A performance model for data throughput with adaptive modulation

This paper presents a comprehensive analytical model for determining the maximum average data throughput achievable through adaptive modulation in a wireless TDMA network environment. The model allows for designing rate adaptation algorithms that exploit higher transmission rates for data devices operating in superior channel conditions. It incorporates the probability of retransmissions due to device movement and channel estimation errors such that data throughput loss resulting from such retransmissions can be accounted for in designing rate adaptation algorithms while seeking to maximize transmission rate. The model can be used in various system and channel environments to determine the potential benefits of adaptive modulation and to derive the conditions under which these benefits are attainable. As a demonstrating example, the paper considers a rate assignment policy that employs adaptive modulation for every fresh transmission but switches to a safe mode should retransmissions occur, where channel state inaccuracy is accounted for when choosing the transmission rate in the adaptive mode and the lowest transmission rate is used in the safe mode. The results show how the model can be used to assess data throughput for a given system and how it can be used to optimize a rate assignment algorithm by containing retransmissions