No abstract

With the rapid development of vehicular networks, vehicle-to-everything (V2X) communications have huge number of tasks to be calculated, which brings challenges to the scarce network resources. Cloud servers can alleviate the terrible situation regarding the lack of computing abilities of vehicular user equipment (VUE), but the limited resources, the dynamic environment of vehicles, and the long distances between the cloud servers and VUE induce some potential issues, such as extra communication delay and energy consumption. Fortunately, mobile edge computing (MEC), a promising computing paradigm, can ameliorate the above problems by enhancing the computing abilities of VUE through allocating the computational resources to VUE. In this paper, we propose a joint optimization algorithm based on a deep reinforcement learning algorithm named the double deep Q network (double DQN) to minimize the cost constituted of energy consumption, the latency of computation, and communication with the proper policy. The proposed algorithm is more suitable for dynamic scenarios and requires low-latency vehicular scenarios in the real world. Compared with other reinforcement learning algorithms, the algorithm we proposed algorithm improve the performance in terms of convergence, defined cost, and speed by around 30%, 15%, and 17%.

No abstract

This paper considers a scenario in which an unmanned aerial vehicle (UAV) collects the sensed data from the terrestrial Internet of things (IoT) devices in an untraveled and base station (BS)-uncovered area.These IoT devices, in this scenario, are of various types and deployed in a spare manner.Consequently, how to design an effective and suitable resource allocation scheme is a key part of dispatching a UAV to collect the sensed data from IoT devices.The goal of this paper is to minimize the UAV's cruise time with the joint optimization of IoT devices' communication scheduling, UAV trajectory, and bandwidth allocation.To facilitate data collection by UAVs, a logarithmic kernel-based mean shift (LKMS) clustering algorithm is proposed to group IoT devices into multiple clusters.Based on the clustering result, a mixed-integer joint non-convex problem is formulated.To avoid the difficulties caused by solving the aforementioned problem directly, the block coordinate descent (BCD)-based method, as an alternative, is adopted to decouple the variables and decompose the non-convex problem.To tackle the non-convex subproblems, a successive convex approximation (SCA)-based algorithm is also proposed.Numerical results demonstrate that the proposed scheme is able to achieve significant performance over other schemes for scenarios of UAV-assisted wireless IoT networks to collect a huge amount of data.

This paper investigates the unmanned aerial vehicle (UAV)‐assisted wireless communication network that collects the data information of Internet of things (IoT) devices deployed in the region, where the cellular networks cannot cover. Due to the numerous variety and number of IoT devices, a large amount of data generated by IoT networks needs to be collected by UAV. The goal of this paper is to minimize the UAV's cruise time with the joint optimization of IoT devices communication scheduling, UAV trajectory, and transmit bandwidth allocation. To facilitate data collection by UAVs, the data‐distance‐k‐means (d2‐k‐means) algorithm is proposed to divide IoT devices into multiple initial clusters. However, the formulated problem is mixed‐integer joint non‐convex, so it is difficult to solve directly. Since it may be with relatively high computational complexity, as an alternative, a block coordinate descent (BCD)‐based method is designed. To tackle the non‐convex problem, a successive convex approximation (SCA)‐based algorithm is also proposed. Numerical results demonstrate that the proposed scheme is able to achieve significant performance over other schemes for scenarios of UAV‐assisted wireless IoT networks to collect massive amount of data.