项目名称: 非重构框架下的认知MIMO频谱感知算法研究
项目编号: No.61301101
项目类型: 青年科学基金项目
立项/批准年度: 2014
项目学科: 无线电电子学、电信技术
项目作者: 高玉龙
作者单位: 哈尔滨工业大学
项目金额: 24万元
中文摘要: MIMO和认知无线电的结合提高了系统容量和频谱利用率,但同时也增加了频谱感知的软硬件复杂度。简单地把单天线频谱感知方法直接应用到认知MIMO中,只能有限提高感知性能。因此,研究计算复杂度较低、性能较好的频谱感知算法具有重要的理论意义和实际应用价值。课题首先根据MIMO信号间和信号内相关性,扩展联合稀疏模型,提出测量矩阵优化方法,利用它们对多信号进行压缩采样得到压缩采样矩阵。然后,在非重构框架下利用非渐进随机矩阵理论直接对压缩采样矩阵进行分析,得到有无信号时压缩采样矩阵极端奇异值的不同特征,以此进行频谱感知。非重构框架下的频谱感知无需数据重构算法恢复采样数据,能降低计算复杂度,非渐进随机理论使频谱感知算法减少对主用户、信道和噪声等先验信息的依赖,克服噪声不确定性,提高算法在低信噪比下的感知性能。该研究课题能充分挖掘认知MIMO的潜在优势,为认知MIMO的应用提供具有实用价值的频谱感知算法。
中文关键词: 认知MIMO;频谱感知;联合稀疏;非重构;随机矩阵
英文摘要: The combination of MIMO and cognitive radio can improve system capacity and spectrum efficiency, but also increases hardware and software complexity of spectrum sensing. Direct application of single antenna spectrum sensing method to cognitive MIMO only finitley improve sensing performance. So, research on spectrum sensing algorithm with lower computational complexity and better performance has important theoretical significance and practical value. Firstly, according to the intra-signal and inter-signal correlation, we expand the joint sparsity model and propose the optimization method for measurement matrix, and then exploit them to compress and sample the multiple signals of MIMO system to obtain the compression sample matrix. And then, compression sample matrix is processed directly using non-asymptotic RTM under non-reconstruction frame to get the different character of extreme singular values for different receiving signals, spectrum sensing is performed based on them. The spectrum sensing algorithm under non-recontruction frame does not need data reconstruction algorithm to reconstruct sample data, which will decrease computational complication.Spectrum sensing algorithm using non-asymptotic RTM does not depend on some prior information, such as primary user, channel information and noise, and can overcom
英文关键词: cognitive MIMO;spectrum sensing;joint sparsity;non-reconstruction;random matrix