Balancing Fairness and Energy Efficiency in SWIPT-Based D2D Networks: Deep Reinforcement Learning Based Approach

Han, Eunyoung; Muy, Sengly; Lee, Jung-Ryun

doi:10.1109/access.2022.3182686

Cited by 11 publications

(5 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a method based on learning with DNN, the work in [24] proposed a long shortterm memory (LSTM) recurrent neural network (RNN)-based mode-switching algorithm to maximize the achievable rate under the energy-causality constraint for its dual mode SWIPT system. In [25], the authors determine the subchannel allocation, power splitting ratio (PR), and transmit power for the SWIPT-based device-to-device (D2D) networks through the deep-reinforcement-learning (DRL)-based algorithm developed therein. For similar D2D SWIPT-based networks, an EE optimization problem was formulated in [26], and the authors adopted exhaustive search (ES) and gradient search (GS), respectively, to obtain the global optimum and local optimum for the formulated nonconvex optimization problem.…”

Section: Related Workmentioning

confidence: 99%

“…In the above, we assume that N t antennas of each BS are arranged as a uniform linear array (ULA). In addition, similar to [9,11,25,26], we consider that each UE i with PS on the received signal y i (t) can simultaneously perform ID and EH as shown in Figure 1. Specifically, with θ i (t) ∈ (0, 1) to denote the PR adopted by UE i at time t, the instantaneous signal to interference and noise ratio (SINR) for ID can be formulated as [32]:…”

Section: Network and Channel Modelsmentioning

confidence: 99%

“…Given this, the conventional works, such as [9,11,25,33,34], usually convert the received signal y EH i (t) into the DC power with a linear function. However, a nonlinear function for the energy conversion would be more practical, and the previous works, such as [30,31], adopted the logarithmic nonliner EH model for the ith IoT device on the jth sub-carrier as follows:…”

Section: Network and Channel Modelsmentioning

confidence: 99%

See 2 more Smart Citations

Joint Data Transmission and Energy Harvesting for MISO Downlink Transmission Coordination in Wireless IoT Networks

Liu

Lin

et al. 2023

Sensors

View full text Add to dashboard Cite

The advent of simultaneous wireless information and power (SWIPT) has been regarded as a promising technique to provide power supplies for an energy sustainable Internet of Things (IoT), which is of paramount importance due to the proliferation of high data communication demands of low-power network devices. In such networks, a multi-antenna base station (BS) in each cell can be utilized to concurrently transmit messages and energies to its intended IoT user equipment (IoT-UE) with a single antenna under a common broadcast frequency band, resulting in a multi-cell multi-input single-output (MISO) interference channel (IC). In this work, we aim to find the trade-off between the spectrum efficiency (SE) and energy harvesting (EH) in SWIPT-enabled networks with MISO ICs. For this, we derive a multi-objective optimization (MOO) formulation to obtain the optimal beamforming pattern (BP) and power splitting ratio (PR), and we propose a fractional programming (FP) model to find the solution. To tackle the nonconvexity of FP, an evolutionary algorithm (EA)-aided quadratic transform technique is proposed, which recasts the nonconvex problem as a sequence of convex problems to be solved iteratively. To further reduce the communication overhead and computational complexity, a distributed multi-agent learning-based approach is proposed that requires only partial observations of the channel state information (CSI). In this approach, each BS is equipped with a double deep Q network (DDQN) to determine the BP and PR for its UE with lower computational complexity based on the observations through a limited information exchange process. Finally, with the simulation experiments, we verify the trade-off between SE and EH, and we demonstrate that, apart from the FP algorithm introduced to provide superior solutions, the proposed DDQN algorithm also shows its performance gain in terms of utility to be up to 1.23-, 1.87-, and 3.45-times larger than the Advantage Actor Critic (A2C), greedy, and random algorithms, respectively, in comparison in the simulated environment.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Network and Channel Modelsmentioning

confidence: 99%

Section: Network and Channel Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Joint Data Transmission and Energy Harvesting for MISO Downlink Transmission Coordination in Wireless IoT Networks

Liu

Lin

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Therefore, in the process of education, it is necessary to cultivate students' abilities in independent thinking, self-learning, acquiring new information and dealing with relevant problems. At the same time, teachers are also required to guide and help them understand the basic content of their major and make their own reasonable explanations after in-depth understanding, so as to promote personalized development to adapt to social needs and improve the awareness of independent learning [17][18].…”

Section: The Role Of Deep Learning In Intelligent Educationmentioning

confidence: 99%

Appliance of Deep Learning in Intelligent Education

2022

View full text Add to dashboard Cite

Intelligent education is the main direction of China's future development. In the information age, the traditional teaching mode can no longer meet the higher requirements for information and knowledge quality in students' learning. Based on this, this paper proposes a method to help improve neural network training speed, reduce training time and improve memory efficiency. First, it introduces the relevant theories of intelligent pedagogy through literature analysis, and then designs a set of experimental scheme according to the actual situation, and verifies its effectiveness and accuracy. Finally, the feasibility of the design scheme is verified by the experimental results. The test results show that the data processing time of the model is relatively fast, the delay time is relatively low, and the compatibility rate of the model is up to more than 90%. This provides a reference for further exploration of AI education in the future, and also contributes to the future development of China.

show abstract

“…The authors in [19] applied a learning-based method to maximize both SE and EE for wireless-powered D2D networks. In [20] and [21], reinforcement-learning methods addressed energy-efficient resource allocation by making agents interact with unknown wireless environments.…”

Section: Introductionmentioning

confidence: 99%

Energy Price-Based Resource Scheduling for Network Lifetime Extending in Cooperative D2D Communications Overlaying Cellular Networks

Yu,

Tang

2023

IEEE Access

View full text Add to dashboard Cite

This paper considers device-to-device (D2D) communications overlaying cellular networks. Combined with cooperative relay technology, a D2D user can serve as a relay for a cellular link and gain the opportunity to access the corresponding cellular spectrum. Considering mobile terminals' limited battery capacity and motivated by green communication, we propose a resource scheduling scheme, which comprises joint power and spectrum allocation, mode selection, and link matching. Unlike previous works, to balance energy utilization and extend the network lifetime, this scheme sets a price for each node and minimizes the system cost during each frame instead of energy efficiency (EE). Specifically, we investigate the power and spectrum allocation problem to characterize the minimum cost of matched links. We prove it to be nonconvex and obtain the optimal solution by the graphic method and Newton's method with a small number of iterations. We also discuss the energy pricing strategy. In order to reduce communication and computation overhead, we propose a heuristic pricing strategy that involves no iteration and supposes nodes' energy prices are inversely proportional to their residual energy. Through simulations under various cases, we verify that the proposed scheme significantly improves network lifetime and suits real-time operations while guaranteeing the quality of service (QoS) requirement. Moreover, the simulation results demonstrate that the proposed scheme performs the same as the maximum EE scheme if the energy prices are equal at each node.

show abstract

Balancing Fairness and Energy Efficiency in SWIPT-Based D2D Networks: Deep Reinforcement Learning Based Approach

Cited by 11 publications

References 26 publications

Joint Data Transmission and Energy Harvesting for MISO Downlink Transmission Coordination in Wireless IoT Networks

Joint Data Transmission and Energy Harvesting for MISO Downlink Transmission Coordination in Wireless IoT Networks

Appliance of Deep Learning in Intelligent Education

Energy Price-Based Resource Scheduling for Network Lifetime Extending in Cooperative D2D Communications Overlaying Cellular Networks

Contact Info

Product

Resources

About