Packet-Error-Rate Analysis Using Markov Models of the Signal-to-Interference Ratio for Mobile Packet Systems

To analyze the packet error rate (PER) of interference-limited mobile packet systems, in this paper, two Markov models of the time-varying signal-to-interference ratio (SIR) are derived over frequency-selective fading channels. The two Markov models are first-order two-state for the following two cases: 1) infinitesimal interval and 2) finite-time interval. The state and state transition probabilities are obtained from both the level crossing rate and average outage duration of the fluctuating SIR. The PER is analytically derived using the Markov model in the case of a finite time interval, and the analysis can be applied not only to cellular communication systems but to ad hoc mobile networks as well. The analytical results are exact when the average powers of both the desired and interfering signal components are equal for all paths but include an approximation in the other cases. The analytical results of the PER agree well with the results that were obtained by computer simulations. It is also shown that the PER characteristics of cellular systems and of ad hoc networks are almost the same.