一种自适应阈值选择的宽带压缩频谱感知重构算法

针对宽带压缩频谱感知算法在未知稀疏度条件下重构频谱效果不理想的问题，提出一种自适应阈值选择的改进型分段正交匹配追踪(Adaptive Threshold Option-Improved Stagewise Orthogonal Matching Pursuit, ATO-IStOMP)算法，该算法根据迭代残差的分布特性自适应地调整原子选择判决门限，使其每次迭代能够高效选择多个原子作为候选集，同时该算法利用残差比阈值对迭代终止条件进行修正，能够实现重构算法的盲停止，增强算法在低信噪比环境下的鲁棒特性。仿真结果表明，ATO-IStOMP算法能够实现对原始信号的盲重构，且在低信噪比环境下的重构性能良好。 

An Adaptive Threshold Option Reconstruction Algorithm for Wideband Compressed Spectrum Sensing

An Adaptive Threshold Option-Improved Stagewise Orthogonal Matching Pursuit (ATO- IStOMP) algorithm is proposed to solve the problems that wideband compressed spectrum sensing reconstruction algorithm is not ideal under the condition of unknown sparsity. According to the distribution characteristics of iterative residuals, the algorithm adaptively adjusts the threshold of the atomic selection decision, so that each iteration can select multiple atoms efficiently as a candidate set, At the same time, this algorithm modifies the iterative termination condition by using the residual ratio threshold, which can stop the iteration without knowing sparsity and enhance the robustness of the algorithm in low signal-to-noise ratio. The simulation results show that the ATO-IStOMP algorithm can realize the blind reconstruction of the original signal and the reconstruction performance is good under the low SNR environment.