Combined predistortion and coded modulation in evolutionary digital satellite transmission systems
Abstract
Most communication theory is based on the assumption of a linear channel. However, nonlinearity results in a significant performance degradation in the digital satellite link, in which the satellite repeater is equipped with a nonlinear amplifier. The traveling wave tube amplifier (TWTA) in the satellite is driven near its maximum output power for efficiency, and exhibits a highly nonlinear characteristic which cannot be ignored in the evaluation of system performance. This thesis proposal is focussed on the design and analysis of higher level modulation system in the digital satellite link. The main contributions of this research can be summarized as follows. First we assess the performance of data predistorters with (or without) memory which is a powerful technique to reduce the performance degradation due to the nonlinear channel. We also propose a nonsimulation-based method which allows for efficient design of a memoryless data predistorter and various filters. Then, we consider trellis-coded modulation (TCM) in the nonlinear satellite channel as a solution to the limited power and bandwidth problem and optimize its performance by combining data predistortion and constellation optimization. We quantify the performance gain of (12,4)-PSK over 16-QAM in the nonlinear channel for both 2-dimensional TCM and 8-dimensional TCM. We also present analytical methods upper bounding the BER of TCM when mismatched Viterbi decoder is used for receiver, which is applicable when ISI cancellation technique at the transmitter side is not perfect. Although TCM performs well, it is not compatible with the existing QPSK based digital satellite services. As an alternative to the TCM scheme, we assess the performance of hierarchical coding scheme which uses multiple convolutional encoders and maintains the backward compatibility. We generalize the conventional transfer function method for the analysis of the performance of hierarchical decoding and derive bounds on the bit error rate (BER) in linear AWGN and Rayleigh fading channel, which turns out very helpful to the determination of system parameters, like decoding algorithm or constellation labeling schemes. We also investigate a method to balance the error protection for each level and assess its performance in nonlinear satellite channel by simulation.
Degree
Ph.D.
Advisors
Gelfand, Purdue University.
Subject Area
Electrical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.