Predicting the outage performance of an equalized-QPSK system over indoor wireless channels

The need to design wide-band wireless indoor systems with low outage probabilities (on the order of 1% to 2%) implies the need to study system performance at the 99th percentile as a function of several variables. We investigate the performance of a 15-MBaud quantenary phase-shift-keying (QPSK) system combined with decision-feedback equalization. The study attempts to identify, and quantify, trends in the system outage performance as a function of the equalizer span, the ensemble averaged RMS-delay-spread (/spl tau/~/sub RMS/) of the environment, and the ensemble averaged input signal-to-noise ratio (SNR). We introduce a novel figure of merit 99%-SNR-loss, which when plotted as a function of the normalized-equalizer-span shows a trend in the system´s outage performance as a function of the desired parameters. These suggest a pattern that can help predict, to the first order, the performance of such a system without the need to resort time-consuming simulations. The results suggest a 0.79-dB improvement in the 99%-SNR-loss per normalized equalizer and a 2-dB improvement with a doubling of /spl tau/~/sub RMS/ with the same normalized equalizer span.