A New Class of Generative Models for Burst-Error Characterization in Digital Wireless Channels

Abstract

Accurate and efficient generative models are significant for the design and performance evaluation of wireless communication protocols as well as error-control schemes. In this paper, deterministic processes are used to derive a new class of hard and soft generative models for simulation of digital wireless channels with hard and soft decision outputs, respectively. The proposed deterministic-process-based generative models (DPBGMs) are all based on a properly parameterized and sampled deterministic process followed by a threshold detector and two parallel mappers. The target hard and soft error sequences are provided by computer simulations of uncoded enhanced general packet radio service (EGPRS) systems with typical urban and rural area channels. Simulation results indicate that the proposed DPBGMs enable us to approximate very closely all the interested burst-error statistics of the target hard and soft error sequences. The validity of the suggested DPBGMs is further confirmed by the excellent match of the simulated frame-error rates and residual bit-error rates of coded EGPRS systems obtained from the target and generated error sequences