Research On Algorithms Of Spectrum Sensing And Allocation In Cognitive Radio Networks

With the development of information technology and wireless communication technology, our demands for radio spectrum resources increase rapidly. Cognitive radio technology breaks the bondage of current fixed spectrum allocation policy. Spectrum sensing and allocation are two of the main research directions of cognitive radio technology. By dynamic spectrum sensing and opportunistic spectrum accessing, cognitive radio can vastly increase the spectrum efficiency.This article is focused on the intelligent optimization algorithms which could be used in the technology of spectrum sensing in cooperative cognitive network and allocation in cellular heterogeneous network, and three intelligent optimization algorithms in cognitive radio network have been put forward.For spectrum sensing in cooperative cognitive network, this article explains the model of cooperative cognitive radio network then proposes an intelligent optimization algorithm called Continuous Quantum-inspired Frog Leaping algorithm (CQSFL) based on frog leaping algorithm in order to accurately detect the weak primary signal. Different secondary users have different weights according to their performance. Simulation results for cognitive radio system are provided to show that the designed spectrum sensing algorithm is superior to some previous spectrum sensing algorithm in convergence speed, precision and probability of detection.For spectrum allocation in cellular heterogeneous network, this paper focuses on how to get the largest network benefits, including max-sum-reward and max-proportional-fair reward. Firstly, two intelligent optimization algorithms, Multivariable Quantum-inspired Particle Swarm Optimization algorithm (M-QPSO) and Grouped Quantum-inspired Particle Swarm Optimization algorithm (G-QPSO) based on Particle Swarm Optimization algorithm have been proposed to solve the problem of single-objective spectrum allocation. In these two algorithms, each solution vector of quantum particles is regarded as a scheme of spectrum allocation according to different network benefit functions. Secondly, a multi-objective optimization algorithm to get both Max-Sum-Reward network benefits and Max-Fair-Reward network benefits is proposed. Simulation results show that the network benefits are improved by both of these single-objective optimization algorithms and multi-objective optimization algorithm.