Experimental BER Performance Evolution of Equalizations and Matched Filter Bounds for Single-Carrier WiMax Radio in 3.5 GHz

This paper presents an experimental evaluation of bit-error-rate (BER) performances for adaptive training algorithms and obtaining matched filter bound in real-time WiMax (3.5 GHz) radio channels. Without involving with any modified or improved versions, two conventional adaptive equalizer training methods, the least mean squares (LMS) and recursive least squares (RLS), are implemented in a C code running over sampled received sequence and their performances are compared in different modulation techniques before and after coding. In experimental trails the obtained channel impulse responses were generally non-minimum phase and the obtained BER performance of receiver using LMS training has been quite close to whom using RLS particularly after coding, in real-time channel conditions.