摘要
测量矩阵在压缩感知中有着重要作用,因而如何构造出性能优异的测量矩阵一直是该领域的研究重点。因此,通过里德-索罗门(Reed Solomon,RS)码得到一个生成矩阵,然后将张量积应用到RS码生成矩阵上产生一个新的矩阵,最后利用该矩阵构造出一个新的测量矩阵,使得该测量矩阵的相关性渐近Welch界,从而可以达到性能渐近最优。仿真结果表明,构造的测量矩阵相对于BCH矩阵和高斯矩阵在性能方面有较大提升。
Measurement matrix plays an important role in compressed sensing, and thus how to construct a high-performance measurement matrix is always the hot research topic in this field. The RS （Reed Salomon） code is used to obtain a code generator matrix, then a new matrix acquired by applying the tensor product on RS（Reed-Solomon） generator matrix. Based on it, a new measurement matrix is constructed, making the correlation coefficient of this matrix asymptotic Welch, so as to achieve asymptotic optimal performance. Simulation results indicate that, as compared to BCH matrix and Gaussian matrix, the constructed measurement matrix is greatly improved in terms of performance.
出处
《通信技术》
2016年第9期1139-1143,共5页
Communications Technology
关键词
压缩感知
测量矩阵
RS码
张量积
compressed sensing
measurement matrix
RS code
tensor product
