Initial estimation of communication efficiency of indoor wireless channels

We report theoretical/numerical estimation of ultimate limits of bandwidth efficient delivery of high bit-rate digital signals indoors. Specifically, we analyze the Crawford Hill Bell Laboratories building. Single omni transmit and receive antennas are assumed. The signals incur attenuation and distortion due to multipath and at high bit-rates this frequency selectivity causes ISI. Moreover, there is impairment by gaussian noise. Transmit power and bandwidth constraints limit communications efficiency. We illustrate this limiting of efficiency in examples assuming 5.2 GHz carrier, 10 MHz bandwidth and up to 1 W transmitted power. The experimentally based WiSE ray-tracing tool models the channels from a base on the ceiling to workspaces in various rooms. Movement within a workspace causes channel changes. Computing the probability distribution of capacity shows substantial capacity for the omni-omni case with 10 MHz bandwidth for 100 mW transmitted. For the most distant offices, for at least 95% of the area in a workspace, we obtain 3.6 bps/Hz. We also examine diversity methods, illustrating substantial gains with nth order optimum combining (OC(n)): OC(2) improves capacity by over 35% in rooms where capacity is the lowest while OC(4) improves it 70%. We put the results in a pragmatic perspective by highlighting the bit-rates achievable with decision feedback equalization