Methods for the evaluation of the outage probability in mobile radio systems have been widely investigated, but analytical results have been published only for Rayleigh fading radio channels. For the case of Rice fading, only approximated results have been found. In this paper, an exact expression of the outage probability for a subscriber of a CDMA cellular mobile radio system, with the radio channel impaired by Rice fading and log-normal shadowing superimposed is given.
1. Introduction
Cellular radio capaciv, i.e. the number of admitted users per cell for a given total available bandwidth, is a key parameter in the design of cellular mobile radio systems. It depends on many factors [I], including available bandwidth, service quality, traffic statistic, access protocol, shape and size of service area, antenna gain and antenna diversity, radio propagation statistic, efficiency of the multiple access scheme and of the modulation format, etc. The effect of some of these factors is summarized into a parameter known as outage probabilky, namely the probability not to achieve a satisfactory signal-to-interference plus noise ratio.
The analysis reported in this paper is based on the evaluation of the probability that the ratio between the power of the useful signal and the powers of the interfering signals exceeds a given threshold; the effect of additive white Gaussian noise is also taken into account. Each received signal (useful or interference) is a multiple stochastic random variable (rv) and, precisely, it depends on the random distance between the transmitter and the receiver locations, on the transmitter power, on the antenna gains, on the presence of multiple antennas (diversity), on thefading and shadowing effects, on the presence or absence of power control [2]. Except for diversity and power control, all previous contributions have been taken into account in this work. In particular, the fading is modeled as Ricean and the shadowing as log-normal; moreover, Rayleigh fading turns out to be a particular case of Rice fading.
Many papers have considered the computation of the outage probability by modeling the fading as a Rayleigh