Deep Learning for Channel Tracking in IRS-Assisted UAV Communication Systems

To boost the performance of wireless communication networks, unmanned aerial vehicles (UAVs) aided communications have drawn dramatically attention due to their flexibility in establishing the line of sight (LoS) communications. However, with the blockage in the complex urban environment, and due to the movement of UAVs and mobile users, the directional paths can be occasionally blocked by trees and high-rise buildings. Intelligent reflection surfaces (IRSs) that can reflect signals to generate virtual LoS paths are capable of providing stable communications and serving wider coverage. This is the first paper that exploits a three-dimensional geometry dynamic channel model in IRS- assisted UAV-enabled communication system. Moreover, we develop a novel deep learning based channel tracking algorithm consisting of two modules: channel pre-estimation and channel tracking. A deep neural network with off-line training is designed for denoising in the pre-estimation module. Moreover, for channel tracking, a stacked bi-directional long short term memory (Stacked Bi-LSTM) is developed based on a framework that can trace back historical time sequence together with bidirectional structure over multiple stacked layers. Simulations have shown that the proposed channel tracking algorithm requires fewer epochs to convergence compared to benchmark algorithms. It also demonstrates that the proposed algorithm is superior to different benchmarks with small pilot overheads and comparable computation complexity.

I. Introduction

Unmanned aerial vehicles (UAVs) acting as aerial base stations can provide a solution on serving wider coverage, supporting reliable connections, and providing energy efficient communications [1], [2]. The flexibility of UAVs has led to plenty of applications such as security surveillance, real-time monitoring, rescue, and emergency communications [3]. Although UAV-aided systems are considered as promising techniques for future wireless communications for smart city, the complex urban environment poses potential blockage problem on LoS links between navigation UAVs and ground users [4]. IRSs that can construct virtual LoS paths to enhance the quality and coverage of wireless propagation has become an invaluable solution on overcoming signal pathloss and on securing communications [5], [6]. This is because that the low-cost IRS can intelligently adjust its phase shifts to steer signal power towards targeted directions and reduce information leakage. Thus, to address the blockage problem in UAV-aided system, intelligent reflection surfaces (IRSs) can be installed to assist the UAV to offer ubiquitous communication services [4], [7]–[9]. Additionally, deploying IRSs in the UAV-aided system can further help with the time- and energy- consuming problem caused by UAV navigation when some users are far away [7]. Therefore, with the appealing advantages of UAV and IRS, deploying both of them in the wireless system can dramatically boost the communication performance.