Approximate Best Linear Unbiased Channel Estimation for Multi-Antenna Frequency Selective Channels with Applications to Digital TV systems Serdar ¨ Ozen a,b , Christopher Pladdy b , S. M. Nerayanuru b , Mark J. Fimoff b , and Michael D. Zoltowski c a ˙ Izmir Institute of Technology, Dept. of Electrical & Electronics Eng., Urla, ˙ Izmir, Turkey; b Zenith Electronics R&D Center, Lincolnshire, IL 60069, USA; c Purdue University, School of ECE, West Lafayette, IN 47907, USA ABSTRACT We provide an iterative and a non-iterative channel impulse response (CIR) estimation algorithm for communication re- ceivers with multiple-antenna. Our algorithm is best suited for communication systems which utilize a periodically trans- mitted training sequence within a continuous stream of information symbols, and the receivers for this particular system are expected work in a severe frequency selective multipath environment with long delay spreads relative to the length of the training sequence. The iterative procedure calculates the (semi-blind) Best Linear Unbiased Estimate (BLUE) of the CIR. The non-iterative version is an approximation to the BLUE CIR estimate, denoted by a-BLUE, achieving almost similar performance, with much lower complexity. Indeed we show that, with reasonable assumptions, a-BLUE channel estimate can be obtained by using a stored copy of a pre-computed matrix in the receiver which enables the use of the initial CIR estimate by the subsequent equalizer tap weight calculator. Simulation results are provided to demonstrate the performance of the novel algorithms for 8-VSB ATSC Digital TV system. We also provide a simulation study of the robustness of the a-BLUE algorithm to timing and carrier phase offsets. Keywords: channel estimation, least squares, best linear unbiased estimator, digital television, multi-antenna receivers, 8-VSB 1. INTRODUCTION For the communications systems utilizing periodically transmitted training sequence, least-squares (LS) based channel estimation or the correlation based channel estimation algorithms have been the most widely used two alternatives. 1 Both methods use a stored copy of the known transmitted training sequence at the receiver. The properties and the length of the training sequence are generally different depending on the particular communication system’s standard specifications. However most channel estimation schemes ignore the baseline noise term which occurs due to the correlation of the stored copy of the training sequence with the unknown symbols adjacent to transmitted training sequence, as well as the additive channel noise. 1, 2 In the sequel, we provide (semi-blind) Best Linear Unbiased Estimate (BLUE) and approximate BLUE (a-BLUE) channel estimators for communication systems using a periodically transmitted training sequence. Our novel CIR estimation algorithms can be considered as semi-blind techniques since these methods take advantage of the statistics of the data. 3 Although the examples following the derivations of the BLUE and the a-BLUE channel estimators will be drawn from the ATSC digital TV 8-VSB system, 4 to the best of our knowledge it could be applied with minor modifications to any digital communication system with linear modulation which employs a periodically transmitted training sequence. The novel algorithm presented in the sequel is targeted for the systems that are desired to work with channels having long delay spreads L d ; in particular we consider the case where (NT + 1)/2 <L d < NT , where NT is the duration of the available training sequence. For instance the 8-VSB digital TV system has 728 training symbols, whereas the delay spreads of the terrestrial channels have been observed to be at least 400-500 symbols long. 5, 6 The a-BLUE algorithm can Further author information: (Send correspondence to S. ¨ O) S. ¨ O: E-mail: serdarozen@iyte.edu.tr, Tel: +90 232 750 6510; Fax: +90 232 750 6505 C.P., S.M.N., M.J.F.: E-mail: {christopher.pladdy, snerayanuru, mark.fimoff}@zenith.com, Telephone: (847) 941 8911 M.D.Z: E-mail: mikedz@ecn.purdue.edu, Telephone: (765) 494 3512 Digital Wireless Communications VI, edited by Raghuveer M. Rao, Sohail A. Dianat, Michael D. Zoltowski, Proceedings of SPIE Vol. 5440 (SPIE, Bellingham, WA, 2004) 0277-786X/04/$15 · doi: 10.1117/12.543164 33 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 06/17/2016 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx