Neda Beheshti

Center of Gravity Feedback in Fading Channels

In almost all wireless communication systems, the communication channel is modeled as a fading channel with time-varying characteristics. In these channels, the receiver and/or the transmitter usually have some form of information about the current characteristics of the channel, which is usually referred to as channel state information (CSI). The main question of interest is how to optimally employ this information in order to increase the transmission rate and decrease the channel error probability .

In this thesis, we consider a simple model for fading channels and investigate the performance of a coding scheme known as ``center of gravity feedback'' under the assumption that the transmitter and the receiver have either perfect or no knowledge about the channel state and a noiseless, delayless feedback link is available from the receiver to the transmitter. In the case where both the transmitter and the receiver have perfect knowledge of the fading coefficients, two suboptimal adaptive methods are examined. These methods are based on adapting the transmit power to the channel variations and have very simple encoder and decoder structures.

We also study this coding scheme under the condition that only the receiver knows the fading coefficients and compare the average transmitted energy and the probability of error with their corresponding values in the AWGN channel.