本文選題：信號(hào)處理 + 壓縮感知�。� 參考：《電子與信息學(xué)報(bào)》2016年07期

【摘要】：構(gòu)造確定性測(cè)量矩陣對(duì)壓縮感知理論的推廣與應(yīng)用具有重要的意義。該文源于代數(shù)編碼理論,提出一種基于二進(jìn)制序列族的確定性測(cè)量矩陣構(gòu)造算法。相關(guān)性是描述矩陣性質(zhì)的重要準(zhǔn)則,減小相關(guān)性可使重建性能提高。該文推導(dǎo)出所構(gòu)造測(cè)量矩陣的相關(guān)性小于同條件下的高斯隨機(jī)矩陣和伯努利隨機(jī)矩陣。理論分析和仿真實(shí)驗(yàn)表明,該方式構(gòu)造的測(cè)量矩陣的重建性能優(yōu)于同條件下的高斯隨機(jī)矩陣和伯努利隨機(jī)矩陣;所構(gòu)造矩陣可由線(xiàn)性反饋移位寄存器結(jié)構(gòu)實(shí)現(xiàn),易于硬件實(shí)現(xiàn),有利于壓縮感知理論的實(shí)用化。
[Abstract]:The construction of deterministic measurement matrix is of great significance to the extension and application of compression perception theory. Based on algebraic coding theory, a deterministic measurement matrix construction algorithm based on binary sequence family is proposed in this paper. Correlation is an important criterion to describe the properties of matrix. Reducing the correlation can improve the reconstruction performance. This paper deduces that the correlation of the constructed measurement matrix is less than that of the Gao Si random matrix and Bernoulli random matrix under the same conditions. Theoretical analysis and simulation experiments show that the reconstruction performance of the measurement matrix constructed by this method is better than that of Gao Si random matrix and Bernoulli random matrix under the same conditions, and the constructed matrix can be realized by linear feedback shift register structure, which is easy to be implemented in hardware. It is beneficial to the practical application of the theory of compressed perception.
【作者單位】： 西安電子科技大學(xué)ISN國(guó)家重點(diǎn)實(shí)驗(yàn)室;
【基金】：國(guó)家自然科學(xué)基金(61372069) 高等學(xué)校學(xué)科創(chuàng)新引智計(jì)劃(111計(jì)劃)(B08038)~~
【分類(lèi)號(hào)】：TN911.7

【相似文獻(xiàn)】

相關(guān)期刊論文 前6條

1 何善寶,趙春明,史志華,姜明;基于稀疏二進(jìn)制序列的低密度奇偶校驗(yàn)碼[J];通信學(xué)報(bào);2005年06期

2 周悅;朱燦焰;汪一鳴;;混沌二進(jìn)制序列性能分析[J];蘇州大學(xué)學(xué)報(bào)(工科版);2006年02期

3 D.L.Key;穆良知;;非線(xiàn)性二進(jìn)制序列的結(jié)構(gòu)及復(fù)雜度分析[J];通信技術(shù);1979年03期

4 E.L.Key;立早;;非線(xiàn)性二進(jìn)制序列產(chǎn)生器的構(gòu)造和復(fù)雜度分析[J];通信保密;1985年01期

5 聞斌;李上釗;唐鋒;歐衛(wèi)華;;基于二進(jìn)制序列構(gòu)造非規(guī)則LDPC碼[J];常熟理工學(xué)院學(xué)報(bào);2009年04期

6 ;[J];;年期

，

本文編號(hào)：1932824