论文部分内容阅读
为了解决信号重构性能差的问题,提出了一种基于广义Jaccard系数的广义正交匹配追踪(generalized orthogonal matching pursuit,g OM P)重构算法。该算法利用广义Jaccard系数相似性匹配准则替换g OM P算法中的内积度量准则,优化了通过感知矩阵来选择与残差余量最匹配原子的匹配方式。实验结果表明,该算法的重构成功率不仅高于g OMP算法,同时也高于OMP、St OMP等算法。
In order to solve the problem of poor signal reconstruction performance, a generalized orthogonal matching pursuit (gOMP) reconstruction algorithm based on generalized Jaccard coefficients is proposed. The algorithm replaces the inner product metric in g OM P algorithm by using the generalized Jaccard coefficient similarity matching criterion, and optimizes the matching method of choosing the best matching atom with residual margin by perceptual matrix. Experimental results show that the reconstruction success rate of this algorithm is not only higher than g OMP algorithm, but also higher than OMP, St OMP and other algorithms.