Channel estimation for massive multiple‐input and multiple‐output system based on different measurement matrices. Issue 3 (1st May 2019)

• Record Type: Journal Article
• Authors: Ma, Ruijin; Zhang, Huisheng
• Is Part Of: IET networks. Volume 8:Issue 3(2019)
• Page Start: 187
• Page End: 194
• Publication Date: 2019-05-01
• Subjects: channel estimation -- MIMO communication -- compressed sensing -- iterative methods -- matrix algebra -- least squares approximations; massive multiple‐input -- multiple‐output system -- different measurement matrices -- deterministic measurement matrix -- complementary sequence -- compressed sensing signal model -- traditional massive MIMO channels -- traditional models -- CS‐based channel estimation model -- channel estimation results -- ROMP‐based channel estimation -- orthogonal matching pursuit -- matching pursuit algorithms; Computer network architectures -- Periodicals; Computer network protocols -- Periodicals; Information networks -- Periodicals; Telecommunication systems -- Periodicals; 004.605
• DOI: 10.1049/iet-net.2018.5198
• Languages: English
• ISSNs: 2047-4954

Abstract

Abstract : This study proposes a construction method of deterministic measurement matrix based on a correlation criterion. On this basis, the authors investigated the massive multiple‐input and multiple‐output (MIMO) channel estimation algorithm using the complementary sequence as measurement matrix and obtained the compressed sensing (CS) signal model through the analysis on traditional massive MIMO channels. In this model, the measurement matrix plays the role of the pilot sequence in the traditional models. Then, the complementary sequence was determined as the measurement matrix for channel estimation. Meanwhile, the regularised orthogonal matching pursuit (ROMP), which realises accurate signal recovery through reselection of support set, was employed. Finally, the CS‐based channel estimation model was analysed, and the channel estimation results of different measurement matrices were compared through simulations. The results show that the complementary sequence outperforms the other two sequences when serving as a measurement matrix; the ROMP‐based channel estimation surpassed the orthogonal matching pursuit and matching pursuit algorithms in both computing accuracy and runtime.