Authors: Songnan Xi, Hsiao-Chun Wu, Tho Le-Ngoc, Arjan Durresi
Addresses: Department of Electrical and Computer Engineering, Louisiana State University, Baton Rouge, LA 70803, USA. ' Department of Electrical and Computer Engineering, Louisiana State University, Baton Rouge, LA 70803, USA. ' Department of Electrical and Computer Engineering, McGill University, Montreal, Quebec, H3A 2A7, Canada. ' Department of Computer Science and Information Science, Purdue School of Science, IUPUI, Indianapolis, IN 46202, USA
Abstract: Least-square channel estimation techniques usually involve the large-dimensional matrix inversion, whose heavy computational complexity cannot be extendable for long channel filters. Maximum-Length Shift-Register (MLSR) sequences, or m-sequences, possess the well controlled second–order cyclic statistics and have been used as the training sequences for least-square channel estimators. In this paper, we analyse the statistical characteristics of m-sequences and design a corresponding highly computationally-efficient channel estimation algorithm. Two crucial measures, namely mean-square error and computational complexity, are evaluated thereupon. It can be justified that our proposed algorithm can achieve both efficiency and satisfactory performance for communication channel estimation.
Keywords: channel estimation; m-sequences; least square; fast estimation algorithm; shift-register sequences; communication channels.
International Journal of Wireless and Mobile Computing, 2010 Vol.4 No.2, pp.148 - 152
Published online: 07 May 2010 *Full-text access for editors Access for subscribers Purchase this article Comment on this article