IEEE INFOCOM 2023

MERCURY: Fast Transaction Broadcast in High Performance Blockchain Systems

Mingxun Zhou (Carnegie Mellon University, USA); Yilin Han (Shanghai Tree-Graph Blockchain Research Institute, China); Liyi Zeng (Tsinghua University, China); Peilun Li (Shanghai Tree-Graph Blockchain Research Institute, China); Fan Long (University of Toronto, Canada); Dong Zhou (IMO Ventures, China); Ivan Beschastnikh (University of British Columbia, Canada); Ming Wu (Shanghai Tree-Graph Blockchain Research Institute, China)

0

Blockchain systems must be secure and offer high performance. A key mechanism that these systems rely on to provide both of these features is transaction broadcast. Unfortunately, in today's systems, the broadcast protocols are highly inefficient. The wide adoption of public blockchain systems increases the importance of high performance blockchain infrastructure that offers both high throughput and low transaction confirmation latency. Underlying such infrastructures is a fast, efficient, and robust transaction broadcast mechanism.

We present Mercury, a new transaction broadcast protocol designed for high performance blockchains. Mercury shortens the transaction propagation delay using two techniques: a virtual coordinate system (VCS) and an early outburst strategy. Our simulation results show that Mercury outperforms prior propagation schemes and decreases overall propagation latency by up to 44%. When implemented in Conflux, an open-source high-throughput blockchain system, Mercury reduces transaction propagation latency by over 50% with a bandwidth overhead of less than 5%.

Speaker Liyi Zeng (Tsinghua University)

Liyi Zeng is a PhD candidate from IIIS, Tsinghua University, advised by Prof. Wei Xu. Her research areas include Blockchain System, Data Mining and Network Security.

PROPHET: Conflict-Free Sharding Blockchain via Byzantine-Tolerant Deterministic Ordering

Zicong Hong (The Hong Kong Polytechnic University, China); Song Guo and Enyuan Zhou (The Hong Kong Polytechnic University, Hong Kong); Jianting Zhang (Purdue University, USA); Chen Wuhui (Sun Yat-sen University, China); Jinwen Liang and Jie Zhang (The Hong Kong Polytechnic University, Hong Kong); Albert Zomaya (The University of Sydney, Australia)

0

Sharding scales throughput by splitting blockchain nodes into parallel groups. However, the independent and random scheduling for cross-shard transactions among different shards results in numerous conflicts and aborts, since cross-shard transactions from different shards may access the same account. Deterministic ordering can eliminate conflicts by predetermining a global order for transactions before processing, as proved in the database. However, due to the intertwining of the Byzantine environment and shard isolation, there is no trusted party able to predetermine such an order for cross-shard transactions. Therefore, this paper proposes Prophet, a conflict-free sharding blockchain with a new idea of Byzantine deterministic ordering. It first depends on untrusted self-organizing coalitions of nodes from different shards to pre-execute cross-shard transactions for prerequisite information about ordering. Then, it proposes a trusted global order for transactions by performing a stateless ordering in one of the shards and post-verifying pre-executed results through shard cooperation. Based on the order, the shards thus orderly commit the transactions without conflicts. We rigorously prove the determinism and serializability of transactions under the Byzantine and sharded environment. An evaluation shows that Prophet improves the throughput by 3.11X and achieves nearly no aborts on 1 million Ethereum transactions compared with state-of-the-art sharding.

Speaker Zicong Hong (The Hong Kong Polytechnic University)

Zicong Hong is pursuing his Ph.D. degree in the Department of Computing at Hong Kong Polytechnic University, Hong Kong SAR, China. Before that, in 2020, he received B.E. degree in software engineering from Sun Yat-sen University, Guangzhou, China. Now, he is a visiting student of Distributed Computing Group at ETH Zurich, Switzerland. His research interest broadly lies in the areas of blockchain, edge computing and incentive mechanism.

A Decentralized Truth Discovery Approach to the Blockchain Oracle Problem

Yang Xiao (University of Kentucky, USA); Ning Zhang (Washington University in St. Louis, USA); Wenjing Lou and Thomas Hou (Virginia Tech, USA)

3

Blockchain applications enable distrustful parties to execute business logic without relying on a trusted intermediary. When a blockchain application runs on data from the real world, it relies on an oracle mechanism that transports data from external sources to the blockchain. The blockchain oracle problem arises around the need to procure trustworthy data from external sources without introducing a central point of trust. The truthful data challenge, which emerges when legitimate external sources submit fraudulent or deceitful data, remains unsolved. In this paper, we introduce a new decentralized truth discovering oracle architecture called DecenTruth to address the truthful data challenge using a data-centric approach. DecenTruth aims to enable decentralized oracle nodes to discover and reach consensus on truthful values of common data objects from multi-sourced inputs in an off-chain manner. It harmonizes techniques in both the data plane and consensus plane---truth discovery (TD) and asynchronous BFT consensus---and enables nodes to finalize the same estimated truths on data objects with high accuracy, amid the harsh asynchronous network condition and presence of Byzantine sources and nodes. We implemented DecenTruth and the evaluation results demonstrate DecenTruth's significantly higher Byzantine resilience and long-term data feed accuracy compared to existing median-based aggregation methods.

Speaker Yang Xiao (University of Kentucky)

Yang Xiao is an Assistant Professor of Computer Science at the University of Kentucky. His research interests lie in network security, distributed systems, blockchain and decentralized systems, and mobile network security. He received his Ph.D. degree from Virginia Tech in 2022.

CoChain: High Concurrency Blockchain Sharding via Consensus on Consensus

Mingzhe Li (Hong Kong University of Science and Technology, Hong Kong); You Lin (Southern University of Science and Technology, China); Jin Zhang (Southern University of Science and Technology, USA); Wei Wang (Hong Kong University of Science and Technology, Hong Kong)

0

Sharding is an effective technique to improve the scalability of blockchain. It splits nodes into multiple groups so that they can process transactions in parallel. To achieve higher parallelism and concurrency at large scales, it is desirable to maintain a large number of small shards. However, simply configuring small shards easily results in a higher fraction of malicious nodes inside shards, causing shard corruption and compromising system security. Existing sharding techniques hence demand large shards, at the expense of limited concurrency. To address this limitation, we propose CoChain: a blockchain sharding system that can securely configure small shards for enhanced concurrency. CoChain allows some shards to be corrupted. For security, each shard is monitored by multiple other shards. The latter reach a cross-shard Consensus on the Consensus results of their monitored shard. Once a corrupted shard is found, its subsequent consensus will be taken over by another shard, hence recovering the system. Via Consensus on Consensus, CoChain allows the existence of shards with more fraction of malicious nodes (<2/3) while securing the system, thus reducing the shard size safely. We implement CoChain based on Harmony and conduct extensive experiments. Compared with Harmony, CoChain achieves 35x throughput gain with 6,000+ nodes.

Speaker **Mingzhe LI (A*STAR)**

Mingzhe Li is currently a Scientist with the Institute of High Performance Computing (IHPC), A*STAR, Singapore, and a Champion of its blockchain group. He received his Ph.D. degree from the Department of Computer Science and Engineering, Hong Kong University of Science and Technology in 2022. Prior to that, he received his B.E. degree from Southern University of Science and Technology. His interests are mainly in scalable and secure blockchain protocols, Web 3.0, privacy-preserving blockchain, network economics, etc.

Session Chair

Ruidong Li

OBLIVION: Poisoning Federated Learning by Inducing Catastrophic Forgetting

Chen Zhang (The Hang Seng University of Hong Kong, Hong Kong); Boyang Zhou, Zhiqiang He, Zeyuan Liu, Yanjiao Chen and Wenyuan Xu (Zhejiang University, China); Baochun Li (University of Toronto, Canada)

0

Federated learning is exposed to model poisoning attacks as compromised clients may submit malicious model updates to pollute the global model. To defend against such attacks, robust aggregation rules are designed for the centralized server to winnow out outlier updates, and to significantly reduce the effectiveness of existing poisoning attacks. In this paper, we develop an advanced model poisoning attack against defensive aggregation rules. In particular, we exploit the catastrophic forgetting phenomenon during the process of continual learning to destroy the memory of the global model. Our proposed framework, called OBLIVION, features two special components. The first component prioritizes the weights that have the most influence on the model accuracy for poisoning, which induces a more significant degradation on the global model than equally perturbing all weights. The second component smooths malicious model updates based on the number of selected compromised clients in the current round, adjusting the degree of poisoning to suit the dynamics of each training round. We implement a fully-functional prototype of OBLIVION in PLATO, a real-world scalable federated learning framework. Our extensive experiments over three datasets demonstrate that OBLIVION can boost the attack performance of model poisoning attacks against unknown defensive aggregation rules.

Speaker Yanjiao Chen (Zhejiang University)

Yanjiao Chen received her B.E. degree in Electronic Engineering from Tsinghua University in 2010 and Ph.D. degree in Computer Science and Engineering from Hong Kong University of Science and Technology in 2015. She is currently a Bairen researcher in Zhejiang University, China. Her research interests include AI security, network economics, and IoT security.

SplitGP: Achieving Both Generalization and Personalization in Federated Learning

Dong-Jun Han and Do-Yeon Kim (KAIST, Korea (South)); Minseok Choi (Kyung Hee University, Korea (South)); Christopher G. Brinton (Purdue University & Zoomi Inc., USA); Jaekyun Moon (KAIST, Korea (South))

1

A fundamental challenge to providing edge-AI services is the need for a machine learning (ML) model that achieves personalization (i.e., to individual clients) and generalization (i.e., to unseen data) properties concurrently. Existing techniques in federated learning (FL) have encountered a steep tradeoff between these objectives and impose large computational requirements on edge devices during training and inference. In this paper, we propose SplitGP, a new split learning solution that can simultaneously capture generalization and personalization capabilities for efficient inference across resource-constrained clients (e.g., mobile/IoT devices). Our key idea is to split the full ML model into client-side and server-side components, and impose different roles to them: the client-side model is trained to have strong personalization capability optimized to each client's main task, while the server-side model is trained to have strong generalization capability for handling all clients' out-of-distribution tasks. We analytically characterize the convergence behavior of SplitGP, revealing that all client models approach stationary points asymptotically. Further, we analyze the inference time in SplitGP and provide bounds for determining model split ratios. Experimental results show that SplitGP outperforms existing baselines by wide margins in inference time and test accuracy for varying amounts of out-of-distribution samples.

Speaker Dong-Jun Han (Purdue University)

Dong-Jun Han is currently a postdoctoral researcher at Purdue University working with Prof. Christopher G. Brinton and Prof. Mung Chiang. His research interest lies in the intersection of machine learning and communications/networking, and published papers in top-tier ML conferences (NeurIPS, ICML, ICLR), communications/networking conferences (INFOCOM) and journals (JSAC, TWC). He received his B.S., M.S., and Ph.D. degrees all from KAIST, South Korea.

Network Adaptive Federated Learning: Congestion and Lossy Compression

Parikshit Hegde and Gustavo de Veciana (The University of Texas at Austin, USA); Aryan Mokhtari (University of Texas at Austin, USA)

1

In order to achieve the dual goals of privacy and learning across distributed data, Federated Learning (FL) systems rely on frequent exchanges of large files (model updates) between a set of clients and the server. As such FL systems are exposed to, or indeed the cause of, congestion across a wide set of network resources. Lossy compression can be used to reduce the size of exchanged files and associated delays, at the cost of adding noise to model updates. By judiciously adapting clients' compression to varying network congestion, an FL application can reduce wall clock training time. To that end we propose a Network Adaptive Compression (NAC-FL) policy, which dynamically varies the client's lossy compression choices to network congestion variations. We prove, under appropriate assumptions, that NAC-FL is asymptotically optimal in terms of directly minimizing the expected wall clock training time. Further we show via simulation that NAC-FL achieves robust performance improvements with higher gains in settings with positively correlated delays

Speaker Parikshit Hegde (The University of Texas at Austin)

Parikshit Hegde is a 4th year PhD student in the Department of Electrical and Computer Engineering at The University of Texas at Austin. Previously he received his Bachelors and Masters from the Indian Institute of Technology Madras in Electrical Engineering. He is advised by Gustavo de Veciana in the Wireless, Networking and Communications Group (WNCG). His current research interests are in Federated Learning and Networks.

TVFL: Tunable Vertical Federated Learning towards Communication-Efficient Model Serving

Junhao Wang, Lan Zhang, Yihang Cheng and Shaoang Li (University of Science and Technology of China, China); Hong Zhang, Dongbo Huang and Xu Lan (Tencent, China)

0

Vertical federated learning (VFL) enables multiple participants with different data features and the same sample ID space to collaboratively train a model in a privacy-preserving way. However, the high computational and communication overheads hinder the adoption of VFL in many resource-limited or delay-sensitive applications. In this work, we focus on reducing the communication cost and delay incurred by the transmission of intermediate results in VFL model serving. We investigate the inference results, and find that a large portion of test samples can be predicted correctly by the active party alone, thus the corresponding communication for federated inference is dispensable. Based on this insight, we theoretically analyze the "dispensable communication" and propose a novel tunable vertical federated learning framework, named TVFL, to avoid "dispensable communication" in model serving as much as possible. TVFL can smartly switch between independent inference and federated inference based on the features of the input sample. We further reveal that such tunability is highly related to the importance of participants' features. Our evaluations on seven datasets and three typical VFL models show that TVFL can save 57.6% communication cost and reduce 57.1% prediction latency with little performance degradation.

Speaker Junhao Wang (University of Science and Technology of China)

PHD candidate at University of Science and Technology of China

Session Chair

Carla Fabiana Chiasserini

SkyNet: Multi-Drone Cooperation for Real-Time Identification and Localization

Junkun Peng (Tsinghua University, China); Qing Li (Peng Cheng Laboratory, China); Yuanzheng Tan (Sun Yat-Sen University, China); Dan Zhao (Peng Cheng Laboratory, China); Zhenhui Yuan (Northumbria University, United Kingdom (Great Britain)); Jinhua Chen (Sun Yat-Sen University, China); Hanling Wang (Tsinghua University & Peng Cheng Laboratory, China); Yong Jiang (Graduate School at Shenzhen, Tsinghua University, China)

0

Aerial images from drones have been used to detect and track suspects in the crowd for the public safety purpose. However, using a single drone for human identification and localization faces many challenges including low accuracy and long latency, due to poor visibility, varying field of views (FoVs) and limited on-board computing resources. In this paper, we propose SkyNet, a multi-drone cooperative system for accurate and realtime human identification and localization. SkyNet computes the 3D position of a person by cross searching from multiple views. To achieve high accuracy in identification, SkyNet fuses aerial images of multiple drones according to their legibility. Moreover, by predicting the estimated finishing time of tasks, SkyNet schedules and balances workloads among edge devices and the cloud server to minimize processing latency. We implement and deploy SkyNet in real life, and evaluate the performance of identification and localization with 20 human participants. The results show that SkyNet can locate people with an average error within 0.18m on a square of 554m² . The identification accuracy reaches 91.36%, and the localization and identification process is completed within 0.84s.

Speaker Junkun Peng (Tsinghua University)

Junkun Peng received his B.S. degree in Information Management and Information Systems from Shanghai University, China, in 2015.

He is currently pursuing a Ph.D. in Computer Science at Tsinghua University, China.

His research focuses on real-time video transmission and analysis, and robot learning.

A²-UAV: Application-Aware Content and Network Optimization of Edge-Assisted UAV Systems

Andrea Coletta (JP Morgan AI Research, Italy); Flavio Giorgi, Gaia Maselli, Matteo Prata and Domenicomichele Silvestri (Sapienza University of Rome, Italy); Jonathan Ashdown (United States Air Force, USA); Francesco Restuccia (Northeastern University, USA)

1

To perform object/person detection and target tracking, Unmanned Autonomous Vehicles (UAVs) require the continuous execution of edge-assisted vision tasks based on Deep Neural Networks (DNNs). In multi-hop UAV networks, the execution of these tasks fall short of expectations, owing to the typical bandwidth constraints which affect the quality of received data. In this paper, we propose a novel framework to optimize the number of correctly executed tasks. We consider the UAV application at hand (i.e., the specific DNN and the classes of interest) and formulate an optimization problem that takes into account (i) the relationship between DNN accuracy and image resolution for the classes of interest based on the available dataset, (ii) the target positions, (iii) the current energy/position of the UAVs to optimize routing, data pre-processing and target assignment for each UAV. We extensively evaluate our approach through large-scale simulations, as well as real-world experiments with a testbed. We consider state-of-the-art image classification tasks with four different DNN models. Results show that our approach outperform state-of-the-art approaches. To allow full reproducibility, we pledge to share datasets and code with the research community.

Speaker Matteo Prata (Sapienza University of Rome, Italy)

Matteo Prata is a PhD student in the Department of Computer Science at Sapienza University of Rome, Italy. He has authored numerous research papers in the field of computer network performance, with a particular focus on unmanned aerial vehicle networks. Additionally, his research interests encompass AI applications in finance.

WiSwarm: Age-of-Information-based Wireless Networking for Collaborative Teams of UAVs

Vishrant Tripathi (MIT, USA); Igor Kadota (Columbia University, USA); Ezra Tal (MIT, USA); Muhammad Shahir Abdurrahman (Stanford University & Massachusetts Institute of Technology, USA); Alexander Warren, Sertac Karaman and Eytan Modiano (MIT, USA)

0

The Age-of-Information (AoI) metric has been widely studied in the theoretical communication networks and queuing systems literature. However, experimental evaluation of its applicability to complex real-world time-sensitive systems is largely lacking. In this work, we develop, implement, and evaluate an AoI-based application layer middleware that enables the customization of WiFi networks to the needs of time-sensitive applications. By controlling the storage and flow of information in the underlying WiFi network, our middleware can: (i) prevent packet collisions; (ii) discard stale packets that are no longer useful; and (iii) dynamically prioritize the transmission of the most relevant information. To demonstrate the benefits of our middleware, we implement a mobility tracking application using a swarm of UAVs communicating with a central controller via WiFi. Our experimental results show that, when compared to WiFi-UDP/WiFi-TCP, the middleware can improve information freshness by a factor of 109x/48x and tracking accuracy by a factor of 4x/6x, respectively. Most importantly, our results also show that the performance gains of our approach increase as the system scales and/or the traffic load increases.

Speaker Vishrant Tripathi (MIT)

Vishrant Tripathi is a Ph.D. candidate in the EECS department at MIT, working with Prof. Eytan Modiano at the Laboratory for Information and Decision Systems (LIDS). His research is on modeling, analysis and design of communication networks, with emphasis on wireless and real-time networks. His current focus is on scheduling problems in networked control systems, multi-agent robotics and federated learning.

FlyTracker: Motion Tracking and Obstacle Detection for Drones Using Event Cameras

Yue Wu, Jingao Xu and Danyang Li (Tsinghua University, China); Yadong Xie (Beijing Institute of Technology, China); Hao Cao (Tsinghua University, China); Fan Li (Beijing Institute of Technology, China); Zheng Yang (Tsinghua University, China)

0

Location awareness in environments is one of the key parts for drones' applications and have been explored through various visual sensors. However, standard cameras easily suffer from motion blur under high moving speeds and low-quality image under poor illumination, which brings challenges for drones to perform motion tracking. Recently, a kind of bio-inspired sensors called event cameras emerge, offering advantages like high temporal resolution, high dynamic range and low latency, which motivate us to explore their potential to perform motion tracking in limited scenarios. In this paper, we propose FlyTracker, aiming at developing visual sensing ability for drones of both individual and circumambient location-relevant contextual, by using a monocular event camera. In FlyTracker, background-subtraction-based method is proposed to distinguish moving objects from background and fusion-based photometric features are carefully designed to obtain motion information. Through multilevel fusion of events and images, which are heterogeneous visual data, FlyTracker can effectively and reliably track the 6-DoF pose of the drone as well as monitor relative positions of moving obstacles. We evaluate performance of FlyTracker in different environments and the results show that FlyTracker is more accurate than the state-of-the-art baselines.

Speaker Yue Wu

Yue Wu received the doctor degree in Computer Science and Technology from Beijing Institute of Technology, in 2021. Currently she is a post-doctoral in the School of Software, Tsinghua University, Beijing, China. Her research interests include visual localization, mobile computing, and Internet of things.

Session Chair

Enrico Natalizio

A Framework for Wireless Technology Classification using Crowdsensing Platforms

Alessio Scalingi (IMDEA Networks, Spain); Domenico Giustiniano (IMDEA Networks Institute, Spain); Roberto Calvo-Palomino (Universidad Rey Juan Carlos, Spain); Nikolaos Apostolakis (IMDEA Networks, Spain); Gérôme Bovet (Armasuisse, Switzerland)

1

Spectrum crowdsensing systems do not provide labeled data near real-time yet. We propose a framework that relies solely on Power Spectrum Density (PSD) data collected by low-cost RTL-SDR receivers. A major hurdle is to design a system that is computationally efficient for near real-time operation, yet using only the limited knowledge of 2 MHz bandwidth in low-cost spectrum sensors. First, we present a method for unsupervised transmission detection that works with PSD data already collected by the backend of the crowdsensing platform and that provides stable detection of transmission boundaries. Second, we introduce a data-driven deep learning solution to classify the radio frequency communication technology used by the transmitter, using transmission features in a compressed space extracted from single PSD measurements over at most 2 MHz band for inference to support near real-time operation. We build an experimental platform, and evaluate our framework with real-world data collected from 47 different sensors deployed across Europe. We show that our framework yields an average classification accuracy close to 94.25% over the testing dataset, with a maximum latency of 3.4 seconds when running in a major crowdsensing network.

Speaker Alessio Scalingi (IMDEA Networks)

Alessio Scalingi is Ph.D. student of the Pervasive Wireless Systems Group at IMDEA Networks Institute since January 2020.

He completed both his Bachelor's and Master's degrees in Computer Engineering at the University of Naples Federico II in 2015 and 2019, respectively.

During his Master's program, Alessio conducted research for his thesis at the Computer Science Lab of Saint Louis University in the United States. He also gained valuable experience as a visiting PhD at the Wireless Networks and Embedded Systems (WiNES) Laboratory in Boston, USA, for a period of six months. His primary research interests encompass Collaborative Spectrum Sensing, Machine Learning, Spectrum Anomaly Detection, Open-RAN, and Security in 5G and Beyond Networks.

MagFingerprint: A Magnetic Based Device Fingerprinting in Wireless Charging

Jiachun Li, Yan Meng, Le Zhang and Guoxing Chen (Shanghai Jiao Tong University, China); Yuan Tian (University of California Los Angeles, USA); Haojin Zhu (Shanghai Jiao Tong University, China); Sherman Shen (University of Waterloo, Canada)

0

Wireless charging is a promising solution for charging battery-driven devices pervasively deployed in the Internet of Things (IoT). However, the wide deployment of wireless charging stations is vulnerable to the device masquerade attack, which causes financial loss when billing or charging system damages like overheating and explosion. Device fingerprinting is a classical technique to thwart the device masquerade attack. But existing works either are vulnerable to forging or require specialized equipment, which is not suitable for wireless charging.

In this paper, we design a magnetic based fingerprinting system MAGFINGERPRINT, which utilizes the alternating magnetic signals as the fingerprint and is compatible with existing wireless charging systems. MAGFINGERPRINT is convenient for the user since it only employs commercial-off-the-shelf (COTS) magnetic sensors and requires no action from users. In particular, for the charging device, based on its intrinsic manufacturing errors, MAGFINGERPRINT generates a unique fingerprint according to the distinct magnetic changes during the wireless charging process. It is shown that MAGFINGERPRINT can achieve an accuracy of 98.90% on wireless charging exposed coils, while it is also effective on different commercial wireless charging pads of Apple, Huawei, and Xiaomi.

Speaker Jiachun Li (Shanghai Jiao Tong University)

Jiachun Li is a Ph.D. candidate in the Department of Computer Science and Engineering, Shanghai Jiao Tong University, China. He received the B.S. degree in Communication Engineering from Huazhong University of Science and Technology in 2020. His research interests include smart home security and smart healthcare security.

Plug and Power: Fingerprinting USB Powered Peripherals via Power Side-channel

Riccardo Spolaor and Hao Liu (Shandong University, China); Federico Turrin (University of Padua, Italy); Xiuzhen Cheng (Shandong University, China); Mauro Conti (University of Padua, Italy; TU Delft, Netherlands)

0

The literature and the news regularly report cases of exploiting Universal Serial Bus (USB) devices as attack tools for malware injections and private data exfiltration. To protect against such attacks, security researchers proposed different solutions to verify the identity of a USB device via side-channel information (e.g., timing or electromagnetic emission). However, such solutions often make strong assumptions on the measurement (e.g., electromagnetic interference-free area around the device), on a device's state (e.g., only at the boot or during specific actions), or are limited to one particular type of USB device (e.g., flash drive or input devices).

In this paper, we present PowerID, a novel method to fingerprint USB peripherals based on their power consumption. PowerID analyzes the power traces from a peripheral to infer its identity and properties. We evaluate the effectiveness of our method on an extensive power trace dataset collected from 82 USB peripherals, including 35 models and eight types. Our experimental results show that PowerID accurately recognizes a peripheral type, model, activity, and identity.

Speaker Federico Turrin (University of Padova)

Federico Turrin received his Ph.D. in Brain, Mind, and Computer Science, in 2023 at the University of Padova. He is currently a Post Doc Researcher at the University of Padova and a Cybersecurity Engineer at SPRITZ Matter Srl. He has been visiting researcher at SUTD, in Singapore in 2022. His research interests lie primarily in Cyber-Physical System security with a particular focus on Industrial Control systems security, Vehicles Security, and Anomaly detection.

Contrastive learning with self-reconstruction for channel-resilient modulation classification

Erma Perenda (KU Leuven, Belgium); Sreeraj Rajendran (Sirris, Belgium); Mariya Zheleva (UAlbany SUNY, USA); Gérôme Bovet (Armasuisse, Switzerland); Sofie Pollin (KU Leuven, Belgium)

0

Despite the substantial success of deep learning for modulation classification, models trained on a specific transmitter configuration and channel model often fail to generalize well to other scenarios with different transmitter configurations, wireless fading channels, or receiver impairments such as clock offset. This paper proposes Contrastive Learning with Self-Reconstruction called CLSR-AMC to learn good representations of signals resilient to channel changes. While contrastive loss focuses on the differences between individual modulations, the reconstruction loss captures representative features of the signal. Additionally, we develop three data augmentation operators to emulate the impact of channel fading and receiver imperfection impairments without exhaustive modeling of different channel profiles. We perform extensive experimentation with commonly used datasets. We show that CLSR-AMC outperforms its counterpart based on contrastive learning for the same amount of labelled data by significant average accuracy gains of 24.29%, 17.01%, and 15.97% in Additive White Gaussian Noise (AWGN), Rayleigh+AWGN, and Rician+AWGN channels, respectively.

Speaker Mariya Zheleva (University at Albany – SUNY, New York, USA)

Mariya Zheleva is an Associate Professor in Computer Science at University at Albany – SUNY. She graduated with her PhD in Computer Science from University of California Santa Barbara in 2014. She leads the UbiNET Lab, which conducts research at the intersection of wireless communications and Information and Communication Technology for Development. Mariya is the recipient of the NSF CAREER award, the Dynamic Spectrum Alliance 2019 Award for University Research on New Opportunities for Dynamic Spectrum Access, and the University at Albany 2019 President’s Award for Exemplary Public Engagement. She is the co-lead for the NSF-supported National Radio Dynamic Zones Partnership and Workshop Series; and a founding member of SpectrumX.

Session Chair

Francesco Restuccia

Robustified Learning for Online Optimization with Memory Costs

Pengfei Li (UC Riverside, USA); Jianyi Yang and Shaolei Ren (University of California, Riverside, USA)

1

Online optimization with memory costs has many real-world applications, where sequential actions are made without knowing the future input. Nonetheless, the memory cost couples the actions over time, adding substantial challenges. Conventionally, this problem has been approached by various expert-designed online algorithms with the goal of achieving bounded worst-case competitive ratios, but the resulting average performance is often unsatisfactory. On the other hand, emerging machine learning (ML) based optimizers can improve the average performance, but suffer from the lack of worst-case performance robustness. In this paper, we propose a novel expert-robustified learning (ERL) approach, achieving both good average performance and robustness. More concretely, for robustness, ERL introduces a novel projection operator that robustifies ML actions by utilizing an expert online algorithm; for average performance, ERL trains the ML optimizer based on a recurrent architecture by explicitly considering downstream expert robustification process. We prove that, for any $\lambda\geq1$, ERL can achieve $\lambda$-competitive against the expert algorithm and $\lambda\cdot C$-competitive against the optimal offline algorithm (where$C$ is the expert's competitive ratio). Additionally, we extend our analysis to a novel setting of multi-step memory costs. Finally, our analysis is supported by empirical experiments for an energy scheduling application.

Speaker Pengfei Li (University of California, Riverside)

Pengfei Li is a third-year CS Ph.D. student in University of California, Riverside, under the supervision of Prof. Shaolei Ren. He obtained a M.S.E degree in Laboratory for Computational Sensing and Robotics (LCSR), Johns Hopkins University, under the supervision of Prof. Alan Yuille and Prof. Gregory Hager. Pengfei’s research focuses on online optimization, scheduling, Machine Learning and applications in sustainable AI. His recent work on AI water footprint has been widely reported by the major media (e.g. Barron’s, Forbes, Fox News, CNN News 18, etc)

Online Distributed Optimization with Efficient Communication via Temporal Similarity

Juncheng Wang and Ben Liang (University of Toronto, Canada); Min Dong (Ontario Tech University, Canada); Gary Boudreau and Ali Afana (Ericsson, Canada)

0

We consider online distributed optimization in a networked system, where multiple devices assisted by a server collaboratively minimize the accumulation of a sequence of global loss functions that can vary over time. To reduce the amount of communication, the devices send quantized and compressed local decisions to the server, resulting in noisy global decisions. Therefore, there exists a tradeoff between the optimization performance and the communication overhead. Existing works separately optimize computation and communication. In contrast, we jointly consider computation and communication over time, by encouraging temporal similarity in the decision sequence to control the communication overhead. We propose an efficient algorithm, termed Online Distributed Optimization with Temporal Similarity (ODOTS), where the local decisions are both computation- and communication-aware. Furthermore, ODOTS uses a novel tunable virtual queue, which completely removes the commonly assumed Slater's condition through a modified Lyapunov drift analysis. ODOTS delivers provable performance bounds on both the optimization objective and constraint violation. As an example application, we apply ODOTS to enable communication-efficient federated learning. Our experimental results based on real-world image classification demonstrate that ODOTS obtains higher classification accuracy and lower communication overhead compared with the current best alternatives for both convex and non-convex loss functions.

Speaker Ben Liang

Ben Liang received honors-simultaneous B.Sc. (valedictorian) and M.Sc. degrees in Electrical Engineering from Polytechnic University (now the engineering school of New York University) in 1997 and the Ph.D. degree in Electrical Engineering with a minor in Computer Science from Cornell University in 2001. He was a visiting lecturer and post-doctoral research associate at Cornell University in the 2001 - 2002 academic year. He joined the Department of Electrical and Computer Engineering at the University of Toronto in 2002, where he is now Professor and L. Lau Chair in Electrical and Computer Engineering. His current research interests are in networked systems and mobile communications. He is an associate editor for the IEEE Transactions on Mobile Computing and has served on the editorial boards of the IEEE Transactions on Communications, the IEEE Transactions on Wireless Communications, and the Wiley Security and Communication Networks. He regularly serves on the organizational and technical committees of a number of conferences. He is a Fellow of IEEE and a member of ACM and Tau Beta Pi.

A Bayesian Framework for Online Nonconvex Optimization over Distributed Processing Networks

Zai Shi (The Ohio State University, USA); Yilin Zheng (the Ohio State University, USA); Atilla Eryilmaz (The Ohio State University, USA)

0

In many applications such as machine learning, reinforcement learning, and optimization for data centers, the increasing data size and model complexity have made it impractical to run optimizations over a single machine. Therefore, solving the distributed optimization problem has become an important task. In this work, we consider a distributed processing network $G=(\mathcal{V},\mathcal{E})$ with $n$ nodes, where each node $i$ can only evaluate the values of a local function and can only communicate with its neighbors. The objective is to reach consensus on the global optimizer of $\max_{x\in\mathcal{X}} \frac{1}{n}\sum_{i=1}^n f_i(x)$. Previous methods either assume gradient information which is not suitable for model-free learning, or consider the zeroth-order information but assume convexity of the objective functions and can only guarantee convergence to a stationary point for non-convex objectives. To address these limitations, we drop both the known gradient assumption and convexity assumption. Instead, we propose a distributed Bayesian framework for the problem with only zeroth-order information and nonconvex objective functions in a Matern Reproducing Kernel Hilbert Space. Under this framework, we propose an algorithm and show that with high probability it reaches consensus and has a sublinear regret with regard to the global optimal. The results are validated under numerical studies.

Speaker Yilin Zheng (The Ohio State University)

Yilin Zheng is a graduate student at the Ohio State University in the ECE department under the supervision of Professor Atilla Eryilmaz. His research is mainly focusing on online learning and optimization.

Constrained Bandit Learning with Switching Costs for Wireless Networks

Juaren Steiger (Queen's University, Canada); Bin Li (The Pennsylvania State University, USA); Bo Ji (Virginia Tech, USA); Ning Lu (Queen's University, Canada)

0

Bandits with arm selection constraints and bandits with switching costs have both gained recent attention in wireless networking research. Pessimistic-optimistic algorithms, which combine bandit learning with virtual queues to track the constraints, are commonly employed in the former. Block-based algorithms, where switching is disallowed within a block, are commonly employed in the latter. While efficient algorithms have been developed for both problems, it remains challenging to guarantee low regret and constraint violation in a bandit problem that includes both arm selection constraints and switching costs due to the tight coupling between the two. Here, switching may be necessary to decrease the constraint violation but comes at the cost of increased switching regret. In this paper, we tackle the constrained bandits with switching costs problem, for which we design a block-based pessimistic-optimistic algorithm. We identify three timely wireless networking applications for this framework in edge computing, mobile crowdsensing, and wireless network selection. We also prove that our algorithm achieves sublinear regret and vanishing constraint violation and corroborate these results with synthetic simulations and extensive trace-based simulations in the wireless network selection setting.

Speaker Juaren Steiger (Queen's University)

Juaren Steiger is a PhD student at Queen's University in Canada who is studying online learning and its applications to wireless communications.

Session Chair

Ruozhou Yu

BreathSign: Transparent and Continuous In-ear Authentication Using Bone-conducted Breathing Biometrics

Feiyu Han, Panlong Yang and Shaojie Yan (University of Science and Technology of China, China); Haohua Du (Beihang University, China); Yuanhao Feng (University of Science and Technology of China, China)

0

As one of the most natural physiological activities, breathing provides an effective and ubiquitous approach for continuous authentication. Inspired by that, this paper presents BreathSign, which reveals a novel biometric characteristic using bone-conducted human breathing sound and provides an anti-spoofing and transparent authentication mechanism based on inward-facing microphones on commercial earphones. To explore the breathing differences among persons, we first analyze how the breathing sound propagating in the body, and then derive unique body physics-level features from breathing-induced body sounds. Furthermore, to eliminate the impact of behavioral biometrics, we design a triple network model to reconstruct breathing behavior-independent features. Extensive experiments with 20 subjects over a month period have been conducted to evaluate the accuracy, robustness, and vulnerability of BreathSign. The results show that our system accurately authenticates users with an average authentication accuracy rate of 95.17% via only one breathing cycle, and effectively defends against various spoofing attacks with an average spoofing attack detection rate of 98.25%. Compared with other continuous authentication solutions, BreathSign extracts hard-to-forge biometrics in the effortless human breathing activity for authentication and can be easily implemented on commercial earphones with high usability and enhanced security.

Speaker Feiyu Han (University of Science and Technology of China)

Feiyu Han is a PhD student in the LINKE lab at USTC, China. His research area is wearable computing and authentication. He received his B.S. degree from the school of computer science and engineering, Nanjing University of Science and Technology in 2019.

FacER: Contrastive Attention based Expression Recognition via Smartphone Earpiece Speaker

Guangjing Wang, Qiben Yan, Shane Patrarungrong, Juexing Wang and Huacheng Zeng (Michigan State University, USA)

1

Facial expression recognition has been applied to reveal users' emotional status when they interact with digital content. Previous studies consider using cameras or wearable sensors for expression recognition. However, these approaches bring considerable privacy concerns or extra device burdens. Moreover, the recognition performance of camera-based methods deteriorates when users are wearing masks. In this paper, we propose FacER, an active acoustic facial expression recognition system. As a software solution on a smartphone, FacER avoids extra costs of external microphone arrays. FacER extracts facial expression features by modeling the echoes of emitted near-ultrasound signals between the earpiece speaker and the 3D facial contour. Besides isolating a range of background noises, FacER is designed to identify different expressions from various users with a limited set of training data. To achieve this, we propose a contrastive external attention-based model to learn consistent expression feature representations across different users. Extensive experiments with 20 volunteers with or without masks show that FacER can recognize 6 common facial expressions with more than 85% accuracy, outperforming the state-of-the-art acoustic sensing approach by 10% in various real-life scenarios. FacER provides a more robust solution for recognizing the users' expressions in a convenient and usable manner.

Speaker Guangjing Wang (Michigan State University)

Guangjing Wang is a Ph.D. candidate in Computer Science and Engineering, at Michigan State University. His research mainly focuses on mobile sensing, security and privacy. More information can be found at https://guangjing.wang/

Wider is Better? Contact-free Vibration Sensing via Different COTS-RF Technologies

Zhe Chen (China-Singapore International Joint Research Institute, China); Tianyue Zheng (Nanyang Technological University, Singapore); Chao Cai (Huazhong University of Science and Technology, China); Yue Gao (University of Surrey, United Kingdom (Great Britain)); Pengfei Hu (Shandong University, China); Jun Luo (Nanyang Technological University, Singapore)

0

Vibration sensing is crucial to human life and work, as vibrations indicate the status of their respective sources (e.g., heartbeat to human health condition). Given the inconvenience of contact sensing, both academia and industry have been intensively exploring contact-free vibration sensing, with several major developments leveraging radio-frequency (RF) technologies made very recently. However, a measurement study systematically comparing these options is still missing. In this paper, we choose to evaluate five representative commercial off-the-shelf (COTS) RF technologies with different carrier frequencies, bandwidths, and waveform designs. We first unify the sensing data format and processing pipeline, and also propose a novel metric v-SNR to quantify sensing quality. Then our extensive evaluations start from controlled experiments for benchmarking, followed by investigations on two real-world applications: machinery vibration measurement and vital sign monitoring. Our comprehensive study reveals that Wi-Fi performs the worst among all five technologies, while a lesser-known UWB-based technology achieves the best overall performance, and others have respective pros and cons in different scenarios.

Speaker Zhe Chen (Fudan University)

Dr. Zhe Chen is the Co-Founder of AIWiSe Ltd. Inc. He obtained his Ph.D. degree in Computer Science from Fudan University, China, with a 2019 ACM SIGCOMM China Doctoral Dissertation Award. Before joining AIWiSe, he worked as a research fellow at NTU for several years, and his research achievements, along with his efforts in launching products based on them, have thus earned him the 2021 ACM SIGMOBILE China Rising Star Award recently. His current research interests include wireless networking, deep learning, mobile and pervasive computing, and embedded systems.

WakeUp: Fine-Grained Fatigue Detection Based on Multi-Information Fusion on Smart Speakers

Zhiyuan Zhao, Fan Li and Yadong Xie (Beijing Institute of Technology, China); Yu Wang (Temple University, USA)

0

With the development of society and the gradual increase of life pressure, the number of people engaged in mental work and working hours have increased significantly, resulting in more and more people in a state of fatigue. It not only reduces people's work efficiency, but also causes health and safety related problems. The existing fatigue detection systems either have different shortcomings in diverse scenarios or are limited by proprietary equipment, which is difficult to be applied in real life. Motivated by this, we propose a multi-information fatigue detection system named WakeUp based on commercial smart speakers, which is the first to fuse physiological and behavioral information for fine-grained fatigue detection in a non-contact manner. We carefully design a method to simultaneously extract users' physiological and behavioral information based on the MobileViT network and VMD decomposition algorithm respectively. Then, we design a multi-information fusion method based on the statistical features of these two kinds of information. In addition, we adopt an SVM classifier to achieve fine-grained fatigue level. Extensive experiments with 20 volunteers show that WakeUp can detect fatigue with an accuracy of 97.28%. Meanwhile, WakeUp can maintain stability and robustness under different experimental settings.

Speaker Zhiyuan Zhao (Beijing Institute of Technology )

Zhiyuan Zhao received the ME degree in computational mathematics from the Hefei University of Technology in 2020, and BE degree in applied mathematics from the Henan Normal University in 2017. Currently he is a Ph.D. student in the School of Computer Science, Beijing Institute of Technology, Beijing, China. His research interests include mobile computing, mobile health, human-computer interaction, and deep learning.

Session Chair

Zhichao Cao

Harry: A Scalable SIMD-based Multi-literal Pattern Matching Engine for Deep Packet Inspection

Hao Xu (Fudan University, China); Harry Chang, Wenjun Zhu, Yang Hong, Geoff Langdale and Kun Qiu (Intel, China); Jin Zhao (Fudan University, China)

0

Deep Packet Inspection (DPI) is a significant network security technique. It examines traffic workloads by searching for specific rules. Since every byte of packets needs to be examined by many literal rules, multi-literal matching becomes the performance bottleneck of DPI. FDR, the fastest multi-literal matching engine on CPUs, takes advantage of Single-Instruction-Multiple-Data (SIMD) to alleviate this bottleneck and achieves a performance boost over the widely-used Aho-Corasick (AC) algorithm. However, FDR does not deeply exploit the data-level parallelism of SIMD and its SIMD vector utilization is only 50%. Besides, limited by certain SIMD shift instructions, it cannot benefit from advanced SIMD instruction sets. To overcome these issues, we propose Harry, a scalable and SIMD-based multi-literal matching engine. Harry adopts a column-vector-based matching algorithm to improve the data-level parallelism and SIMD vector utilization. To support the algorithm, it takes two encoding methods to compress the mask table. Also, it utilizes shuffle instruction to implement shift. We implement Harry on commodity CPU and evaluate it with real network traffic and DPI rules. Our evaluation shows that Harry reaches a throughput of 30∼70 Gbit/s, up to 52x that of AC and 2.09x of FDR. It has been successfully deployed in Hyperscan.

Speaker Hao Xu (Fudan University)

Hao Xu is currently a second year Master student in School of Computer Science Technology, Fudan University, advised by Prof. Jin Zhao. He received his B.Eng. from Northeastern University in 2019. His research interests lie in computer networking and systems.

COIN: Cost-Efficient Traffic Engineering with Various Pricing Schemes in Clouds

Gongming Zhao, Jingzhou Wang and Hongli Xu (University of Science and Technology of China, China); Zhuolong Yu (Johns Hopkins University, USA); Chunming Qiao (University at Buffalo, USA)

0

The rapid growth of cloud services has brought a significant increase in inter-datacenter traffic. To transfer data among geographically distributed datacenters, cloud providers need to purchase bandwidth from ISPs. The data transferring cost has become one of the major expenses for cloud providers. Therefore, it is essential for a cloud provider to carefully allocate inter-datacenter traffic among the ISPs' links to minimize the costs. Exiting solutions mainly focus on the situations where all links adopt the same pricing scheme. However, in practice, ISPs usually provide multiple pricing schemes for their links due to market competition, which makes the existing solutions non-optimal. Thus, a new traffic engineering approach that considers various pricing schemes is needed. This paper presents COIN, a new framework for cost-efficient traffic engineering with various pricing schemes. We propose a partition rounding traffic engineering algorithm based on linear independence analysis. The approximation factors and time complexity are formally analyzed. We further conduct large-scale simulations with real-world topologies and datasets. Extensive simulation results show that COIN can save the data transferring cost by up to 54.54% compared with the state-of-the-art solutions.

Speaker Jingzhou Wang (University of Science anf Technology of China)

Jingzhou Wang is pursuing his Ph.D. degree in Computer Science in USTC. His interest includes cloud networks, quality of service and quantum networks. He has published 5 papers in top-ranked conferences and journals, including INFOCOM, ICDCS and ToN.

DeeP4R: Deep Packet Inspection in P4 using Packet Recirculation

Sahil Gupta (Rochester Institute of Technology, USA); Devashish Gosain (Max Planck Institute for Informatics, Germany); Minseok Kwon and Hrishikesh B Acharya (Rochester Institute of Technology, USA)

1

Software-defined networks are useful for multiple tasks, including firewalling, telemetry, and flow analysis. In particular, the P4 language makes it possible to carry out some simple packet processing tasks in the data plane, i.e., on the switch itself (without real-time support from the SDN controller or a server). However, owing to the limitations of packet parsing in P4, these tasks involve only the packet headers. In this paper, we present a novel approach that allows Deep Packet Inspection (DPI) - i.e., inspection of the packet payload - in the data plane, using P4 alone. We make use of the fact that in P4, a switch can clone and recirculate packets. One copy (clone) can be recirculated, slicing off a byte in each round, and using a finite-state machine to check if a target string has yet been seen. If the target string is found, the other copy (original packet) is discarded; if not, it is passed through. Our approach allows us to build the first application-layer firewall (URL filter) in the data plane, and to achieve essentially line-rate performance while filtering thousands of URLs, on a commodity programmable switch. It may in future also be used for other DPI tasks.

Speaker Sahil Gupta (Rochester Institute of Technology)

The author is a Ph.D. student at the Rochester Institute of Technology in the computer science department. The author is interested in Networks, Systems, and Network Security as research areas.

Burst can be Harmless: Achieving Line-rate Software Traffic Shaping by Inter-flow Batching

Danfeng Shan, Shihao Hu and Yuqi Liu (Xi'an Jiaotong University, China); Wanchun Jiang (Central South University, China); Hao Li, Peng Zhang, Yazhe Tang and Huanzhao Wang (Xi'an Jiaotong University, China); Fengyuan Ren (Tsinghua University, China)

2

Traffic shaping is a common function at end hosts. Compared with hardware ones, software shapers are more flexible to be developed and deployed, and thus are very attractive. Nevertheless, software approaches are still unsatisfactory as they struggle to saturate 40Gbps and higher speed. While much effort has been made to reduce the intrinsic overhead of software traffic shaping, we find that it is the extrinsic overhead, such as PCIe communications and interrupts, that hinders shaping from achieving 40Gbps - 100Gbps speed. Batching is an effective way to amortize these overheads. However, blindly batching can degrade the network performance, as it introduces bursts into the network. Diving into the dilemma, we find that intra-flow burst is to blame for harming the network performance, while inter-flow burst, consisting of packets from different flows, can be naturally demultiplexed in the network. Based on the insight, we present FlowBundler, which can achieve efficient traffic shaping by inter-flow batching. Testbed experiments show that FlowBundler can achieve an accurate shaping of 98Gbps with a single CPU core, which is 2.6× better than state-of-the-art approaches. Large-scale simulations show that FlowBundler can batch packet transmissions without harming the network performance.

Speaker Danfeng Shan (Xi'an Jiaotong University)

Danfeng Shan is an associate professor at School of Computer Science and Technology, Xi'an Jiaotong University. He received his Ph.D. degree from Department of Computer Science and Technology, Tsinghua University in 2018, and the B.E. degree from Department of Computer Science and Technology from Xi'an Jiaotong University in 2013. His research interests include traffic management, data center networking, congestion control.

Session Chair

Marco Fiore

Demo Abstract: Predictive Radio Environment for Digital Twin Communication Platform via Enhanced Sensing

Yinghe Miao (Beijing University of Posts and Telecommunications, China); Yuxiang Zhang (Beijing University of Posts & Telecommunications, China); Jianhua Zhang, Yutong Sun, Yixuan Tian and Li Yu (Beijing University of Posts and Telecommunications, China); Guangyi Liu (Research Institute of China Mobile, China)

0

With the accelerated development of digitalization and intelligentization of communications, sixth generation (6G) network is envisioned to support digital twin (DW) for intelligent network autonomy. However, higher carrier frequencies, more complex technologies, and diverse scenarios lead to dynamic radio environment, which makes 6G network intractable to design and optimize. In this demonstration, we propose a platform architecture and implementation via enhanced sensing to achieve radio environment prediction for 6G digital twin communication. The platform consists of three modules. First, the environmental information is captured by environment sensing module, and then the radio propagation path can be calculated by radio propagation prediction module. At last, the channel impulse response (CIR) is calculated and visualized by data processing and visualization module. In a nutshell, this platform is promising to realize the prediction of radio environment to improve the adaptability of the 6G network in environmental changes from bottom to top.

Speaker Yinghe Miao; Yuxiang Zhang; Yutong Sun

Speaker biography is not available.

Demo Abstract: Demonstrating Resource-Efficient SLAM in Virtual Spacecraft Environments

Ying Chen (Duke University, USA); John Sarik (Columbia University, USA); Hazer Inaltekin (Macquarie University, Australia); Maria Gorlatova (Duke University, USA)

0

The performance of simultaneous localization and mapping (SLAM) systems is impacted by the constrained computation capabilities of mobile devices. Given that the advancement of these systems relies on accurate evaluation of SLAM performance, this issue is exacerbated by the difficulty in evaluating SLAM performance in practice, due to the unavailability of ground truth data. In this demo, we present SpacecraftWalk, a resource-efficient SLAM framework that constructs maps (of the environments) with minimal uncertainty under resource budgets. SpacecraftWalk is evaluated within virtual spacecraft environments (meeting NASA lighting standards) in game engine-based emulators that generate ground truth automatically. Demo participants will navigate in virtual environments while creating their own moving trajectories for evaluating SLAM. They will develop an intuition for how uncertainty-based map construction improves resource efficiency. This demonstration accompanies [1].

Speaker Ying Chen

Speaker biography is not available.

A Scalable Byzantine Consensus Parallelism and its Practical Implementation

Xiao Chen (University of Edinburgh, United Kingdom (Great Britain))

0

Lack of scalability is one of the major obstacles to the broader adoption of Byzantine fault tolerance (BFT)-based blockchain consensus for large-scale networks. Recent blockchains use sharding to improve scalability at the cost of safety and responsiveness. To address this issue, we have proposed SharBFT which is a novel sharding-based BFT consensus parallelism with improved safety and responsiveness while retaining scalability. This demo presents a practical implementation of SharBFT, i.e., SharBFT-SMaRt which is a scalable Byzantine fault-tolerant state machine replication library developed in Java, which prioritises simplicity and robustness. Our main objective is to offer a code base that can be expanded to develop new protocols and utilised to construct blockchain systems.

Speaker

Speaker biography is not available.

AR-Span: A Multi-Screen Adaptive Displaying System Using Smart Handheld Devices Based on Mobile Networks

Lien-Wu Chen, Ai-Ni Li and Yu-An Shi (Feng Chia University, Taiwan)

0

In this paper, we design and implement a multi-screen adaptive displaying system, called AR-Span, which can enable the assembling of heterogeneous mobile devices into one multi-screen displaying equipment. The AR-Span system consists of multiple mobile devices with various screen sizes and a dedicated server communicating and coordinating among these mobile devices. AR-Span can cooperatively display a complete image on multiple mobile devices. In addition, AR-Span can precisely partition a video and synchronically play the video on multiple mobile devices. Furthermore, AR-Span can arbitrarily customize an electronic marquee on multiple mobile devices. In particular, AR-Span can immediately adjust and control the displayed content in an augmented reality based manipulation manner. Experimental results show that AR-Span manipulates the desired content much faster than existing methods/systems and can significantly reduce the total operating time. Moreover, AR-Span can improve the displaying time inconsistency and unsynchronized content gap on heterogeneous mobile devices.

Speaker Yu-An Shi

Speaker biography is not available.

Demonstrating Flow-Level In-Switch Inference

Michele Gucciardo (IMDEA Networks Institute, Spain); Aristide Tanyi-Jong Akem (IMDEA Networks Institute, Spain & Universidad Carlos III de Madrid, Spain); Beyza Butun (Universidad Carlos III de Madrid & IMDEA Networks, Spain); Marco Fiore (IMDEA Networks Institute, Spain)

0

Existing approaches for in-switch inference with Random Forest (RF) models that can run on production-level hardware do not support flow-level features and have limited scalability to the task size. This leads to performance barriers when tackling complex inference problems with sizable decision spaces. Flowrest is a complete RF model framework that fills existing gaps in the existing literature and enables practical flow-level inference in commercial programmable switches. In this demonstration, we exhibit how Flowrest can classify individual traffic flows at line rate in an experimental platform based on Intel Tofino switches. To this end, we run experiments with real-world measurement data, and show how Flowrest yields improvements in accuracy with respect to solutions that are limited to packet-level inference in programmable hardware.

Speaker Michele Gucciardo (IMDEA Networks Institute)

Powering Inaccessible IoT Devices Through a WPT-enabled Sustainable UAV Network

Prodromos-Vasileios Mekikis, Pavlos Bouzinis, Nikos Mitsiou, Sotiris A. Tegos and Vasilis K. Papanikolaou (Aristotle University of Thessaloniki, Greece); Dimitrios Tyrovolas (Aristotle University of Thessaloniki & Technical University of Chania, Greece); Panagiotis D. Diamantoulakis and George K. Karagiannidis (Aristotle University of Thessaloniki, Greece)

0

Powering Internet of Things (IoT) devices in hard to reach or hazardous locations could be prohibitive in terms of cost and safety. In this work, we demonstrate a thorough solution that tackles this problem using a network of unmanned aerial vehicles (UAVs) with wireless power transfer (WPT) capabilities. Given the large-scale IoT deployments of the future and the energy needs of the UAVs, we consider an infrastructure of charging stations for the UAVs that allow an uninterrupted and flexible UAV deployment based on the current energy needs of the IoT devices. Then, we exhibit an orchestrator that communicates with the entire infrastructure and handles the UAV traffic and energy decisions, as well as the digital representation and interaction of the network with the user.

Speaker Prodromos-Vasileios Mekikis

Speaker biography is not available.

Demo: Collaborative Mixed-Reality-Based Firefighter Training

Zhanchen Dong, Jiangong Chen and Bin Li (The Pennsylvania State University, USA)

0

Wireless collaborative mixed reality (WCMR) has many fascinating applications in education, training, manufacturing, and gaming. In this demo, we develop a WCMR-based firefighter training system that provides firefighters with experiences in extreme and diverse fire accidents without any safety concerns. In such a system, it is important to ensure that all firefighters can see almost the same status of the fire accident to facilitate collaborative training. This is challenging due to the heterogeneous communication delays of different network users. We propose the latency compensation algorithm that determines when the edge server should transmit the message to users based on the estimated latency of each user. Our experiment demonstrates around 55% synchronization performance improvement while guaranteeing at least 60 frames per second (FPS).

Speaker

Speaker biography is not available.

Demo: Missed An Exit? Confusion Zone Detection

Seyhan Ucar (Toyota Motor North America R&D, InfoTech Labs, USA); Takamasa Higuchi (Toyota Motor North America R&D, USA); Onur Altintas (Toyota Motor North America R&D, InfoTech Labs, USA)

0

Drivers may get confused for any number of reasons. For example, due to confusion, they miss their intended exits in some regions. In such cases, confused drivers should loop back around. However, some drivers exhibit risky driving (e.g., back up on the highway or perform abrupt lane changes). Inferring such zones where most drivers get confused could be helpful. Drivers can use such prior knowledge and plan their exits or turns accordingly. In this paper, we focus on this use case. Connected vehicles share their location data with the remote server, and the remote server examines data to identify turn loops in real time. The region becomes a confusion zone when a turn loop is executed by several vehicles. We demonstrate the feasibility of confusion zone detection through a simulation-based demonstration.

Speaker Seyhan Ucar

Dr. Uçar is currently working as a Principal Researcher in Intelligent Mobility Systems at InfoTech Labs, Toyota Motor North America USA. He received his B.Sc. degree in Computer Engineering from İzmir Institute of Technology in 2011. He received his M.Sc. and a Ph.D. degree in Computer Science and Engineering from Koç University in 2013 and 2017, respectively. Throughout his M.Sc. studies, he worked on developing multi-hop clustering algorithms and Long-Term Evaluation (LTE) based heterogeneous architectures for vehicular ad hoc networks. During his Ph.D., he focuses on Visible Light Communication (VLC) and automated car following (or platooning) where a group of vehicles travels within close proximity through communication. He is now working on intelligent transportation systems and applications and analyzing the impact of connected vehicles on transportation safety and management.

A Runtime Anomaly Detector via Service Communication Proxy for 5G Mobile Networks

Yin-Chi Li, Ping-Tsan Liu, Yi-An Tai, Che-Hung Liu, Man-Hsin Chen and Chi-Yu Li (National Yang Ming Chiao Tung University, Taiwan); Guan-Hua Tu (Michigan State Unversity, USA)

0

With the growing popularity of the 5G mobile network, its security is becoming important. Although the newly introduced 5G security mechanisms have addressed many legacy security issues, there may be still vulnerabilities in the 5G network due to newly deployed components and used technologies. To detect security threats, we develop a runtime anomaly detector (RAD) platform, designated as 5G-RAD, to cooperate with the operational 5G core network via the service communication proxy (SCP). It validates the core network operation in terms of state machine and message content by analyzing control-plane messages. We demonstrate its effectiveness by building a 5G mobile network architecture with SCP based on the open-source free5GC and UERANSIM. The 5G-RAD is tested with three attacks, including DoS, authentication bypass, and invalid message injection; it can successfully detect them at run time.

Speaker Yin-Chi Li

Speaker biography is not available.

AI/ML Data-driven Control Loop for Managing O-RAN SDR-based RANs

Jaswanth S. R. Mallu (Virginia Tech Commonwealth Cyber Initiative, USA); Joao F. Santos (Virginia Tech & Commonwealth Cyber Initiative, USA); Aloizio Pereira Da Silva (Virginia Tech, USA & Commonwealth Cyber Initiative, USA); Prateek Sethi (Virginia Tech Commonwealth Cyber Initiative, USA); Vikas Krishnan Radhakrishnan (Virginia Tech, USA); Luiz DaSilva (Virginia Tech, USA & Trinity College Dublin, Ireland)

1

Open Radio Access Network (O-RAN) introduced a common control and management overlay, allowing mobile network operators to embed networking intelligence using different types of third-party applications: xApps for real-time control loops, and rApps for Artificial Intelligence (AI)/Machine Learning (ML)-based classification and decision-making. However, the development of reference implementations for rApps lags behind the progress in other O-RAN-related standardization efforts. In this demonstration, we showcase a proof-of- concept rApp capable of generating policies to steer the behavior of xApps, and detail how we extended a RAN slicing xApp to react to such policies, creating the first experimental ML-based RAN slicing platform based on O-RAN.

Speaker

Speaker biography is not available.

Networking: An Agenda for the Next Decade

Panelists: Ann C Von Lehmen (NSF, USA); Deepankar Medhi (NSF, USA); Ananthram Swami (Army Research Laboratory, USA); Edward Knightly (Rice University, USA); Leandros Tassiulas (Yale University, USA); Moderator: Srikanth Krishnamurthy (University of California Riverside, USA)

1

Research has always been ahead of commercial/DoD prototype deployments. We have already seen investment in emerging technologies such as software defined networking for cellular, solutions for 360 degrees video, machine learning for networking including federated learning, smart CPS, and proactive in-network solutions to security. However, gaps remain between where these technologies are today, and where we need to be, to see them everywhere. In this panel, we seek to discuss topics that we need to delve into in the near future, and the questions that we need answered, to shape the networks that we will see a decade from now. We have a team of seminal researchers from academia and leaders from DoD and NSF, who will provide their vision towards where they think networking research should be oriented going forward, and what are the critical questions that we as researchers should focus upon.

Speaker

Speaker biography is not available.

Session Chair

Srikanth Krishnamurthy (University of California Riverside, USA)

AGO: Boost Mobile AI Inference Performance by Removing Constraints on Graph Optimization

Zhiying Xu, Hongding Peng and Wei Wang (Nanjing University, China)

0

Traditional deep learning compilers rely on heuristics for subgraph generation, which impose extra constraints on graph optimization, e.g., each subgraph can only contain at most one complex operator. In this paper, we propose AGO, a framework for graph optimization with arbitrary structures to boost the inference performance of deep models by removing such constraints. To create new optimization opportunities for complicated subgraphs, we propose intensive operator fusion, which can effectively stitch multiple complex operators together for better performance. Further, we design a graph partitioning scheme that allows an arbitrary structure for each subgraph while guaranteeing the acyclic property among all generated subgraphs. Additionally, to enable efficient performance tuning on complicated subgraphs, we devise a novel divide-and-conquer tuning mechanism to orchestrate different system components. Through extensive experiments on various neural networks and mobile devices, we show that our system can improve the inference performance by up to 3.3× when compared with state-of-the-art deep compilers.

Speaker Zhiying Xu (Nanjing University)

I am a PhD candidate at Nanjing University, China. I am interested in machine learning system, compilation, and code generation techniques.

OpticNet: Self-Adjusting Networks for ToR-Matching-ToR Optical Switching Architectures

Caio Alves Caldeira (Universidade Federal de Minas Gerais, Brazil); Otavio Augusto de Oliveira Souza and Olga Goussevskaia (UFMG, Brazil); Stefan Schmid (University of Vienna, Austria)

1

Demand-aware reconfigurable datacenter networks can be modeled as a ToR-Matching-ToR (TMT) two-layer architecture, in which each top-of-rack (ToR) is represented by a static switch, and $n$ ToRs are connected by a set of reconfigurable optical circuit switches (OCS). Each OCS internally connects a set of in-out ports via a matching that may be updated at runtime. The matching model is a formalization of such networks, where the datacenter topology is defined by the union of matchings over the set of nodes, each of which can be reconfigured at unit cost.
In this work we propose a scalable matching model for scenarios where OCS have a constant number of ports. Furthermore, we present OpticNet, a framework that maps a set of $n$ static ToR switches to a set of $p$-port OCS to form any constant-degree topology. We prove that OpticNet uses a minimal number of reconfigurable switches to realize any desired network topology and allows to apply any existing self-adjusting network (SAN) algorithm on top of it, also preserving amortized performance guarantees. Our experimental results based on real workloads show that OpticNet is a flexible and efficient framework to design efficient SANs.

Speaker Otávio Augusto de Oliveira Souza (Universidade Federal of Minas Gerais)

Otávio A. de O. Souza received the M.Sc degree in Computer Science in 2020 from Universidade Federal de Minas Gerais (UFMG), Brazil, where he is currently a Ph.D. candidate in Computer Science. His research interest includes modeling, algorithm design, and analysis in communication networks, with emphasis on distributed systems.

An End-to-end Learning Framework for Joint Compensation of Impairments in Coherent Optical Communication Systems

Rui Zhang (University of Electronic Science and Technology of China, China); Min Liao and Jun Chen (Huawei Chengdu Research Center, China); Xusong Ning, Lin Li and Qinli Yang (University of Electronic Science and Technology of China, China); Yongsheng Xu (Huawei Chengdu Research Center, China); Junming Shao (University of Electronic Science and Technology of China, China)

1

The application of machine learning techniques in Coherent Optical Communication (COC) systems has gained increasing attention in recent years. One representative and successful application is to employ neural networks to compensate the signal impairments of devices in the COC system. However, existing studies usually concentrate on each individual device or one impairment, the various impairments sourced from multiple devices (e.g., non-linear distortion, memory, and crosstalk effects) are not well investigated. More importantly, due to the impairment isolation caused by frequency offset, traditional studies only compensate the impairments of the transmitter or receiver individually. Different from previous works, in this paper, we consider a more practical and challenging experimental setup environment: joint compensation of multiple impairments associated with all devices of transmitter and receiver simultaneously. To this end, we propose an end-to-end compensation framework from the transmitter to the receiver in COC systems with three associated modules: an auxiliary channel neural network for impairment modeling, a pre-compensation neural network deployed in the transmitter, and a post-compensation neural network deployed in the receiver. The solution has been successfully verified by the high baud rate (120Gbaud) coherent optical professional test platform and shows impressive optical Signal-to-Noise Ratio (SNR) gains.

Speaker Rui Zhang( University of Electronic Science and Technology of China)

I am a Ph.D. candidate at the School of Computer Science of the USETC under the supervision of Prof. Junimg Shao. My main research area is adversarial machine learning and its application with a particular focus on adversarial defense and the deep learning model's robustness.

Minimizing Age of Information for Underwater Optical Wireless Sensor Networks

Yu Tian, Lei Wang, Chi Lin, Yang Chi, Bingxian Lu and Zhenquan Qin (Dalian University of Technology, China)

0

In Underwater Optical Wireless Sensor Networks (UOWSNs), the Age of information (AoI) is critical as it manifests information freshness, which is extremely useful for real-time monitoring applications. Prior works extensively focus on improving AoI in single-hop or multi-hop networks with fixed paths. However, the issue of multi-path transmission is overlooked. In this paper, we make the first attempt to address the issue of minimizing AoI through link scheduling in UOWSNs (termed MAKE problem). To minimize AoI, we propose an AoI model and corresponding link constraint model for multi-path and multi-hop UOWSNs. Then, we formalize the MAKE problem into a submodular function maximization problem, and propose a greedy method with an approximation ratio guarantee to solve it. Theoretical analyses are presented to explore the features of our scheme. Finally, to demonstrate the effectiveness of the proposed scheme, extensive simulations are conducted. The results reveal that the proposed scheme has 21.4% lower total average AoI compared with other baseline algorithms. Furthermore, test-bed experiments are carried out to verify the applicability of the proposed scheme in practical applications.

Speaker Yu Tian (Dalian University of Technology)

Yu Tian received an M.S. degree from Inner Mongolia University, Hohhot, China, in 2019. He is currently pursuing a Ph.D. degree in software engineering from the Dalian University of Technology, Dalian, China. His research interest focuses on the underwater optical wireless network.

Session Chair

Yan Zhang

A Better Cardinality Estimator with Fewer Bits, Constant Update Time, and Mergeability

Yang Du, He Huang and Yu-e Sun (Soochow University, China); Kejian Li (Soochow University, Hong Kong); Boyu Zhang and Guoju Gao (Soochow University, China)

1

Cardinality estimation is a fundamental problem with diverse practical applications. HyperLogLog (HLL) has become a standard in practice because it offers good memory efficiency, constant update time, and mergeability. Some recent work achieved better memory efficiency, but typically at the cost of impractical update time or losing mergeability, making them incompatible with applications like network-wide traffic measurement. This work presents SpikeSketch, a better cardinality estimator that reduces memory usage of HLL by 37% without sacrificing other crucial metrics. We adopt a bucket-based data structure to promise constant update time, design a smoothed log4 ranking and a spike coding scheme to compress cardinality observables into buckets, and propose a lightweight mergeable lossy compression to balance memory usage, information loss, and mergeability. Then we derive an unbiased estimator for recovering cardinality from the lossy-compressed sketch. We further implement SpikeSketch on the NetFPGA-SUME board. Theoretical and empirical results show that SpikeSketch can work as a drop-in replacement for HLL because it achieves a near-optimal MVP (memory-variance-product) of 4.08 (37% smaller than HLL) with constant update time and mergeability. Its memory efficiency even defeats ACPC and HLLL, the state-of-the-art lossless-compressed sketches using linear-time compression to reduce memory usage.

Speaker Yang Du

Yang Du is currently a postdoctoral fellow in the School of Computer Science and Technology at Soochow University, P. R. China. He received his B.E. degree from Soochow University in 2015 and Ph.D. degree from University of Science and Technology of China in 2020. His research interests include network traffic measurement and sketch.

RecMon: A Deep Learning-based Data Recovery System for Network Monitoring

Huaiyi Zhao (Institute of Computing Technology, Chinese Academy of Sciences, China); Xinyi Zhang (CNIC & Chinese Academy of Sciences, China); Kun Xie (Hunan University, China); Dong Tian (CNIC Chinese Academy of Sciences, China); Gaogang Xie (CNIC Chinese Academy of Sciences & University of Chinese Academy of Sciences, China)

0

Network monitoring systems struggle with the issue that the measurement data is incomplete, with only a subset of origin-destination (OD) pairs or time slots observed, due to the high deployment and measurement cost. Recent studies show that the missing data can be inferred from partial measurements using neural network models and tensor methods. However, these recovery methods fail to achieve accuracy, adaptability and high speed simultaneously. In this paper, we propose RecMon, a deep learning-based data recovery system that satisfies all three criteria. Global spatio-temporal attention and a data augmentation algorithm are proposed to improve model accuracy. A semisupervised learning-based scheme is devised to quickly update the model. We conduct extensive experiments on three real-world datasets to compare RecMon with four state-of-the-art methods in terms of online recovery performance. The experimental results show that RecMon can adapt to the latest state of the network and accurately recover network measurement data in less than 100 milliseconds. When 90% of the data is missing, the recovery accuracy of RecMon improves over the strongest baseline method by 22.7%, 16.0%, and 8.2% in the three datasets, respectively.

Speaker Huaiyi Zhao (Institute of Computing Technology, Chinese Academy of Sciences)

Huaiyi Zhao is a Ph.D candidate at Institute of Computing Technology, Chinese Academy of Sciences. His research interests include network architecture, network measurement and AI for network.

LightNestle: Quick and Accurate Neural Sequential Tensor Completion via Meta Learning

Yuhui Li (Hunan University, China); Wei Liang (Hunan University of Science and Technology, China); Kun Xie, Dafang Zhang and Songyou Xie (Hunan University, China); Kuan-Ching Li (Hunan University of Science and Technology, China)

0

Network operation and maintenance rely heavily on network traffic monitoring. Due to the measurement overhead reduction, lack of measurement infrastructure, and unexpected transmission error, network traffic monitoring systems suffer from incomplete observed data and high data sparsity problems. Recent studies model missing data recovery as a tensor completion task and show good performance. Although promising, the current tensor completion models adopted in network traffic data recovery lack of an effective and efficient retraining scheme to adapt to newly arrived data while retaining historical information. To solve the problem, we propose LightNestle, a novel sequential tensor completion scheme based on meta-learning, which designs (1) an expressive neural network to transfer spatial knowledge from previous embeddings to current embeddings; (2) an attention-based module to transfer temporal patterns into current embeddings in linear complexity; and (3) an meta-learning-based algorithms to iteratively recover missing data and update transfer modules to catch up with learned knowledge. We conduct extensive experiments on two real-world network traffic datasets to assess our performance. The result demonstrates that our proposed methods achieve both fast retraining and high recovery accuracy.

Speaker Yuhui Li (Hunan University)

Excalibur: A Scalable and Low-Cost Traffic Testing Framework for Evaluating DDoS Defense Solutions

Xiang Chen and Hongyan Liu (Zhejiang University, China); Tingxin Sun (Fuzhou University, China); Qun Huang (Peking University, China); Dong Zhang (Fuzhou University, China); Xuan Liu (Yangzhou University & Southeast University, China); Boyang Zhou (Zhejiang Lab, China); Haifeng Zhou (Zhejiang University, China); Chunming Wu (College of Computer Science, Zhejiang University, China)

1

To date, security researchers evaluate their solutions of mitigating denial-of-service (DDoS) attacks via kernel-based or kernel-bypassing testing tools. However, kernel-based tools exhibit poor scalability in attack traffic generation while kernel-bypassing tools introduce unacceptable monetary cost. We propose Excalibur, a scalable and low-cost testing framework for evaluating DDoS defense solutions. The key idea is to leverage the programmable switch to perform testing tasks with Tbps-level scalability and low cost. Specifically, Excalibur offers intent-based primitives to enable academic researchers to customize testing tasks on demand. Moreover, in view of switch resource limitations, Excalibur coordinates both a server and a programmable switch to jointly perform testing tasks. It realizes flexible attack traffic generation, which requires a large number of resources, in the server while using the switch to increase the sending rate of attack traffic to Tbps-level. Our experiments on a 64×100 Gbps Tofino switch demonstrate that Excalibur achieves orders-of-magnitude higher scalability and lower cost than existing tools.

Speaker Xiang Chen

Xiang is a first-year PhD student at Zhejiang University. His advisors are Prof. Chunming Wu, Prof. Qun Huang, and Prof. Dong Zhang. He has received a Best Paper Award from IEEE/ACM IWQoS 2021 and a Best Paper Candidate from IEEE INFOCOM 2021. His research interests include programmable networks, network virtualization, and network security.

Session Chair

Gang Zhou

Secure and Robust Two Factor Authentication via Acoustic Fingerprinting

Yanzhi Ren, Tingyuan Yang and Zhiliang Xia (University of Electronic Science and Technology of China, China); Hongbo Liu (Electronic Science and Technology of China, China); Yingying Chen (Rutgers University, USA); Nan Jiang and Zhaohui Yuan (East China Jiaotong University, China); Hongwei Li (University of Electronic Science and Technology of China, China)

0

The two-factor authentication (2FA) has become pervasive as the mobile devices become prevalent. In this work, we propose a secure 2FA that utilizes the individual acoustic fingerprint of the speaker/microphone on enrolled device as the second proof. The main idea behind our system is to use both magnitude and phase fingerprints derived from the frequency response of the enrolled device by emitting acoustic beep signals alternately from both enrolled and login devices and receiving their direct arrivals for 2FA. Given the input microphone samplings, our system designs an arrival time detection scheme to accurately identify the beginning point of the beep signal from the received signal. To achieve a robust authentication, we develop a new distance mitigation scheme to eliminate the impact of transmission distances from the sound propagation model for extracting stable fingerprint in both magnitude and phase domain. Our device authentication component then calculates a weighted correlation value between the device profile and fingerprints extracted from run-time measurements to conduct the device authentication for 2FA. Our experimental results show that our proposed system is accurate and robust to both random impersonation and Man-in-the-middle (MiM) attack across different scenarios and device models.

Speaker Yingying Chen（Rutgers University）

Yingying Chen is the Department Chair and Professor of Electrical and Computer Engineering and the Peter D. Cherasia Faculty Scholar at Rutgers University. She is the Associate Director of the Wireless Information Network Laboratory (WINLAB). She also leads the Data Analysis and Information Security Laboratory (DAISY). She is a National Academy of Inventors (NAI) Fellow, an Institute of Electrical and Electronics Engineers (IEEE) Fellow, and an Asia-Pacific Artificial Intelligence Association (AAIA) Fellow. She is also named as an ACM Distinguished Scientist. Her background is a combination of Computer Science, Computer Engineering and Physics. She has co-authored three books Securing Emerging Wireless Systems (Springer 2009) and Pervasive Wireless Environments: Detecting and Localizing User Spoofing (Springer 2014) and Sensing Vehicle Conditions for Detecting Driving Behaviors (Springer 2018), and published 240+ journal articles and referred conference papers. She also obtained many patents with multiple of them being licensed and commercialized by industry.

Secur-Fi: A Secure Wireless Sensing System Based on Commercial Wi-Fi Devices

Xuanqi Meng, Jiarun Zhou, Xiulong Liu, Xinyu Tong, Wenyu Qu and Jianrong Wang (Tianjin University, China)

0

Wi-Fi sensing technology plays an important role in numerous IoT applications such as virtual reality, smart homes and elder healthcare. The basic principle is to extract physical features from the Wi-Fi signals to depict the user's locations or behaviors. However, current research focuses more on improving the sensing accuracy but neglects the security concerns. Specifically, current Wi-Fi router usually transmits a strong signal, so that we can access the Internet even through the wall. Accordingly, the outdoor adversaries are able to eavesdrop on this strong Wi-Fi signal, and infer the behavior of indoor users in a non-intrusive way, while the indoor users are unaware of this eavesdropping. To prevent outside eavesdropping, we propose Secur-Fi, a secure Wi-Fi sensing system. Our system meets the following two requirements: (1) we can generate fraud signals to block outside unauthorized Wi-Fi sensing; (2) we can recover the signal, and enable authorized Wi-Fi sensing. We implement the proposed system on commercial Wi-Fi devices and conduct experiments in three applications including passive tracking, behavior recognition, and breath detection. The experiment results show that our proposed approaches can reduce the accuracy of unauthorized sensing by 130% (passive tracking), 72% (behavior recognition), 86% (breath detection).

Speaker Xuanqi Meng(Tianjin University)

I Can Hear You Without a Microphone: Live Speech Eavesdropping From Earphone Motion Sensors

Yetong Cao and Fan Li (Beijing Institute of Technology, China); Huijie Chen (Beijing University of Technology, China); Xiaochen Liu and Chunhui Duan (Beijing Institute of Technology, China); Yu Wang (Temple University, USA)

0

Recent literature advances motion sensors mounted on smartphones and AR/VR headsets to speech eavesdropping due to their sensitivity to subtle vibrations. The popularity of motion sensors in earphones has fueled a rise in their sampling rate, which enables various enhanced features. This paper investigates a new threat of eavesdropping via motion sensors of earphones by developing EarSpy, which builds on our observation that the earphone's accelerometer can capture bone conduction vibrations (BCVs) and ear canal dynamic motions (ECDMs) associated with speaking; they enable EarSpy to derive unique information about the wearer's speech. Leveraging a study on the motion sensor measurements captured from earphones, EarSpy gains abilities to disentangle the wearer's live speech from interference caused by body motions and vibrations generated when the earphone's speaker plays audio. To enable user-independent attacks, EarSpy involves novel efforts, including a trajectory instability reduction method to calibrate the waveform of ECDMs and a data augmentation method to enrich the diversity of BCVs. Moreover, EarSpy explores effective representations from BCVs and ECDMs, and develops a convolutional neural model with Connectionist Temporal Classification (CTC) to realize accurate speech recognition. Extensive experiments involving 14 participants demonstrate that EarSpy reaches a promising recognition for the wearer's speech.

Speaker Yetong Cao

Yetong Cao is a PhD student in the Wireless and Mobile Computing Lab supervised by Prof. Fan Li in the School of Computer Science, Beijing Institute of Technology. She also works with Prof. Jun Luo in the School of Computer Science and Engineering, Nanyang Technological University. Her research fields include Human physiological signals sensing and mobile/wearable computing.

HeartPrint: Passive Heart Sounds Authentication Exploiting In-Ear Microphones

Yetong Cao (Beijing Institute of Technology, China); Chao Cai (Huazhong University of Science and Technology, China); Fan Li (Beijing Institute of Technology, China); Zhe Chen (China-Singapore International Joint Research Institute, China); Jun Luo (Nanyang Technological University, Singapore)

0

Biometrics has been increasingly integrated into wearable devices to enhance security in recent years. Meanwhile, the popularity of wearables in turn creates a unique opportunity for capturing novel biometrics leveraging various embedded sensing modalities. In this paper, we study a new biometrics combining the uniqueness of heart motion, bone conduction, and body asymmetry. Specifically, we design \pname as a passive yet secure user authentication system: it exploits the bone-conducted heart sounds captured by (widely available) dual \textit{in-ear microphone}s (IEMs) to authenticate users, while neatly leveraging IEMs renders itself transparent to users without impairing earphones' normal functions. To suppress the interference from other body sounds and audio produced by the earphones, we develop a interference elimination method using modified non-negative matrix factorization to separate heart sounds from background interference. We further explore the uniqueness of IEM-recorded heart sounds in three aspects to extract a biometric representation, based on which \pname leverages a convolutional neural model equipped with a continual learning method to achieve accurate authentication under drifting body conditions. Extensive experiments with 18 pairs of commercial earphones on 45 participants confirm that \pname can achieve 1.6% FAR and 1.8% FRR, while effectively coping with major attacks, complicated interference, and hardware diversity.

Speaker Yetong Cao

Yetong Cao is a PhD student in the Wireless and Mobile Computing Lab supervised by Prof. Fan Li in the School of Computer Science, Beijing Institute of Technology. She also works with Prof. Jun Luo in the School of Computer Science and Engineering, Nanyang Technological University. Her research fields include Human physiological signals sensing and mobile/wearable computing.

Session Chair

Tamer Nadeem

Poster: Digital Network Twin via Learning-Based Simulator

Yuru Zhang (University of Nebraska Lincoln, USA); Yongjie Xue and Qiang Liu (University of Nebraska-Lincoln, USA); Nakjung Choi (Nokia & Bell Labs, USA)

0

Digital network twin (DNT) allows network operators to test their network management policy before their actual deployment in real-world networks. Achieving DNT, however, can be challenging and compute-intensive if every detail needs to be replicated exactly. In this work, we propose a new compute-efficient approach to realize DNT by augmenting existing network simulators. First, we build a real-world testbed by using OpenAirInterface and replicate its settings with the NS-3 simulator. Second, we observe the non-trivial distributional discrepancy between the simulator and the real-world testbed. Third, we use deep learning techniques to bridge the sim-to-real discrepancy under different network states. The experimental results show our method can reduce up to 91% sim-to-real discrepancy.

Speaker

Speaker biography is not available.

The Architectural Design of Service Management and Orchestration in 6G Communication Systems

Mohammad Asif Habibi (University of Kaiserslautern, Germany); Adrián Gallego Gallego Sánchez and Ignacio Labrador Pavon (Atos Research and Innovation, Spain); Bin Han (RPTU Kaiserslautern-Landau, Germany); Pablo Serrano and Jesús Pérez-Valero (Universidad Carlos III de Madrid, Spain); Antonio Virdis (University of Pisa, Italy); Hans D. Schotten (University of Kaiserslautern, Germany)

0

In this poster paper, we propose and demonstrate an architectural framework for service Management and Orchestration (M&O) in Sixth-Generation (6G) communication systems. This architecture was designed by the Hexa-X project, which is a European flagship project dedicated to developing a vision and technological enablers for 6G. To provide a comprehensive and high-level description, we consider three views: (i) Functional View; (ii) Structural View; and (iii) Deployment View. We first discuss 6G service M&O before delving deeper into each view.

Speaker Mohammad Asif Habibi

Speaker biography is not available.

Meta Learning for Meta-Surface: A Fast Beamforming Method for RIS-Assisted Communications Adapting to Dynamic Environments

Qinpei Luo and Boya Di (Peking University, China)

1

Recently reconfigurable intelligent surface~(RIS) has been proposed as a promising technique to enhance the capacity of wireless networks by reshaping the electromagnetic characteristics of the environment. However, given numerous RIS elements, it is non-trivial to design an efficient beamforming scheme especially for the real-time mobile applications that require fast response to varying environments. In this paper, aiming to maximize the sum rate of a multi-user system via the RIS-enabled beamforming design, a meta-critic network is proposed to recognize the environment change and automatically perform the self-updating of the learning model. We also develop a stochastic Explore and Reload procedure to alleviate the high-dimensional action space issue. Simulation results demonstrate that the proposed scheme can converge to a higher sum rate more rapidly compared to the state-of-the-art methods in dynamic settings. The robustness of our proposed scheme against different RIS sizes is also verified.

Speaker Qinpei Luo

Speaker biography is not available.

Vetting Privacy Policies in VR: A Data Minimization Principle Perspective

Yuxia Zhan and Yan Meng (Shanghai Jiao Tong University, China); Lu Zhou (Xidian University, China); Haojin Zhu (Shanghai Jiao Tong University, China)

0

Virtual Reality is thought to be the prototype of the next-generation Internet, consisting of more I/O devices and interactive methods than traditional mobile systems. Hence VR developers need to inform users what data is collected and shared, which is generally conveyed by privacy policies. Existing research has examined the consistency between the VR app's privacy policy and its corresponding actual behaviors. However, few studies paid attention to the data minimization principle, i.e., whether a privacy policy claims to collect no more data than it practically needs to implement the app's functionalities. In this poster, we targeted a mainstream VR platform and analyzed the data minimization principle compliance of privacy policies for all 1,726 VR apps in this platform. Experiment results show that 48.1% VR apps potentially violate the data minimization principle. Moreover, the comparative experiments reveal significant differences in the distribution of data collection between VR and non-VR apps.

Speaker Yuxia Zhan (Shanghai Jiao Tong University, China)

Yuxia Zhan is a second-year master student at Shanghai Jiao Tong University. Her research interests include security and privacy issues in virtual reality.

Towards Robust Pedestrian Detection with Roadside Millimeter-Wave Infrastructure

Hem Regmi, Vansh Nagpal and Sanjib Sur (University of South Carolina, USA)

0

We present MilliPED, a system that uses a millimeter-wave device to identify pedestrians at traffic intersections and enhance road safety during inclement weather, such as low visibility and heavy rain, when vision cameras are ineffective. We evaluate it with 3000 millimeter-wave reflection samples of pedestrian crossing traffic intersections and show that accurate pedestrian detection is feasible with millimeter-wave devices.

Speaker Hem Regmi

I am currently working as a Graduate Research Assistant at SyReX Lab under the supervision of Dr. Sanjib Sur at the Department of Computer Science & Engineering, University of South Carolina (UofSC). My research interest includes Deep Learning for Imaging Classification, Generative Adversarial Networks (GANs) for Millimeter Wave, Autonomous Driving, and Artificial Intelligence. Before joining UofSC, I worked as Controls Engineer at Tesla, Inc for 2 years. I have completed my M.Sc. in Electrical Engineering from the University of Toledo, Ohio, the USA in, and my undergraduate degree in Electronics and Communication from Tribhuvan University, Nepal. Please look at my resume for detail and follow my research works on Github.

Value of Updates: Which Packets Are Worth Transmitting?

Polina Kutsevol (Technische Universität München, Germany); Onur Ayan (Technical University of Munich, Germany); Wolfgang Kellerer (Technische Universität München, Germany)

1

In the context of control systems over a communication network, status updates can be discarded based on their content to unload the network and prevent network congestion. In this work, we propose a transport layer scheme that not only considers the current system state, but also its significance w.r.t already transmitted updates, including those that are not yet acknowledged. The benefit of admitting a packet to be sent is compared to its transmission cost to obtain the value of update (VoU). Using Zolertia Re-Mote sensors, we show that the consideration of VoU allows improving the control performance by at least $70\%$.

Speaker Polina Kutsevol

POLINA KUTSEVOL received the B.Sc. degree in applied mathematics and physics from the Moscow Institute of Physics and Technology, Moscow, Russia, in 2019, and the M.Sc. degree in communication engineering from the Technical University of Munich, Munich, Germany, in 2021, where she is currently pursuing the Ph.D. degree with the Chair of Communication Networks. Her current research interests include resource management for wireless and mobile communication networks, cyber-physical systems, and networked control systems.

GNN for Wireless Link Anomaly Detection

Blaz Bertalanic (Jozef Stefan Institute, Slovenia); Mihael Mohorcic (Jozef Stefan Institute & Jozef Stefan International Postgraduate School, Slovenia); Carolina Fortuna (Jozef Stefan Institute, Slovenia)

0

In this paper, we present a new approach for detecting wireless link layer anomalies in large-scale IoT networks based on graph neural networks (GNN). We propose a method that transforms time series data into graphs with Markov Transition Field representation. The transformed data is then used to train a new GNN architecture that can successfully distinguish between 4 different link-layer anomalies and outperforms state-of-the-art shallow and deep learning methods.

Speaker Blaz Bertalanic

Blaz Bertalanic received his Master's degree in electrical engineering from the Faculty of Electrical engineering, University of Ljubljana. He is currently pursuing his PhD at the same faculty and is working as a researcher at the Department of Communication Systems, Jožef Stefan Institute. His main research interests are in solving classification problems with the help of machine learning and AI, wireless networking, electronics, and signal processing.

Data Transport for the Orbiting Internet

Aiden David Valentine (184 Iron Mill Lane & University of Sussex, United Kingdom (Great Britain)); George Parisis (University of Sussex, United Kingdom (Great Britain))

1

In this paper, we introduce Orbiting TCP (OrbTCP), a novel multipath data transport protocol for Low Earth Orbit (LEO) satellite networks. OrbTCP utilises in-network telemetry (INT) to obtain per-hop congestion information for each of its active subflows running on edge-disjoint paths. As a result, OrbTCP (1) enables network operation with low buffer capacity and low latency for end-hosts; (2) maximises application throughput and network utilisation; and (3) swiftly reacts to network hotspots due to bursty traffic or path reconfiguration. In this paper, we present early results showcasing the limitations of state-of-the-art data transport in LEO satellite networks, motivate the need for a novel data transport protocol and offer initial evidence that OrbTCP could overcome the identified limitations.

Speaker

Speaker biography is not available.

ARES-WiGR: An Attention-enhanced ResNet based Wi-Fi Gesture Recognition

Kexin Yao and Han Li (Beijing Jiaotong University, China); Ming Liu (Beijing Jiaotong University & Beijing Key Lab of Transportation Data Analysis and Mining, China); Bo Gao and Ke Xiong (Beijing Jiaotong University, China); Pingyi Fan (Tsinghua University, China)

0

This paper proposes an Attention-enhanced ResNet based Wi-Fi Gesture Recognition (ARES-WiGR), in which it first extracts Doppler Frequency Shift (DFS) vector parameters from Channel State Information (CSI) via conjugate matrix multiplication and antenna pair selection. Then, the DFS spectrogram is obtained by Short Time Fourier Transform (STFT), and the DFS spectrogram features are used as the input of the proposed neural network model. Moreover, with the attention mechanism, ARES-WiGR is able to automatically recognize the important information and achieve better recognition effect. The performance is examined in real environment, which shows that the proposed ARES-WiGR effectively extracts gesture features and improves recognition accuracy to about 96%.

Speaker

Speaker biography is not available.

Adversarial Attack and Defense for WiFi-based Apnea Detection System

Harshit Ambalkar (California State University, Sacramento, USA); Tianya Zhao and Xuyu Wang (Florida International University, USA); Shiwen Mao (Auburn University, USA)

0

WiFi sensing systems have gained enormous interest in extensive areas, including vital sign monitoring. By using deep neural networks (DNNs), WiFi sensing systems can perform very well. However, the security and vulnerability of DNNs under adversarial attack would greatly influence the WiFi sensing performance. In this paper, we develop a DNN-based apnea detection system using WiFi channel state information (CSI) and then evaluate its robustness under three different attacks. The experimental results show that adversarial attacks can significantly impact the model performance, and the defense scheme (i.e. adversarial training) can improve the system robustness.

Speaker

Speaker biography is not available.

On the Capacity Region of a Quantum Switch with Entanglement Purification

Nitish K. Panigrahy (Yale University, USA); Thirupathaiah Vasantam (Durham University, United Kingdom (Great Britain)); Don Towsley (University of Massachusetts at Amherst, USA); Leandros Tassiulas (Yale University, USA)

0

Quantum switches are envisioned to be an integral component of future quantum networks. They can provide high quality entanglement distribution service to end-users by performing quantum operations such as entanglement swapping and entanglement purification. In this work, we characterize the capacity region of such a quantum switch under noisy channel transmissions and imperfect quantum operations. These regions describe the set of achievable end-to-end entanglement request rates that the switch can support with finite request backlogs under any entanglement scheduling policy. We express the capacity region as a function of the channel and network parameters, entanglement purification yield and application level parameters. In particular, we provide necessary conditions to verify if a set of request rates belong to the capacity region of the switch. We use these conditions to find the maximum achievable end-to-end user entanglement generation throughput by solving a set of linear optimization problems. We also develop a max-weight scheduling policy and prove that the policy stabilizes the switch for all feasible request arrival rates. From numerical experiments, we discover that performing link-level purification followed by entanglement swaps provides a larger capacity region than doing entanglement swaps followed by purification.

Speaker Nitish Kumar Panigrahy (University of Massachusetts Amherst)

Dr. Nitish Kumar Panigrahy is currently a postdoctoral researcher at NSF ERC Center for Quantum Networks, working jointly with Prof. Leandros Tassiulas (Yale University) and Prof. Don Towsley (University of Massachusetts Amherst). He earned his PhD degree in Computer Science at University of Massachusetts Amherst in 2021. Nitish’s research interests lie in modeling, optimization, and performance evaluation of networked systems with applications to Internet of Things (IoT), cloud computing, content delivery, and quantum information networking.

Qubit Allocation for Distributed Quantum Computing

Yingling Mao, Yu Liu and Yuanyuan Yang (Stony Brook University, USA)

0

With the advancements in quantum communication, optically connected quantum processors can form a distributed quantum computing system. Distributed quantum computing provides a scalable path to execute more complicated computational tasks that a single quantum processor cannot handle. Yet, distributed quantum computing needs a new compiler to map logical qubits of a quantum circuit to different quantum processors in the system. This paper formulates and studies the qubit allocation problem for distributed quantum computing (QA-DQC). We prove the NP-hardness of the formulated problem. Moreover, we show there is no polynomial-time n^a-approximation algorithm for any a<1 unless P=NP, where n is the number of processors in the quantum network. We first propose a heuristic local search algorithm for QA-DQC. Furthermore, we design a multistage hybrid simulated annealing algorithm (MHSA) by combining the local search algorithm and a simulated annealing meta-heuristic algorithm. Lastly, we perform extensive simulations to evaluate the proposed MHSA algorithm under various real quantum circuits and different network topologies. Results show that MHSA outperforms popular baselines.

Speaker Yingling Mao (Stony Brook University)

Yingling Mao received her B.S. degree in Mathematics and Applied Mathematics in Zhiyuan College from Shanghai Jiao Tong University, Shanghai, China, in 2018. She is currently working toward the Ph.D degree in the Department of Electrical and Computer Engineering, Stony Brook University. Her research interests include network function virtualization, edge computing, cloud computing and quantum networks.

A Quantum Overlay Network for Efficient Entanglement Distribution

Shahrooz Pouryousef (University of Massachusetts Amherst MA, USA); Nitish K. Panigrahy (Yale University, USA); Don Towsley (University of Massachusetts at Amherst, USA)

2

Distributing quantum entanglements over long distances is essential for the realization of a global scale quantum Internet. Most of the prior work and proposals assume an on-demand distribution of entanglements which may result in significant network resource under-utilization. In this work, we introduce Quantum Overlay Networks (QONs) for efficient entanglement distribution in quantum networks. When the demand to create end-to-end user entanglements is low, QONs can generate and store maximally entangled Bell pairs (EPR pairs) at specific overlay storage nodes of the network. Later, during peak demands, requests can be served by performing entanglement swaps either over a direct path from the network or over a path using the storage nodes. We solve the link entanglement and storage resource allocation problem in such a QON using a centralized optimization framework. We evaluate the performance of our proposed QON architecture over a wide number of network topologies under various settings using extensive simulation experiments. Our results demonstrate that QONs fare well by a factor of 40% with respect to meeting surge and changing demands compared to traditional non-overlay proposals. QONs also show significant improvement in terms of average entanglement request service delay over non-overlay approaches.

Speaker Shahrooz Pouryousef (UMass Amherst)

Shahrooz is a PhD student in the College of Information and Computer Sciences (CICS) at UMass Amherst working with Prof. Don Towsley. His research interests lie in classical and quantum networks.

Asynchronous Entanglement Provisioning and Routing for Distributed Quantum Computing

Lan Yang, Yangming Zhao and Liusheng Huang (University of Science and Technology of China, China); Chunming Qiao (University at Buffalo, USA)

0

In Distributed Quantum Computing (DQC), quantum bits (qubits) used in a quantum circuit may be distributed on multiple Quantum Computers (QCs) connected by a Quantum Data Network (QDN). To perform a quantum gate operation involving two qubits on different QCs, we have to establish an Entanglement Connection (EC) between their host QCs. Existing EC establishment schemes result in a long EC establishment time, and low quantum resource utilization.

In this paper, we propose an Asynchronous Entanglement Routing and Provisioning (AEPR) scheme to minimize the task completion time in DQC systems. AEPR has three distinct features: (i). Entanglement Paths (EPs) for a given SD pair are predetermined to eliminate the need for runtime calculation; (ii). Entanglement Links (ELs) are established proactively to reduce the time needed to successfully create ELs; and (iii). For a given EC request, quantum swapping along an EP is performed by a repeater whenever two adjacent ELs are created, so precious quantum resources at the repeater can be released immediately thereafter for other ELs and ECs. Extensive simulations show that AEPR can save up to 76.05% of the average task completion time in DQC systems compared with the state-of-the-art entanglement routing schemes designed to maximize QDN throughput.

Speaker Lan Yang (University of Science and Technology of China)

Lan Yang is a Master student in the department of Computer Science and Technology, University of Science and Technology of China. She received the bachelor’s degree in University of Science and Technology of China in 2022. Her research interests include network optimization, traffic engineering, and quantum networks. Currently, she is primarily working with Prof. Yangming Zhao on routing protocol design in quantum networks.

Session Chair

Jianqing Liu

More than Enough is Too Much: Adaptive Defenses against Gradient Leakage in Production Federated Learning

Fei Wang, Ethan Hugh and Baochun Li (University of Toronto, Canada)

0

With increasing concerns on privacy leakage from gradients, a variety of attack mechanisms emerged to recover private data from gradients at an honest-but-curious server, which challenged the primary advantage of privacy protection in federated learning. However, we cast doubt upon the real impact of these gradient attacks on production federated learning systems. By taking away several impractical assumptions that the literature has made, we find that gradient attacks pose a limited degree of threat to the privacy of raw data.

Through a comprehensive evaluation on existing gradient attacks in a federated learning system with practical assumptions, we have systematically analyzed their effectiveness under a wide range of configurations. We present key priors required to make the attack possible or stronger, such as a narrow distribution of initial model weights, as well as inversion at early stages of training. We then propose a new lightweight defense mechanism that provides \emph{sufficient} and \emph{self-adaptive} protection against time-varying levels of the privacy leakage risk throughout the federated learning process. Our experimental results demonstrate that \textsc{Outpost} can achieve a much better tradeoff than the state-of-the-art with respect to convergence performance, computational overhead, and protection against gradient attacks.

Speaker Jointly Presented by Fei Wang and Baochun Li (University of Toronto)

Fei Wang is a second-year Ph.D. student at the Edward S. Rogers Sr. Department of Electrical & Computer Engineering, University of Toronto, Canada, under the supervision of Prof. Baochun Li. She received her B.E. degree with honours from Hongyi Honor College, Wuhan University, China. Her research interests lie at the intersections of networking and communication and machine learning, especially deep reinforcement learning and federated learning. Her personal website is located at silviafeiwang.github.io.

Baochun Li is currently a Professor at the Department of Electrical and Computer Engineering, University of Toronto. He is a Fellow of IEEE.

Truthful Incentive Mechanism for Federated Learning with Crowdsourced Data Labeling

Yuxi Zhao, Xiaowen Gong and Shiwen Mao (Auburn University, USA)

0

Federated learning (FL) has emerged as a promising paradigm that trains machine learning (ML) models on clients' devices in a distributed manner without the need of transmitting clients' data to the FL server. In many applications of ML, the labels of training data need to be generated manually by human agents. In this paper, we study FL with crowdsourced data labeling where the local data of each participating client of FL are labeled manually by the client. We consider the strategic behavior of clients who may not make desired effort in their local data labeling and local model computation and may misreport their local models to the FL server. We characterize the performance bounds on the training loss as a function of clients' data labeling effort, local computation effort, and reported local models. We devise truthful incentive mechanisms which incentivize strategic clients to make truthful efforts and report true local models to the server. The truthful design exploits the non-trivial dependence of the training loss on clients' efforts and local models. Under the truthful mechanisms, we characterize the server's optimal local computation effort assignments. We evaluate the proposed FL algorithms with crowdsourced data labeling and the incentive mechanisms using experiments.

Speaker Yuxi Zhao (Auburn University)

Yuxi Zhao is a PhD graduated from the department of Electrical and Computer Engineering at Auburn University, USA. Her main research interests include federated learning and data crowdsourcing. She received IEEE INFOCOM’21 Student Travel Grant. She is a member of the IEEE, IEEE Young Professionals, and IEEE Communications Society.

SVDFed: Enabling Communication-Efficient Federated Learning via Singular-Value-Decomposition

Haolin Wang, Xuefeng Liu and Jianwei Niu (Beihang University, China); Shaojie Tang (University of Texas at Dallas, USA)

0

Federated learning (FL) is an emerging paradigm of distributed machine learning. However, when applied to wireless network scenarios, FL usually suffers from high communication cost because clients need to upload their updated gradients to a server in every training round. Although many gradient compression techniques like sparsification and quantization are proposed, they compress clients' gradients independently, without considering the correlations among gradients. In this paper, we propose SVDFed, a collaborative gradient compression framework for FL. SVDFed utilizes Singular Value Decomposition (SVD) to find a few basis vectors, whose linear combination can well represent clients' gradients at a certain round. Due to the correlations of gradients, these basis vectors can still well approximate clients' gradients in subsequent rounds. Therefore, clients can only transmit the coefficients of linear combination to the server, which greatly reduces communication cost. In addition, SVDFed leverages the classical PID (Proportional, Integral, Derivative) control to determine the proper time to re-calculate the basis vectors to maintain their representation ability. Through experiments, we demonstrate that SVDFed outperforms existing gradient compression methods in FL. For example, compared to a popular gradient quantization method QSGD, SVDFed can reduce the communication overhead by 66 \% and pending time by 99%.

Speaker Haolin Wang (Beihang University)

Haolin Wang received the B.S. degree in Computer Science and Engineering from Beihang University, Beijing, China, in 2022. He is currently working toward the M.S. degree in Computer Science and Engineering in Beihang University, Beijing, China. His research interests include Federated Learning.

Enabling Communication-Efficient Federated Learning via Distributed Compressed Sensing

Yixuan Guan and Xuefeng Liu (Beihang University, China); Tao Ren (Institute of Software Chinese Academy of Sciences, China); Jianwei Niu (Beihang University, China)

0

Federated learning (FL) trains a shared global model by periodically aggregating gradients from local devices. Communication overhead becomes a principal bottleneck in FL since participating devices usually suffer from limited bandwidth and unreliable connections in the uplink transmission. To address the problem, gradient compression based on compressed sensing (CS) has been put forward recently. However, most existing CS-based works compress gradients independently, ignoring the gradient correlations between participants or adjacent rounds, which constrains the achievement of higher compression rates. In view of this observation, we propose a novel gradient compression scheme named Federated Distributed Compressed Sensing (FedDCS), guided by the distributed compressed sensing theory (DCS). FedDCS can fully exploit the correlated gradients from previous rounds, known as side information, to assist the gradient reconstruction currently. Benefiting from this, the reconstruction performance is significantly improved on errors and iterations under the identical compression rate, and the total uploading bits to achieve convergence are considerably reduced. Theoretical analysis and extensive experiments conducted on MNIST and Fashion-MNIST both verify the effectiveness of our approach.

Speaker Yixuan Guan (Beihang University)

Yixuan Guan received the B.S. degree from college of communication engineering at Jilin University, Changchun, China, in 2016, and the M.S. degree from school of electronic and information engineering at South China University of Technology, Guangzhou, China, in 2020. He is currently pursuing the Ph.D. degree from school of computer science and engineering at Beihang University, Beijing, China. His research interests include Federated Learning and Data Compression.

Session Chair

Changqing Luo

Flowrest: Practical Flow-Level Inference in Programmable Switches with Random Forests

Aristide Tanyi-Jong Akem (IMDEA Networks Institute, Spain & Universidad Carlos III de Madrid, Spain); Michele Gucciardo and Marco Fiore (IMDEA Networks Institute, Spain)

2

User-plane machine learning enables low-latency and high-throughput inference at line rate. Yet, data planes are highly constrained environments, and restrictions are especially marked in programmable switches with limited available memory and minimum support for mathematical operations or data types. As a result, current solutions for in-switch inference that are compatible with production-level hardware lack support for complex features or suffer from limited scalability, which creates performance barriers in complex tasks involving large decision spaces. We address this limitation, and present Flowrest, a first complete Random Forest (RF) model implementation that can operate at the level of individual flows in commercial switches. Our solution builds on (i) a novel framework to embed generic flow-level machine learning models into programmable switch ASICs, and (ii) original guidelines for tailoring RF models to operations in programmable switches already at the design stage. We implement Flowrest as open-source software using the P4 language, and assess its performance in an experimental platform based on Intel Tofino switches. Tests with tasks of unprecedented complexity show how our model can improve accuracy by up to 39% over previous approaches to implement RF models in real- world equipment.

Speaker Aristide Tanyi-Jong Akem (IMDEA Networks Institute & Universidad Carlos III de Madrid)

Akem is a PhD student in the Networks Data Science Group at IMDEA Networks Institute in Madrid, Spain. He is also a student at Universidad Carlos III de Madrid, where he is enrolled in the Telematics Engineering program. Prior to his PhD studies, he completed an engineering degree at the University of Yaounde I, in Cameroon and a master's in electrical and computer engineering at Carnegie Mellon University Africa, in Rwanda. He is currently involved with the European Union's Horizon 2020 project "BANYAN" which aims to bring big data analytics to radio access networks. At the moment, he is visiting Orange Labs in Paris, France as part of the secondments of the project. Akem's current research interest is in the area of in-band network intelligence, with a focus on in-network machine learning.

Melody: Toward Resource-Efficient Packet Header Vector Encoding on Programmable Switches

Xiang Chen and Hongyan Liu (Zhejiang University, China); Qingjiang Xiao and Jianshan Zhang (Fuzhou University, China); Qun Huang (Peking University, China); Dong Zhang (Fuzhou University, China); Xuan Liu (Yangzhou University & Southeast University, China); Chunming Wu (College of Computer Science, Zhejiang University, China)

1

Emerging programmable switches provision a limited capacity of packet header vector (PHV) resources to support the user-specified logic of manipulating packet header fields and metadata fields in network functions. However, existing switch compilers typically employ inefficient strategies of encoding fields onto PHV words. These strategies significantly waste scarce PHV resources, leading to possible failures when offloading network functions. In this paper, we propose Melody, a framework that aims to achieve resource efficiency in PHV encoding. The key idea of Melody is to reuse PHV words for as many fields as possible to improve resource efficiency. To realize this idea, we design both a field analyzer and an optimization framework in Melody. The field analyzer automatically identifies which fields can reuse PHV words while preserving the original packet processing logic, avoiding laborious user burdens. The optimization framework integrates analysis results into PHV encoding to produce the resource-optimal decisions. We have implemented Melody and evaluated it with real network functions and large-scale synthetic ones. Our results indicate that compared to existing compilers, Melody improves the resource efficiency by up to 85%.

Speaker Xiang Chen

Xiang is a first-year PhD student at Zhejiang University. His advisors are Prof. Chunming Wu, Prof. Qun Huang, and Prof. Dong Zhang. He has received a Best Paper Award from IEEE/ACM IWQoS 2021 and a Best Paper Candidate from IEEE INFOCOM 2021. His research interests include programmable networks, network virtualization, and network security.

CLIP: Accelerating Features Deployment for Programmable Switch

Tingting Xu, Xiaoliang Wang and Chen Tian (Nanjing University, China); Yun Xiong and Yun Lin (HUAWEI, China); Baoliu Ye (Nanjing University, China)

0

Cloud network serves a large amount of tenants and a variety of applications. The continuous changing demands require a programmable data plane to achieve fast feature velocity. However, the years-long release cycle of traditional function-fixed switches can not meet this requirement. Emerging programmable switches provide the flexibility of packet processing without sacrificing hardware performance. Due to the trade-off between performance and flexibility, the current programmable switches make compromises in some aspects such as limited memory/computation resources, lack of the capacity to realize complicated computation. The programmable switches can not satisfy the demand of network services and applications in production networks. We propose a framework that leverages host servers to extend the capability of network switches quickly, accelerates new feature deployment, and verifies new ideas in production networks. Specifically, to build the unified programmable data plane, we address essential design and implementation challenges including a programming abstraction that allows automatically and effectively deploy network functions on switch and server clusters, allocating traffic to fully utilize the server resources, and supporting flexible scaling of the system. The quick deployment of several self-defined functions in realistic system have verified the feasibility and practicality of the proposed framework.

Speaker Tingting Xu (Nanjing University)

Tingting Xu received the B.E. degree in 2019 from the college of Computer Science and Electronic Engineering, Hunan University, Hunan, China. She is currently working toward the Ph.D. degree in the Department of Computer Science and Technology, Nanjing University under the supervision of Prof.Xiaoliang Wang. Her research interests include programmable network, datacenter network, and network function virtulization.

RED: Distributed Program Deployment for Resource-aware Programmable Switches

Xingxin Jia, Fuliang Li and Songlin Chen (Northeastern University, China); Chengxi Gao (Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, China); Pengfei Wang (Dalian University of Technology, China); Xingwei Wang (Northeastern University, China)

0

Programmable switches allow data plane to program how packets are processed, which enables flexibility for network management tasks, e.g., packet scheduling and flow measurement. Existing studies focus on program deployment at a single switch， while deployment across the whole data plane is still a challenging issue. In this paper, we present RED, a Resource-Efficient and Distributed program deployment solution for programmable switches. First of all, we compile the data plane programs to estimate the resource utilization and divide them into two categories that wait to be further processed. Then, the proposed merging and splitting algorithms are selectively applied to merge or split the pending programs. Finally, we consolidate the scarce resources of the whole data plane to deploy the programs. Extensive experiment results show that 1) RED improves the speedup by two orders of magnitude compared to P4Visor and merges 58.64% more nodes than SPEED; 2) RED makes the overwhelmed programs run normally at a single switch and meanwhile reduces 3% latency of inter-device scheduling; 3) RED achieves network-wide resource balancing in a distributed way.

Speaker Xingxin Jia(Northeastern University, China)

Session Chair

Patrick P. C. Lee

Minimizing Age of Information in Spatially Distributed Random Access Wireless Networks

Nicholas W Jones and Eytan Modiano (MIT, USA)

0

We analyze Age of Information (AoI) in wireless networks where nodes use a spatially adaptive random access scheme to send status updates to a central base station. We show that the set of achievable AoI in this setting is convex, and design policies to minimize weighted sum, min-max, and proportionally fair AoI by setting transmission probabilities as a function of node locations. We show that under the capture model, when the spatial topology of the network is considered, AoI can be significantly improved, and we obtain tight performance bounds on weighted sum and min-max AoI. Finally, we design a policy where each node sets its transmission probability based only on its own distance from the base station, when it does not know the positions of other nodes, and show that it converges to the optimal proportionally fair policy as the size of the network goes to infinity.

Speaker Nicholas Jones (MIT)

Nicholas Jones is a PhD candidate at MIT in the Laboratory for Information and Decision Systems, advised by Professor Eytan Modiano. He is interested in optimizing control of wireless networks for real-time and delay-sensitive applications.

Fresh-CSMA: A Distributed Protocol for Minimizing Age of Information

Vishrant Tripathi, Nicholas W Jones and Eytan Modiano (MIT, USA)

0

We consider the design of distributed scheduling policies that minimize age of information in single-hop wireless networks. The centralized max-weight policy is known to be nearly optimal in this setting. Hence, our goal is to design a distributed CSMA scheme that can mimic its performance. To that end, we propose a distributed protocol called Fresh-CSMA and show that in an idealized setting, our protocol can match the scheduling decisions of the max-weight policy with high probability in each time-slot, and also match the theoretical performance guarantees of the max-weight policy over the entire time horizon. We then consider a more realistic setting and study the impact of protocol parameters on the probability of collisions and the overhead caused by the distributed nature of the protocol. Finally, we provide simulations that support our theoretical results and show that the performance gap between the ideal and realistic versions of Fresh-CSMA is small.

Speaker Vishrant Tripathi (MIT)

Vishrant Tripathi is a Ph.D. candidate in the EECS department at MIT, working with Prof. Eytan Modiano at the Laboratory for Information and Decision Systems (LIDS). His research is on modeling, analysis and design of communication networks, with emphasis on wireless and real-time networks. His current focus is on scheduling problems in networked control systems, multi-agent robotics and federated learning.

Age of broadcast and collection in spatially distributed wireless networks

Chirag Rao (US Army Research Laboratory & Massachusetts Institute of Technology, USA); Eytan Modiano (MIT, USA)

0

We consider a wireless network with a base station broadcasting and collecting time-sensitive data to and from spatially distributed nodes in the presence of wireless interference. The Age of Information (AoI) is the time that has elapsed since the most-recently delivered packet was generated, and captures the freshness of information. In the context of broadcast and collection, we define the Age of Broadcast (AoB) to be the amount of time elapsed until all nodes receive a fresh update, and the Age of Collection (AoC) as the amount of time that elapses until the base station receives an update from all nodes.
We quantify the average broadcast and collection ages in two scenarios: 1) instance-dependent, in which the locations of all nodes and interferers are known, and 2) instance-independent, in which they are not known but are located randomly, and expected age is characterized with respect to node locations. In the instance-independent case, we show that AoB and AoC scale super-exponentially with respect to the radius of the region surrounding the base station. Simulation results highlight how expected AoB and AoC are affected by network parameters such as network density, medium access probability, and the size of the coverage region.

Speaker Chirag Rao

Chirag is a PhD student at MIT's Laboratory for Information and Decision Systems.

Energy-aware Age Optimization: AoI Analysis in Multi-source Update Network Systems Powered by Energy Harvesting

Sujunjie Sun, Weiwei Wu, Chenchen Fu, Xiaoxing Qiu and Luo Junzhou (Southeast University, China)

0

This work studies the Age-of-Information (AoI) minimization problem in the information gathering network systems, where time-sensitive data updates are collected from multiple information sources, which are equipped with a battery and harvest energy from ambient energy sources. In such systems, the transmission is available only when there is energy remained in the battery, which is jointly effected by the energy arrival pattern and transmission scheduling policy. This work studies the fundamental impact of the energy arrival on the AoI-optimization transmission scheduling by developing the closed-form expression of average AoI and analyzing its theoretical properties. For the unit battery case, the closed-form expression of the average AoI is derived and the optimal policy is proposed by analyzing the KKT conditions. For the arbitrary finite battery size, the closed-form expression of AoI under SRS policy space with infinite battery capacity is firstly analyzed. Then based on the analysis of the property in the AoI expression, a policy named Max Energy-Aware Weight (MEAW) is proposed by applying Lyapunov optimization, which achieves $2$-approximation in the full policy space. Experimental results validate the theoretical results and show that MEAW performs close to the theoretical lower bound and outperforms the state-of-the-art schemes.

Speaker Sujunjie Sun (Southeast University)

Sujunjie Sun is currently a Ph.D. student at the Department of Computer Science, Southeast University, Nanjing, China, in 2021. His research interest includes Wireless Networks, Optimization Theroy, Scheduling Algorithm, and Age of Information.

Session Chair

Clement Kam

Secure Device Trust Bootstrapping Against Collaborative Signal Modification Attacks

Xiaochan Xue, Shucheng Yu and Min Song (Stevens Institute of Technology, USA)

1

Bootstrapping security among wireless devices without prior-shared secrets is frequently demanded in emerging wireless and mobile applications. One promising approach for this problem is to utilize in-band physical-layer radio-frequency (RF) signals for authenticated key establishment because of the efficiency and high usability. However, existing in-band authenticated key agreement (AKA) protocols are mostly vulnerable to Man-in-the-Middle (MitM) attacks, which can be launched by modifying the transmitted wireless signals over the air. By annihilating legitimate signals and injecting malicious signals, signal modification attackers are able to completely control the communication channels and spoof victim wireless devices. State-of-the-art (SOTA) techniques addressing such attacks require additional auxiliary hardware or are limited to single attackers. This paper proposes a novel in-band security bootstrapping technique that can thwart colluding signal modification attackers. Different from SOTA solutions, our design is compatible with commodity devices without requiring additional hardware. We achieve this based on internal randomness of each device that is unpredictable to attackers. Any modification to RF signals will be detected with high probabilities. Extensive security analysis and experimentation on USRP platform demonstrate effectiveness of our design under various attack strategies.

Speaker Xiaochan Xue (Stevens Institute of Technology)

Xiaochan Xue is a third-year Ph.D. student in the Department of Electrical and Computer Engineering at Stevens Institute of Technology. She holds a master’s degree and B.S. degree from Stevens Institute of Technology in 2020 and Jilin Univerity in China in 2017, respectively. Her research focuses on physical layer wireless security, data privacy-preserving for federated learning, and millimeter-wave networks.

Voice Liveness Detection with Sound Field Dynamics

Qiang Yang (The Hong Kong Polytechnic University, Hong Kong); Kaiyan Cui (The Hong Kong Polytechnic University, China); Yuanqing Zheng (The Hong Kong Polytechnic University, Hong Kong)

0

Voice assistants are widely integrated into a variety of smart devices, enabling users to easily complete daily tasks and even critical operations like online transactions with voice commands. Thus, once attackers replay an unauthorized voice command by loudspeakers to compromise users' voice assistants, this operation will cause serious consequences, such as information leakage and property loss. Unfortunately, most existing voice liveness detection approaches mainly rely on detecting lip motions or subtle physiological features in speech, which are limited within a very short range. In this paper, we propose VoShield to check whether a voice command is from a real user or a loudspeaker imposter. VoShield measures sound field dynamics, a feature that changes fast as the human mouths dynamically open and close. In contrast, it would remain rather stable for loudspeakers due to the fixed size.
This feature enables VoShield to extend the working distance and remain resilient to user locations. To evaluate VoShield, we conducted comprehensive experiments with various settings in different working scenarios. The results show that VoShield can achieve a detection accuracy of 98.2% and an Equal Error Rate of 2.0%, which serves as a promising complement to current voice authentication systems for smart devices.

Speaker Qiang Yang (The Hong Kong Polytechnic University)

Qiang Yang is a fresh Ph.D. graduate from The Hong Kong Polytechnic University. He has broad research interests in ubiquitous computing and acoustic sensing. His research has been published in well-recognized conferences and journals like UbiComp, INFOCOM, ICDCS, IPSN, and TMC. He is open to postdoc positions.

Expelliarmus: Command Cancellation Attacks on Smartphones using Electromagnetic Interference

Ming Gao (Zhejiang University, China); Fu Xiao (Nanjing University of Posts and Telecommunications, China); Weiran Liu, Wentao Guo, Yangtao Huang and Yajie Liu (Zhejiang University, China); Jinsong Han (Zhejiang University & School of Cyber Science and Technology, China)

0

Human-machine interactions (HMIs), e.g., touchscreens, are essential for users to interact with mobile devices. They are also beneficial in resisting emerging active attacks, which aim at maliciously controlling mobile devices, e.g., smartphones and tablets. With touchscreen-like HMIs, users can notice and interrupt malicious actions conducted by the attackers timely and perform necessary countermeasures, e.g., tapping the ‘Quit' button on the touchscreen. However, the effect of HMI-oriented active attacks has not been investigated yet. In this paper, we present a practical attack towards touchscreen-based devices, namely Expelliarmus. It reveals a new attack surface of active attacks for hijacking users' operations and thus taking full control over victim devices. Expelliarmus neutralizes users' touch commands by producing a reverse current via electromagnetic interference (EMI). Since the reverse current offsets the current change caused by a touch, the touchscreen detects no current change and thus ignores users' commands. Besides this basic denial-of-service attack, we also realize a target cancellation attack, which can neutralize target commands, e.g., ‘Quit' without interference in irrelevant operations. Thus, the active attack can be completely performed without interruption from users, even if they are alerted by the abnormal events. Extensive evaluations demonstrate the effectiveness of Expelliarmus on 29 off-the-shelf devices.

Speaker Ming Gao (Zhejiang University)

Ming Gao is a Ph.D. student at the School of Cyber Science and Technology, Zhejiang University, under the supervision of Prof. Jinsong Han. His research interests lie in mobile computing, wireless sensing, and cyber-physical security.

Nowhere to Hide: Detecting Live Video Forgery via Vision-WiFi Silhouette Correspondence

Xinyue Fang, Jianwei Liu and Yike Chen (Zhejiang University, China); Jinsong Han (Zhejiang University & School of Cyber Science and Technology, China); Kui Ren and Gang Chen (Zhejiang University, China)

0

For safety guard and crime prevention, video surveillance systems have been pervasively deployed in many security-critical scenarios, such as the residence, retail stores, and banks. However, these systems could be infiltrated by the adversary and the video streams would be modified or replaced, i.e., under the video forgery attack. The prevalence of Internet of Things (IoT) devices and the emergence of Deepfake-like techniques severely emphasize the vulnerability of video surveillance systems under such attacks. To secure existing surveillance systems, in this paper we propose a vision-WiFi cross-modal video forgery detection system, namely WiSil. Leveraging a theoretical model based on the principle of signal propagation, WiSil constructs wave front information of the object in the monitoring area from WiFi signals. With a well-designed deep learning network, WiSil further recovers silhouettes from the wave front information. Based on a Siamese network-based semantic feature extractor, WiSil can eventually determine whether a frame is manipulated by comparing the semantic feature vectors extracted from the video's silhouette with those extracted from the WiFi's silhouette. Extensive experiments show that WiSil can achieve 95% accuracy in detecting tampered frames. Moreover, WiSil is robust against environment and person changes.

Speaker Xinyue Fang (Zhejiang University)

Sep 2020 – Now Zhejiang University, Doctoral Student, Computer Science and Technology, Supervisor: Prof. Jinsong Han

Session Chair

Yanjun Pan

Program at a Glance