Sum Rate Maximization for IRS-Assisted Energy Harvesting Cognitive Radio Networks

Cognitive Radio (CR) technology allows secondary users (SUs) to utilize the spectrum resources of primary users (PUs) without affecting PUs' quality of service (QoS), addressing spectrum scarcity in future wireless communication systems. Energy Harvesting (EH)-powered Cognitive Radio Networks (...

Full description

Saved in:
Bibliographic Details
Published in2025 8th World Conference on Computing and Communication Technologies (WCCCT) pp. 357 - 362
Main Authors Kawala, Lilian Chiru, Li, Guoquan, Ting, Liu, Kakati, Amayika
Format Conference Proceeding
LanguageEnglish
Published IEEE 11.04.2025
Subjects
Approximation algorithms
Benchmark testing
Cognitive radio
Cognitive Radio Networks
Convergence
Energy harvesting
Intelligent Reflecting Surfaces
Optimization
Quality of service
Taylor series
The Detection Probability
Throughput
Throughput Maximization
Uncertainty
Online AccessGet full text

Cover

More Information
Summary:Cognitive Radio (CR) technology allows secondary users (SUs) to utilize the spectrum resources of primary users (PUs) without affecting PUs' quality of service (QoS), addressing spectrum scarcity in future wireless communication systems. Energy Harvesting (EH)-powered Cognitive Radio Networks (EH-CRNs) equip secondary base stations with EH capabilities to address spectrum scarcity and RF energy harvesting simultaneously but face challenges in optimizing spectrum and EH efficiency. In this paper, we establish an EH-CRN system model assisted by Intelligent Reflecting Surface (IRS) for better transmission QoS and formulate an optimization problem to maximize secondary networks' achievable throughput subject to constraints on the lowest false alarm probability, SU QoS, and IRS phase shift. To solve the non-convex problem, we propose a resource allocation algorithm based on alternating optimization and divide it into three subproblems to respectively optimize the detection probability, the energy harvesting, and the achievable rate of secondary users by using semidefinite relaxation (SDR), successive convex approximation (SCA), first-order Taylor expansion, and Gaussian randomization with sequential rank-one constraint relaxation (SROCR). Simulation results demonstrate the algorithm's convergence and improved performance compared to benchmark schemes under consistent constraints.
DOI:10.1109/WCCCT65447.2025.11027971