Aim. The introduction of the full paper reviews Refs. 1 through 6 and then proposes the research of this paper, which is explained in sections 1,2, and 3. Their core is: "In the ideal cognitive radio (CR) systems, CR users are capable of sensing wide band spectrum simultaneously. However, due to the physical limits of front end devices, the real time entire spectrum sensing is difficult to be actualized. In order to tackle this problem, we propose a wide band spectrum sensing approach based on compressive sampling theory. First, we design a centralized multi-user parallel compressive sampling framework which realizes the high speed analog signal sampling of wide band spectrum sensing through low speed AD equipped CR users. Moreover, according to the compressive sampling theory, the proposed approach sufficiently utilizes the sparsity of primary signal on frequency domain. Thus, the base station can recover the frequency domain information of primary user by far less samples than Nyquist rate. Finally, the spectrum occupancy status can be determined by the recovery information." The simulation results, presented in Figs. 3 and 4, reveal preliminarily that: ( 1 ) the proposed method has the higher detection probability and shorter sensing time than energy detection; (2) it needs only 10％ memory units compared to conventional sensing schemes.%在理想的认知无线电系统中,认知用户能够同时感知很宽的频带,而受限于前端器件的物理性能,实时全频带检测很难实现.针对这一问题,文章提出一种基于压缩采样的宽带频谱感知方法.首先设计了一种集中式多认知用户分组并行压缩采样架构,使每个认知用户只需要一个低速率AD转换器即可实现宽带频谱感知中的模拟信号高速采样.另外,根据压缩采样理论,该方法充分利用了主用户信号在频域上的稀疏性,基站只需要远少于奈奎斯特采样点数的观测值即可对主用户信号的频域信息进行恢复,并通过恢复出的信息对频谱占用状态进行判定.仿真结果表明,该压缩检测算法相比于传统能量检测具有较高的系统检测概率和较短的检测时间,且只需要10%的存储单元.
展开▼
