Performance of Space-Time Block Coded Communication Under Time Varying Fading Channels

Abstract

ABSTRACT Broadband wireless communication must cope with critical performance limiting challenges that include time-selective and frequency-selective channels, as well as power and bandwidth constraints. Space-time (ST) coding offers an effective transmit-antenna diversity technique to combat fading. This thesis presents a simple two branch transmit diversity scheme and detection of these transmitted signals by using channel estimation based on “least squares with two types of variable forgetting factor i.e. LSn-VFF and LSn-NVFF (LSn with new VFF)” algorithm at the receiver. The data symbols are encoded using space-time block codes, and the encoded data is split into n streams, which are simultaneously transmitted using n transmitter antennas. The received signal at receiver antenna is a linear superposition of n transmitted signals perturbed by noise. Maximum likelihood decision rule at receiver for these received signals is used to choose the signal with minimum Euclidean distance between received and transmitted signal. A complex orthogonal design and the corresponding space-time block code, which provide the full diversity and full transmission rate is not possible for more than two antennas. Therefore, we have used space-time block codes with complex constellation only for two transmitter antenna system and for more than two antennas quasi-orthogonal design is used. Further, space-time block codes (STBC) with real constellation for two, four, eight (higher-order) transmitter antennas are used for advanced wireless communication systems. The channel state information is essential requirement for space-time block code detector/receiver. It is necessary to formulate the new channel estimation algorithms to improve the performance of the space-time codes in the time-varying environment using variable forgetting factor. In this research work, different channel estimation techniques are proposed. The performance characteristics of different algorithms (Least mean squares, Recursive least square, Least square of order one and two, Recursive least squares with variable forgetting factor and least square algorithm of order one and two, with both type of variable forgetting factor i.e. VFF and NVFF) are analyzed and compared. These channel estimates are further used for space-time block iv coding in different simulation examples. Simulation results depict that LSn-NVFF performs well in all time-varying channel models and orthogonal STBC’s performance is better than quasi-orthogonal STBC. LSn-NVFF performance is best for orthogonal ( 2 2 ) STBC and quasi-orthogonal ( 44 ) STBC. For 88 STBC system, LSn outperforms the all algorithms because LSn-NVFF and LSn-VFF fail to track eight channels simultaneously due to increase in computation complexity of the algorithms and increase in number of parameter adjusted by single VFF algorithms for all channels.