A measurement-based characterization of the time variation of an indoor wireless channel

To provide mobile users with multimedia information services via a wireless link, the physical, link, and network protocol layers must be able to adapt to changing quality over the wireless channel. Since feedback used to drive adaptive algorithms introduces additional complexity and system overhead, the rate at which the feedback loop operates must be balanced against the penalty for adapting too slowly. This requires the system designer to have an understanding of the timescales at which these adaptive algorithms execute, and hence an understanding of the dynamic behavior of the wireless channel. Previous efforts on characterizing this behavior have focused primarily on determining the impulse response as a function of location. This work differs in that (1) the primary focus is on characterizing the rate of change in the channel response, and (2) the measurements are performed in-vivo, using an appropriately-sized device (a notebook-sized portable terminal) to gather measurements of the frequency response in the 2.4-2.5 GHz band. For a single, moving user in an indoor office environment, it is found that signal-strength deviations of less than 9 dB can be adequately tracked with a 2 Hz update frequency, and that increasing the update frequency to 20 Hz allows tracking changes to within 3 dB