Bandit-Based Power Control in Full-Duplex Cooperative Relay Networks With Strict-Sense Stationary and Non-Stationary Wireless Communication Channels

Abstract

Full-duplex relaying is an enabling technique of sixth generation (6G) mobile networks, promising tremendous rate and spectral efficiency gains. In order to improve the performance of full-duplex communications, power control is a viable way of avoiding excessive loop interference at the relay. Unfortunately, power control requires channel state information of source-relay, relaydestination and loop interference channels, thus resulting in increased overheads. Aiming to offer a low-complexity alternative for power control in such networks, we adopt reward-based learning in the sense of multi-armed bandits. More specifically, we present bandit-based power control, relying on acknowledgements/negative-acknowledgements observations by the relay. Our distributed algorithms avoid channel state information acquisition and exchange, and can alleviate the impact of outdated channel state information. Two cases are examined regarding the channel statistics of the wireless network, namely, strict-sense stationary and non-stationary channels. For the latter, a sliding window approach is adopted to further improve the performance. Performance evaluation highlights a performance-complexity trade-off, compared to optimal power control with full channel knowledge and significant gains over cases considering channel estimation and feedback overheads, outdated channel knowledge, no power control and random power level selection. Finally, it is shown that the sliding-window bandit-based algorithm provides improved performance in non-stationary settings by efficiently adapting to abrupt changes of the wireless channels.

Bandit-Based Learning-Aided Full-Duplex/Half-Duplex Mode Selection in 6G Cooperative Relay Networks

“…However, these overheads may severely affect network performance and as a remedy, ML integration in wireless networks has introduced promising gains towards low-complexity network operation (see, for example, [12], [29]). Furthermore, reward-based learning, as it is the case of MAB, has already been adopted in various 5G/6G cases; see, e.g., [3], [30], [31], where power control in a single FD relay network was studied, and modelled as a MAB game.…”

“…• Contrary to [30], [31], here, we aim to tackle a more complex and practical problem, by equipping the relay with a buffer to overcome cases when FD operation is infeasible due to excessive SI, thus efficiently switching to HD mode. Under this mode, the relay receives and stores packets in its buffer or extracts packets from its buffer and forwards them towards the destination.…”

“…Here, performance evaluation, in terms of average throughput for MABAMS and other CSI-and learning-based policies are presented. In greater detail, MABAMS is compared to FD relaying with power control, using outdated CSI (out-CSI PC) with ρ = 0.5, FD relaying without power control (no-PC), FD relaying with random power level selection (rnd) and learning-aided BB-PC where only the FD mode is employed [30], [31]. As a performance upper-bound for the FD relay operation, a CSI-based FD relaying policy is included (opt-PC) where power control at the relay sets the transmit power according to eq.…”

A Survey on UAV-Aided Maritime Communications: Deployment Considerations, Applications, and Future Challenges

“…2) Energy efficiency/network lifetime maximization As maritime activities are constantly evolving, an increased number of battery-dependent network nodes, such as UUVs, USVs, UAVs, is deployed. As a result, it is vital to adopt energy-efficient communication techniques with power control [56] and appropriate routing algorithms [57]. The energy efficiency is usually measured in bits/joule, i.e., the number of bits that are transmitted over the energy used for their transmission.…”

“…Still, wireless communication issues might arise related to delay in data exchange, channel quality degradation, energy constraints, and harsh meteorological conditions affecting UAV stability. In this context, ML-based algorithms, e.g., through bandit-based channel prediction can alleviate issues related outdated channel state acquisition [56], while WPT solutions and joint trajectory optimization for swarms of UAV could be considered as promising approaches for optimizing the energy management of these systems. These types of problems become much more complicated to be solved in use cases, where a satellite segment is required to be present, in a research field that a lot of work has to be done.…”