PSO-Based Agent Cooperative Spectrum Sensing in Cognitive Radio Networks

Abstract

Secondary users (SUs) can detect the states of primary users (PUs) and access the idle spectrum in an opportunistic way in cognitive radio networks (CRNs). With the spread of wireless communication devices, the mobility of both PUs and SUs is ubiquitous. To obtain more accurate spectrum sensing data, SUs must be located within the transmission range of PUs. With the unknown mobility, it is difficult to guarantee the efficiency of spectrum sensing. Focused on this issue, we propose a new scheme, called Particle Swarm Optimization (PSO) -based agent cooperative spectrum sensing (PSOA) in this paper. In this scheme, we deploy multiple mobile agents spreading over the network, to cooperate in spectrum sensing instead of SUs. All agents will move according to the latest global optimal agents of the corresponding target PUs with the fitness function calculated by modified PSO. With the optimal movement, the distribution of agents can guarantee that most PUs are within the detection coverage of PSOA. The evaluation results show that our scheme can save over 80% of sensing time and over 80% of energy consumption (affected by the agents' number and max velocity) than the active searching scheme. PSOA also guarantees the sensing probability of 80% and higher in our simulation.

An artificial intelligence‐based spectrum sensing methodology for LoRa and cognitive radio networks

“…The PSO approach is summarized before providing the remaining specifics of the suggested solution. PSO is a metaheuristic technique that manipulates various bird species' natural behavior to achieve the objective and/or optimal value 7,19,41 . A bird always flies toward a better position in its surroundings when it is in a flock.…”

“…PSO is a metaheuristic technique that manipulates various bird species' natural behavior to achieve the objective and/or optimal value. 7,19,41 A bird always flies toward a better position in its surroundings when it is in a flock. Each member of the PSO behaves like a bird, obeying rules and interacting with those around it to fulfill the overall job.…”

“…The coordinated error performance metrics BER, PER, and SER must be defined as LoRa-AWGN compatible to calculate this fitness function. The fitness function F gen is defined for the genetic algorithm and can be calculated as in Equation (19). As can be understood from this equation, F gen consists of the sum of BER, PER, and SER values.…”

“…Despite the sector analysis in the cognitive network, adequate performance criteria were not considered in the simulations, and LPWAN technologies were not taken into account. Chen et al 19 proposed a particle swarm optimization (PSO)‐based collaborative spectrum detection method. In this method, multiple mobile agents deployed over the network are used to cooperate in spectrum sensing instead of SUs.…”

Particle Swarm Optimization: A Comprehensive Survey

“…Since spectrum sensing in CRNs is a non-convex optimization problem, CRNs utilize PSO algorithms to optimize their performance in terms of energy efficiency, spectral efficiency and sensing time. The work in [227] has implemented the standard PSO algorithm to detect the presence of primary users. According to the simulation results, PSO can save more than 80x of energy consumption as well as sensing time.…”