Adaptive equalization of frequency selective indoor wireless channels.

Abstract

In this thesis, we examine the performance of adaptive linear, T/2 fractionally spaced, and decision feedback equalizers (LE, FSE and DFE) for digital communication systems over indoor wireless channels affected by frequency selective fading. The channel is modeled as both fixed and time varying channels, and the data transmission rate is 10 M bits/s. The simulation results show that, with differential encoding, the quadrature phase shift keying (DQPSK) modulation scheme outperforms 4-state and 8-state 8PSK-based trellis coded modulations (D8PTCM4 and D8PTCM8), when the DFE with the least-mean-square (LMS) algorithm (DFELMS) is used. While D8PTCM4 has better BER performance than D8PTCM8 with the DFELMS over time varying channels, the reverse conclusion can be obtained over fixed channels. When the recursive least-square (RLS) algorithm is applied instead of the LMS algorithm, the systems with the DFERLS can achieve slightly better bit error rate (BER) performance than those with the DFELMS.