Capacity analysis of time-varying flat-fading channels using particle methods

We consider the problem of computing the information rate of a first-order approximation of Clarkepsilas time-varying flat-fading channel model, and apply the recently-developed particle-filter method for estimating the information rate of a continuous-valued channel with memory. We compare the results with various bounds, both classic and recent, and we find that the information rates obtained using the particle method fall between recently-established tight bounds. We also investigate the effects of the numbers of particles and of simulated time-steps on the estimated information rate. Such results are useful in providing an upper bound on the data rates achievable in a wireless communication system.