The conventional federated learning (FL) framework usually assumes synchronous reception and fusion of all the local models at the central aggregator and synchronous updating and training of the global model at all the agents as well. However, in a wireless network, due to limited radio resource, inevitable transmission failures and heterogeneous computing capacity, it is very hard to realize strict synchronization among all the involved user equipments (UEs). In this paper, we propose a novel asynchronous FL framework, which well adapts to the heterogeneity of users, communication environments and learning tasks, by considering both the possible delays in training and uploading the local models and the resultant staleness among the received models that has heavy impact on the global model fusion. A novel centralized fusion algorithm is designed to determine the fusion weight during the global update, which aims to make full use of the fresh information contained in the uploaded local models while avoiding the biased convergence by enforcing the impact of each UE’s local dataset to be proportional to its sample share. Numerical experiments validate that the proposed asynchronous FL framework can achieve fast and smooth convergence and enhance the training efficiency significantly.

Real-time path planning can efficiently relieve traffic congestion in urban scenarios. However, how to design an efficient path-planning algorithm to achieve a globally optimal vehicle-traffic control still remains a challenging problem, particularly when we take drivers' individual preferences into consideration. In this paper, we first establish a hybrid intelligent transportation system (ITS), i.e., a hybrid-VANET-enhanced ITS, which utilizes both vehicular ad hoc networks (VANETs) and cellular systems of the public transportation system to enable real-time communications among vehicles, roadside units (RSUs), and a vehicle-traffic server in an efficient way. Then, we propose a real-time path-planning algorithm, which not only improves the overall spatial utilization of a road network but reduces average vehicle travel cost for avoiding vehicles from getting stuck in congestion as well. A stochastic Lyapunov optimization technique is exploited to address the globally optimal path-planning problem. Finally, the transmission delay of the hybrid-VANET-enhanced ITS is evaluated in VISSIM to show the timeliness of the proposed communication framework. Moreover, system-level simulations conducted in Java demonstrate that the proposed path-planning algorithm outperforms the traditional distributed path planning in terms of balancing the spatial utilization and drivers' travel cost.

Content caching at network edge nodes, such as base stations (BSs) and user equipments (UEs), can significantly reduce the traffic load in future cellular networks. Considering the limited caching space, the contents cached at BSs should be selected carefully for improving caching efficiency. In this paper, we study the edge caching at BSs to minimize the transmission cost by considering traffic offloading via the device-to-device (D2D) communications. The traffic offloading reduces the traffic via cellular transmission and thus changes the utility achieved by content caching at BSs. We model the edge caching problem as a Markov decision process and propose a distributed cache replacement strategy based on Q-learning. The proposed strategy further needs the calculations of the following two key parameters: 1) To describe the effect of D2D offloading to the cellular traffic, we define the cellular serving ratio, which is calculated by the iterative maximum weighted independent sets problem for static networks and the stochastic geometry for high-dynamic networks; 2) The cache replacement rewards are calculated by analyzing the relationship between the requested and cached amounts of content data, which are obtained from the messages of the previous data request and transmissions, ignoring any extra information exchange between the BSs. Furthermore, the convergence of the proposed distributed cache replacement strategy is proved by the sequential stage game model. Simulation results verify the convergence of the proposed cache replacement strategy and show its performance gain compared with conventional strategies.

Channel estimation in vehicle-to-everything (V2X) communications is a challenging issue due to the fast time-varying and non-stationary characteristics of wireless channel. To grasp the complicated variations of channel with limited number of pilots in the IEEE 802.11p systems, data pilot-aided (DPA) channel estimation has been widely studied. However, the error propagation in the DPA procedure, caused by the noise and the channel variation within adjacent symbols, limits the performance seriously. In this letter, we propose a deep learning based channel estimation scheme, which exploits a long short-term memory network followed by a multilayer perceptron network to solve the error propagation issue. Simulation results show that the proposed scheme outperforms currently widely-used DPA schemes for the IEEE 802.11p-based V2X communications.

Recently, 360-degree mobile virtual reality video (MVRV) has become increasingly popular because it can provide users with an immersive experience. However, MVRV is usually recorded in a high resolution and is sensitive to latency, which indicates that broadband, ultra-reliable, and low-latency communication is necessary to guarantee the users’ quality of experience. In this paper, we propose a mobile edge computing (MEC)-based 360-degree MVRV streaming scheme with field-of-view (FoV) prediction, which jointly considers video coding, proactive caching, computation offloading, and data transmission. To meet the requirement of stringent end-to-end (E2E) latency, the user’s viewpoint prediction is utilized to cache video data proactively, and computing tasks are partially offloaded to the MEC server. In addition, we propose an analytical model based on diffusion process to study the packet transmission process of 360-degree MVRV in multihop wired/wireless networks and analyze the performance of the MEC-enabled scheme. The simulation results verify the accuracy of the analysis and the effectiveness of the proposed MVRV streaming scheme in reducing the E2E delay. Furthermore, the analytical framework sheds some light on the impacts of system parameters, e.g., FoV prediction accuracy and transmission rate, on the balance between computation delay and communication delay.

Recently, vehicle platooning has attracted a lot of attention as one of the promising solutions to the explosion of vehicle numbers. By exchanging information among vehicles through communication networks, vehicle platooning can significantly improve traffic safety and efficiency. In this paper, we develop a joint systematic design of platoon communication and control to reduce position errors of consecutive vehicles and to improve platoon safety. Through separate information dissemination of the platoon leader and followers, we improve the success probability of the platoon leader's information dissemination. To extend the communication range of the platoon leader and also the platoon scale, we propose to use relays to forward the platoon leader's messages. Further, an adaptive distributed model predictive control (DMPC) based controller is presented, which can adjust its control parameters according to platoon state and dynamic information sources. Simulation results not only verify the proposed systematic design in terms of position errors, but also show that our scheme performs well in vehicle failure cases where collisions can be avoided and platoon safety improved.