Date of Award
Master of Engineering Science
Electrical and Computer Engineering
Dr. Jagath Samarabandu
Dr. Xianbin Wang
Signal transmitted through a wireless channel undergoes distortion due to the pres ence of reflectors in the environment between a transmitter and a receiver as well as due to the Doppler shift caused by the relative movement of the receiver with respect to the transmitter. Distorted signal is recovered at the receiver side by means of predicting the channel responses and performing inverse operation that the channel introduced on the transmitted signal. Prediction of channel responses becomes more complex when the receiver moves with a varying speed since it directly affects the auto-correlation of the channel responses.
First part of this thesis provides a solution for recovering the transmitted data when the receiver is moving with varying speed. The system first tracks the receiver speed variations using the number of deep fadings (nulls) in the received signal enve lope of one sub-carrier during a fixed time period. If there is a significant change in receiver speed then the Kalman filter parameters are calculated and updated. Future channel responses are predicted using the updated Kalman filter parameters and used in equalizer to recover the distorted signal. The performance and computational effi ciency of the proposed system outperforms the conventional system which calculates predictor parameters at a fixed interval.
Second part of the thesis presents an adaptive modulation technique based on the signal-to-noise ratio and the receiver speed. Modulation schemes for different combinations of signal-to-noise ratio and receiver speeds are obtained by selecting the higher modulation scheme with the bit error rate less than target bit error rate. Boundaries of the selected modulation schemes are found using support vector ma chine classifiers. The receiver uses the designed system to select appropriate modula tion scheme by mapping the current modulation scheme and the channel conditions. The proposed system outperforms conventional adaptive modulation technique that uses instantaneous signa-to-noise ratio by a margin of 5 dB.
Christopher, Edwin Niroshan, "Low Complexity AOFDM System for Time-varying Wireless Channels" (2011). Digitized Theses. 3408.