本文選題：認(rèn)知無線電 + 頻譜感知��； 參考：《北京郵電大學(xué)》2015年碩士論文

【摘要】：隨著通信技術(shù)的發(fā)展,人們對無線電頻譜資源的需求日益強烈,為提高頻譜利用率,認(rèn)知無線電技術(shù)應(yīng)運而生。認(rèn)知無線電技術(shù)是指認(rèn)知用戶可以動態(tài)感知空閑頻譜,并在保證不對授權(quán)用戶造成干擾的前提下?lián)駲C接入頻譜進行通信的技術(shù)。頻譜感知與頻譜分配是認(rèn)知無線技術(shù)的兩個重要研究內(nèi)容。 本課題以頻譜感知與分配的智能優(yōu)化算法研究為核心,分別對認(rèn)知網(wǎng)絡(luò)的協(xié)同頻譜感知和蜂窩異構(gòu)網(wǎng)絡(luò)的頻譜分配問題進行了研究,提出了相應(yīng)的智能優(yōu)化算法。 對于認(rèn)知網(wǎng)絡(luò)的協(xié)同頻譜感知技術(shù)的研究,論文首先分析了認(rèn)知網(wǎng)絡(luò)模型,并以使系統(tǒng)獲得較大的檢測概率為目標(biāo),提出了適用于頻譜感知問題的連續(xù)量子蛙跳算法。通過該算法為感知可靠度不同的認(rèn)知用戶賦予不同的權(quán)重,進行協(xié)同感知。該算法具備比傳統(tǒng)智能算法更高的收斂精度和更快的收斂速度,可在相同的虛警概率下,使認(rèn)知網(wǎng)絡(luò)獲得更大的檢測概率。 對于認(rèn)知異構(gòu)網(wǎng)絡(luò)頻譜分配技術(shù),研究分為兩部分：首先進行單目標(biāo)優(yōu)化問題的研究,將量子的概念引入粒子群算法,針對粒子演進過程中易陷入局部最優(yōu)解的問題,提出了兩種離散量子粒子群優(yōu)化算法——多變量量子粒子群優(yōu)化算法和混合量子粒子群優(yōu)化算法,并將其應(yīng)用于認(rèn)知無線電頻譜分配這一離散優(yōu)化問題,其中,算法的每一個即為一種頻譜分配方案；在此基礎(chǔ)上,提出了多目標(biāo)量子粒子群優(yōu)化算法,解決了具有多個網(wǎng)絡(luò)效益目標(biāo)函數(shù)的頻譜分配問題。本文提出的單目標(biāo)量子粒子群優(yōu)化算法具備比現(xiàn)有智能算法更高的收斂精度,可使網(wǎng)絡(luò)獲得更大的網(wǎng)絡(luò)效益；本文提出的多目標(biāo)量子粒子群優(yōu)化算法實現(xiàn)了最大網(wǎng)絡(luò)效益和最大比例公平網(wǎng)絡(luò)效益的兼顧。
[Abstract]:With the development of communication technology, the demand for radio spectrum resources is increasingly strong. In order to improve spectrum efficiency, cognitive radio technology emerges as the times require. Cognitive radio technology refers to the technology that cognitive users can dynamically perceive the idle spectrum and access the spectrum without interfering with the authorized users. Spectrum sensing and spectrum allocation are two important contents of cognitive wireless technology. Based on the research of the intelligent optimization algorithm of spectrum sensing and allocation, this paper studies the cooperative spectrum sensing of cognitive networks and the spectrum allocation of cellular heterogeneous networks, and puts forward the corresponding intelligent optimization algorithms. For the research of cooperative spectrum sensing in cognitive networks, this paper first analyzes the cognitive network model, and proposes a continuous quantum leapfrog algorithm for spectrum sensing. The algorithm assigns different weights to cognitive users with different perceptual reliability and carries out cooperative perception. The proposed algorithm has higher convergence accuracy and faster convergence speed than the traditional intelligent algorithm, which can make the cognitive network obtain a higher detection probability under the same false alarm probability. The spectrum allocation technology of cognitive heterogeneous networks is divided into two parts: firstly, the single objective optimization problem is studied, and the concept of quantum is introduced into particle swarm optimization algorithm to solve the problem that particle evolution is prone to fall into local optimal solution. Two discrete quantum particle swarm optimization algorithms, multivariable quantum particle swarm optimization and hybrid quantum particle swarm optimization, are proposed and applied to the discrete optimization problem of spectrum allocation of cognitive radio. Based on this, a multi-objective quantum particle swarm optimization (QPSO) algorithm is proposed to solve the spectrum allocation problem with multiple network benefit objective functions. The single objective Quantum Particle Swarm Optimization (QPSO) algorithm proposed in this paper has higher convergence accuracy than the existing intelligent algorithm and can make the network gain more network benefits. The multi-objective quantum particle swarm optimization (QPSO) algorithm proposed in this paper realizes both the maximum network benefit and the maximum proportional fair network benefit.
【學(xué)位授予單位】：北京郵電大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2015
【分類號】：TN925

【參考文獻】

相關(guān)期刊論文 前1條

1 李寧,鄒彤,孫德寶,秦元慶;基于粒子群的多目標(biāo)優(yōu)化算法[J];計算機工程與應(yīng)用;2005年23期

，

本文編號：2035732