Comparison between Two Algorithms of 32kb/s ADPCM using QAM Signal at 16.8kb/s

Abstract

This paper studies the  comparison between two
algorithms of 32kb/s ADPCM systems. The first algorithm uses
4-bit quantizer with sampling rate of 8000 sample/sec, and the
second algorithm uses 5-bit quantizer with sampling rate of 6400
sample/sec. The comparison is done using QAM signal at data rate
of 16.8kb/s. Two models of QAM signals are used, the first model
operates at symbol rate of 2400 baud   with each symbol is
represented by 7-bit, while, the second model operates at symbol
rate of 2800 baud with each symbol is represented by 6-bit. The
contribution of this paper is that sending the second model of
QAM signal over ADPCM with 5-bit quantizer. Simulation results
show that the performance of ADPCM with 5-bit quantizer is
better than 4-bit quantizer for both models of QAM signals. Also,
the performance of ADPCM using second model is better than the
first model. Furthermore, the performance with circular
constellation is better than rectangular one