Practical blind demodulators for high-order QAM signals

Abstract

This paper examines the problem of demodulating time-dispersed digitally modulated signals, with particular emphasis on the use of ‘blind’ algorithms for initializing the demodulator in the absence of explicit training by the transmitter. It is shown that the absence of training leads to coupling between the various control loops operating in the demodulator. This coupling impacts the architecture of the demodulator, limiting the array of choices available to the designer. This paper describes a blind demodulator design which has proven practical, and illustrates several implementations of the design. The paper closes with a list of open theoretical questions, which, if answered, can lead to the next generation of blind demodulators.