On Adaptive Threshold Adjustment with Error Rate Constraints for Adaptive Modulation and Coding Systems with Hybrid ARQ

Contrary to real-time services, for data-centric services that are tolerant of latency, transmission with modulation and coding rates selected according to the instantaneous channel condition can increase the system throughput. Nonetheless, the switching thresholds for using different modulation and coding schemes (MCS) highly depend on the channel and system characteristics, and are usually difficult to be determined offline. In this paper, we revisit the adaptive threshold adjustment algorithms proposed previously for adaptive modulation and coding (AMC) systems. By providing an analysis that further considers hybrid ARQ and the channel statistics, we show that the error rate conditions these algorithms use to adjust the thresholds do not necessarily reflect the actual error rates (and throughput) achieved by the MCSs. For systems or applications that have requirements on the error rates, this property is undesirable. In order to offer precise control of the error rates and guarantee that the system meets the desired performance, further conditions and modifications are derived