Reduced rank technique for joint channel estimation and joint data detection in TD-SCDMA systems

• Main Author: Marzook, Ali Kamil
• Format: Thesis
• Language: English
• Published: 2013
• Online Access: http://psasir.upm.edu.my/id/eprint/56192/1/FK%202013%20126RR.pdf; http://psasir.upm.edu.my/id/eprint/56192/

Time Division–Synchronous Code Division Multiple Access (TD-SCDMA) has been classified as one of 3G wireless communication systems in the last years. The physical layers of this system have adopted several advanced technologies, such as:joint detections, uplink synchronization, and smart antennas, making it capable of meeting the requirements of 4G systems directly. One of the essential parameters that affects the performance and reliability of TDSCDMA systems in current wireless applications is channel estimation. A novel channel estimation method is presented in this thesis relying on the reduction of the rank order of the total channel matrix H. We have exploited the rank deficient of H to reduce the number of parameters characterizing the channel matrix and model it with the lower number of parameters that actually require a description of H. The adopted reduced rank technique is based on Singular Value Decomposition (SVD). The proposed method has declared the superiority over the conventional methods that are being used in the current applications of TD-SCDMA systems. The Mean Square Error (MSE) and Bit Error Rate (BER) metrics analyzing the channel matrix precision among the methods is also suggested in this thesis for comparison;Least square (LS), post-processing, and the proposed reduced rank methods. In addition, the technique is proposed as a new algorithm for joint channel est imation of the multi-cell model based on reduced rank technique handling the active users in serving cell and the strong interferers from the neighboring cells. The new technique can improve the TD-SCDMA systems with low rank order processing and low computation complexity. Unlike the conventional models of multi-cell channel parameters, the channel matrix of reduced rank multi-cell model has been estimated in parsimony to involve the detected users from the serving and neighboring cells.This proposed method can acquire and claim a remarkable improvement in the performance especially under the worst interference situations. Finally, a smart antenna system has been merged with reduced rank technique to introduce a robust performance of TD-SCDMA system. A low effective rank order of ST channel can be adopted to construct the low rank based receiver with low complexity and remarkable performance. The simulation results prove that the reduced rank technique of TD-SCDMA systems can lead to considerable system performance enhancements and can efficiently mitigate interference. In single cell networks, rank two based detector has the pioneered order along SNR and it introduces improvement about 4 dB gained over the traditional LS full rank based detectors. In multi-cell networks, rank one estimator is suitable for case one of multi-cell and provide 9 dB enhancement over the multi-cell LS estimator while; rank two is more suitable for case two of the multicell and provide at least 5 dB enhancement along the SNR axis over the multi-cell LS estimator. Finally in the ST TD-SCDMA system, rank three based detector outperforms traditional LS detector by having 4 dB gains at BER of 10-3 with 8 arrays model at 3Km/h. The reduced rank technique is used to control the complexity of channel estimation. The performance of all the schemes are simulated and analyzed by using MatLab 7.9 simulator.