Computer simulation and evaluation of mobile radio voice communication systems
Chia-Chi Huang · 1984
In urban and suburban areas, narrowband mobile radio voice transmission suffers from multipath fading phenomena, which are time-varying and related to geographic environment, carrier frequency, and vehicle speed. Traditionally, mobile radio voice communications have used narrowband frequency modulation (FM). An FM channel usually requires 25-30 kHz of bandwidth, and the performance is heavily affected by multipath. Much recent research effort has been directed toward finding better transmission schemes. On the one hand, advances in solid-state technology and the demand for spectral efficiency have led to the proposal of single-sideband amplitude modulation (SSBAM) with a pilot tone as a better alternative than FM. On the other hand, due to the increasing use of digital techniques and the possibility of more robust transmission, there is an increasing demand for consideration of digital techniques for mobile radio voice transmission. Our research evalutes the performance of various mobile radio voice processing and transmission methods, both analog and digital, using computer simulation. In the analog category, we compare amplitude-companded SSBAM (with pilot) with traditional FM, in the face of receiver front-end noise and co-channel interference. In the digital category, we compare a proposed speech waveform coder, called controlled adaptive prediction delta modulation (CAPDM), with traditional continuously variable slope delta modulation (CVSD); channel coding schemes and bit scrambling techniques are also evaluated. The analog and digital schemes are also compared. From our simulation, we find that SSBAM with two-stage amplitude companding and a pilot tone, where the pilot tone is used to combat fading through a technique called feedforward signal regeneration (FFSR), is more robust than traditional FM in the face of receiver front-end noise and multipath fading. In co-channel interference, however, we find that FM still has an advantage over SSBAM. In the digital category, assuming DPSK modulation at a transmission rate of 15 kbits/sec, we find that our CAPDM coder, which uses both adaptive companding and adaptive prediction, performs much better than the traditional CVSD coder in multipath fading, in both front-end noise and co-channel interference. (Abstract shortened with permission of author.)