This work quantifies the accuracy of bit error rate (BER) estimates produced by Monte Carlo simulations by carefully applying confidence interval estimation techniques. Due to numerical difficulties, some previous work in this area has assumed that the BER statistic possessed a Gaussian distribution. This work demonstrates that in some important regions the estimate is decidedly non-Gaussian, and application of central limit theorem arguments can result in errors in excess of an order of magnitude. This work investigates the accuracy of common approximations and the feasibility of exact calculation of confidence intervals, and presents a novel polynomial class approximation. By combining the new approximation with more conventional approaches, an algorithm is developed for estimating confidence intervals of BER estimates. The algorithm is nonrecursive and numerically stable, requires a trivial amount of compute time to evaluate, has a small margin of error, and can be used for all error rates less than 0.5.
K. L. Kosbar and T. F. Chang, "Interval Estimation and Monte Carlo Simulation of Digital Communication Systems," MILCOM '92 Conference Record - Communications - Fusing Command, Control and Intelligence, pp. 77-81, Institute of Electrical and Electronics Engineers (IEEE), Oct 1992.
The definitive version is available at https://doi.org/10.1109/MILCOM.1992.244088
1992 IEEE Military Communications Conference, MILCOM '92 (1992: Oct. 11-14, San Diego, CA)
Electrical and Computer Engineering
Keywords and Phrases
BER Estimates; Monte Carlo Methods; Monte Carlo Simulation; Accuracy; Bit Error Rate; Central Limit Theorem; Confidence Interval Estimation; Digital Communication Systems; Digital Simulation; Error Statistics; NonGaussian Distribution; Nonrecursive Algorithm; Polynomial Class Approximation; Telecommunications Computing
International Standard Book Number (ISBN)
Article - Conference proceedings
© 1992 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.