In vehicular communication systems, maximizing the number of served vehicles while simultaneously guaranteeing reliable coverage at all the vehicles can be a challenging proposition. A switched-beam at the infrastructure can provide better reliability as the signal-to-interference-plus-noise ratio (SINR) can be improved. However, a simple switched-beam based vehicle-to-infrastructure (V2I) system alone may not suffice for serving all the vehicles, because, (i) the number of vehicles is more than the number of beams, and (ii) vehicles may be out of the coverage region of a beam. Therefore, introducing vehicle-to-vehicle (V2V) communication becomes crucial in extending the number of served vehicles. In this paper, a joint vehicle beam allocation (VBA) and vehicular proximity (VP) algorithms for V2I and V2V, respectively are proposed for reliable coverage for vehicles. VBA is an SINR optimization algorithm, and VP is based on LTE Mode 4, a proximity based service for V2V communications. It is shown that setting a flexible SINR threshold helps in attaining a reliable beam coverage region in switched-beam V2I. It is proven that the outage probability and rates are also directly dependent on SINR thresholds. Lastly, the concept of utility ratio is also introduced as a metric for reliability. It is proven that joint V2I and V2V communication significantly improves the utility ratio.

It is crucial for emergency communication networks to properly manage and coordinate between rescuers and disaster-affected users. Low-cost and high-mobility drones become important contributors to build up such kind of emergency communication networks. In this paper, we consider the drone as a deployed base station to provide communication for ground users in the post-disaster area. Considering the importance of providing guaranteed service for rescuers (compared to normal users), we have investigated power and subcarrier allocation to maximize the downlink system capacity in the drone-assisted emergency communication system. Due to the complexity of the formulated problem. a suboptimal solution is proposed by dividing users into two priority groups: high-priority users (rescuers) and low-priority users (affected people). According to analyses, the complexity of the proposed scheme is much lower than the optimal scheme. Simulation results have shown that when the transmit power of the drone is not very high, the performance of the proposed resource allocation scheme is very close to the optimal scheme.

The commercialization of 5G is increasing in recent days, will provide users with satisfied speed and more reliable QoS. UAV is one possible carrier of wireless base stations and provide flexible coverage and wireless services to target users. In this work, we discuss an intelligent UAV based 5G emergency network model, with dedicated base stations, core networks and smart MIMO antennas located on UAV. As a solution of emergency environment, this model could provide timely access to users with higher speed, more flexible access mode, smart coverage area, customized service and longer time of duration. Field trial and system level simulation have been provided prior to this work and the results seem promising in some main indicators, the coverage area could exceed the radius of 30km, the downlink speed per user could reach 200Mbps when the height is within 4km. The remaining problems of the resource allocation, power control, deployment, etc. are still open for researchers to tackle.

In search and rescue (SAR), survivals' localization information is required to be obtained as quickly and accurately as possible. However, due to the accidental and destructive natural of emergency events, existing network infrastructures and devices may be unable to be use for performing the localization tasks. Furthermore, survivals may be trapped in isolated areas, where human being is unable to reach. To address these issues, in this paper, we propose an integrated platoon and unmanned aerial vehicle (UAV) system for implementing the three-dimensional (3D) localization task in SAR, where the platoon is consisted of self-driving vehicles and UAVs are controlled by these vehicles. Furthermore, in order to facilitate 3D localization, a platoon-aided UAV deployment strategy is proposed, where UAVs are deployed in different places for collecting direction-of-arrival (DoA) information of a target. As a further enhancement, a rotational invariance based beamforming (RIB) scheme is conceived to provide a simple DoA estimation for our multiple-in-multiple-out (MIMO) UAVs. Simulation results are provided for evaluating the system performance, showing that it is capable of achieving a beneficial trade-off between the attainable 3D localization performance and the corresponding time delay.

Intelligent emergency rescue equipment (IERE), e.g., AR/VR aided intelligent helmet, robot, etc., plays an increasingly important role in emergency rescue, which can greatly enhance the capability and efficiency of the rescue activities. However, due to the limitations of its computing and energy capacities, deal with the generated computational-intensive tasks locally is difficult, even more, in disaster areas, the destruction of the terrestrial communication infrastructure also makes the cloud-based computation offloading disabled. Therefore, in this paper, we construct a multi-UAV enabled latency and reliability guaranteed mobile edge computing framework (LR-UMEN) to provide IERE the on-demand computing services. Furthermore, considering the harsh working environment of the IERE, not only the latency but also the reliability must be taken into account. Different from the existing work about the latency and reliability, which are solely paying attention to the optimization of communication resources, or both communication and computing resources, we propose the joint optimization of communication, computing and caching to further decrease the latency as well as enhance the reliability. Besides, to better optimize the latency and reliability, we derive a new quantity, namely Expected Latency, which can effectively reflect the characteristic both of the latency and reliability of the operation unit during a given processing period. And to solve the non-convex and NP-hard optimization problem, a hybrid binary particle swarm optimization algorithm is designed, moreover, extensive simulations validate its effectiveness.

When natural disasters strike, the communication infrastructure may be severely damaged, which leads to the congestion or even destruction of communication networks. Cache can be equipped on the mobile devices of rescuers to provide imperative data for victims via device-to-device (D2D) transmissions. To support the effectiveness and timeliness of emergency communication, network throughput and delay-requirements of victims need to be comprehensively considered, which posts a new challenge for the design of caching placement strategy. In this paper, we formulate the statistical delay-bounded quality of service (QoS) provisioning framework for cache-enabled D2D emergency network. Particularly, we build up the optimization problem to maximize the average effective capacity of D2D transmissions subject to the delay-QoS requirement. Then, we prove that this optimization problem is convex. For solving this problem, we develop the QoS-aware caching placement strategy, which can maximize the average effective capacity of D2D transmission while guaranteeing the delay-QoS requirement. Simulation results verify our theoretical analyses and show that the proposed QoS-aware caching placement scheme can achieve larger average effective capacity as compared with the caching placement strategies without QoS provisioning.

In the emergency scenario, wireless communications will face spectrum shortage and energy cut-off in case of infrastructure destruction. In this paper, a green cognitive Internet of Things (CIoT) with wireless energy harvesting (WEH) is proposed to share the spectrum licensed to primary user (PU) and harvest its radio frequency (RF) energy, in order to achieve self supply of energy and spectrum. Energy-efficient resource allocations are presented to maximize the average transmission rate of CIoT while guaranteeing the energy saving requirements. The underlay and overlay spectrum access models for CIoT with WEH are described, in which the energy-efficient resource allocations are formulated as joint optimization problems that can be solved using alternating direction optimization (ADO) and Karush-Kuhn-Tucker (KKT) conditions. Results show that the CIoT with WEH can consume less power to achieve larger transmission rate.

Radio access technology (RAT) is the decision for user to choose a best target network during network entry or vertical handover. It could be used for the rescuing scenario in the wireless multi-hop UVA networks. Most researches about RAT selection focus on heterogeneous cell networks. The RAT selection method for cell networks is no longer suitable for heterogeneous wireless multi-hop network (HWMN), because the communicating quality to the expected access point (EAP) couldn't reflect the whole network capability due to its multi-hop characteristic. This paper presents a novel metric, named n-hop prediction (NHP), for RAT selection in HWMN. It takes into account of network capability and load within n-hop range of the expected access point, besides users QoS requirement. Simulation results reveal that the proposed approach achieves better performance than that without considering these factors.

This paper studies the noise aggregation scheme with convolutional code in frequency-flat block-fading channel. While there is still a lot of limitation in the noise aggregation scheme, we combine the noise aggregation scheme with convolutional code for further study. By analysing the structure of its encoding system, we derives the optimal decoding algorithm in this scheme, and design three different decoding schemes. Considering the complexity and correctness of the decoding progress, we will adopt different decoding scheme in different situation. Due to the introduction of fading channel, the effect of the noise aggregation scheme has been increased significantly. In the meanwhile, we combine the noise aggregation scheme with adaptive modulation technology, the advantage of noise aggregation become more obvious. Simulation results show that the noise aggregation scheme can achieve a remarkable safety performance improvement and have good application prospects.