【摘要】：認知無線電技術是解決快速發(fā)展的無線應用和低效的固定頻譜分配方式之間矛盾的一種有效手段。然而主用戶的搶占式優(yōu)先權以及空閑信道的空時變化特性導致認知無線電網(wǎng)絡的性能極不穩(wěn)定。隨著網(wǎng)絡融合技術的發(fā)展,異構(gòu)網(wǎng)絡的普及,認知無線電網(wǎng)絡與多個授權網(wǎng)絡交互構(gòu)成的多授權認知無線電網(wǎng)絡逐漸成為可能,其可以給次用戶提供更多的頻譜資源和網(wǎng)絡選擇。針對現(xiàn)有感知技術在多授權認知無線電網(wǎng)絡下存在感知精度較低現(xiàn)象,研究了次用戶的合作感知調(diào)度問題�？紤]到錯誤的感知結(jié)果會對次用戶的接入造成影響,基于合作感知調(diào)度得到的感知精度,研究了非完美感知下次用戶接入時的網(wǎng)絡選擇問題。采用連續(xù)時間馬爾可夫鏈模型對多授權認知無線電網(wǎng)絡進行建模與頻譜特性分析。建立無窮小生成矩陣,計算了網(wǎng)絡的穩(wěn)態(tài)概率以及已知信道上次用戶分布情況的條件穩(wěn)態(tài)概率。針對多授權認知無線電網(wǎng)絡下的頻譜感知精度較低問題,提出了一種基于離散粒子群優(yōu)化算法的合作感知調(diào)度方案。綜合考慮了各個次用戶感知能力的差異性、頻譜使用的統(tǒng)計特性以及主用戶的抗干擾能力等因素,分別從主用戶和次用戶的角度出發(fā),建立了兩個關于次用戶合作感知調(diào)度方案的整數(shù)規(guī)劃問題。采用離散粒子群優(yōu)化算法求解所建問題,并與隨機調(diào)度方案和基于信噪比的貪婪調(diào)度方案進行仿真驗證。仿真結(jié)果表明,本文提出的基于離散粒子群優(yōu)化算法的合作感知調(diào)度方案得到的頻譜感知精度高于另外兩種合作感知調(diào)度方案�；谏鲜龅玫降念l譜感知精度,研究了次用戶接入時的網(wǎng)絡選擇問題,并提出了一種次優(yōu)化網(wǎng)絡選擇方案。分析了非完美感知下次用戶的接入行為,計算了非完美感知下網(wǎng)絡的穩(wěn)態(tài)概率,次用戶的阻塞率和掉話率。以最大化認知無線電網(wǎng)絡的吞吐量為目標,建立了網(wǎng)絡選擇問題的數(shù)學模型。采用次優(yōu)化網(wǎng)絡選擇算法通過降維逐態(tài)優(yōu)化的方式求解目標函數(shù),并與隨機網(wǎng)絡選擇方案和基于空閑信道數(shù)的貪婪網(wǎng)絡選擇方案進行仿真對比。仿真結(jié)果表明,本文提出的次優(yōu)化網(wǎng)絡選擇方案得到的吞吐量高于其他兩種網(wǎng)絡選擇方案。
[Abstract]:Cognitive radio technology is an effective means to solve the contradiction between the rapid development of wireless applications and inefficient fixed spectrum allocation. However, the preemptive priority of the primary user and the space-time characteristics of the free channel lead to the extremely unstable performance of the cognitive radio network. With the development of network fusion technology and the popularity of heterogeneous networks, it is possible that cognitive radio networks interact with multiple authorized networks. It can provide secondary users with more spectrum resources and network choices. In order to solve the problem of low perception precision in multi-authorization cognitive radio networks, the cooperative perceptual scheduling problem of secondary users is studied. Considering the effect of incorrect perception results on the access of secondary users, the problem of network selection for the next user access with imperfect perception is studied based on the perception accuracy obtained by cooperative perceptual scheduling. The continuous time Markov chain model is used to model and analyze the spectrum characteristics of multiple authorized cognitive radio networks. An infinitesimal generating matrix is established to calculate the steady-state probability of the network and the conditional steady-state probability of the known last user distribution of the channel. A cooperative perceptual scheduling scheme based on discrete particle swarm optimization (DPSO) is proposed to solve the problem of low spectral sensing accuracy in multi-authorization cognitive radio networks. Considering the difference of the perception ability of the secondary users, the statistical characteristics of the spectrum usage and the anti-interference ability of the primary users, the paper sets out from the perspective of the primary users and the secondary users, respectively. Two integer programming problems for sub-user cooperative aware scheduling schemes are established. The discrete particle swarm optimization (DPSO) algorithm is used to solve the proposed problem, which is verified by simulations with the stochastic scheduling scheme and the greedy scheduling scheme based on SNR. Simulation results show that the spectrum sensing accuracy of the proposed cooperative perceptual scheduling scheme based on discrete particle swarm optimization is higher than that of the other two cooperative perceptual scheduling schemes. Based on the spectrum sensing accuracy obtained above, the problem of network selection for secondary user access is studied, and a suboptimal network selection scheme is proposed. The access behavior of the next user with imperfect perception is analyzed and the steady-state probability of the network the blocking rate and the drop rate of the secondary user are calculated. Aiming at maximizing the throughput of cognitive radio networks, a mathematical model of network selection is established. The suboptimal network selection algorithm is used to solve the objective function by reducing dimension and state optimization, and compared with the random network selection scheme and the greedy network selection scheme based on the number of idle channels. The simulation results show that the throughput of the proposed sub-optimal network selection scheme is higher than that of the other two network selection schemes.
【學位授予單位】：哈爾濱工業(yè)大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TN925

【參考文獻】

相關期刊論文 前5條

1 杜白;;基于用戶需求和進化博弈的認知無線電網(wǎng)絡選擇[J];鄭州大學學報(工學版);2014年04期

2 王鋼;曾y，

本文編號：2289444