Modeling temporal channel variations in indoor wireless environments

The temporal variations of an indoor wireless channel can have a significant impact on the performance of the underlying wireless system. Therefore, there is a need to understand and model the temporal processes associated with the physical propagation phenomena. Typically, wireless channels are modeled through techniques that do not capture the temporal variability in the underlying channel response, but instead represent a snapshot or ensemble description of the channel. In this paper, we examine the problem of modeling the temporal variations of the channel response in an indoor environment. We have conducted channel sounding measurements for several typical office scenarios, i.e. static environment, people around a lunch table, people sitting and moving in an office. Using these traces, we quantified and compared the time variations under various conditions of the movement. We show that, in an indoor environment, only some taps of the channel impulse response vary significantly. We use autoregressive processes to model the time-varying part of the taps, and show this to be an excellent representation for channel responses in a typical indoor environment.