Detection Guided Decision Feedback IIR Equalizer for sparse channels

Publication Type:
Conference
Issue Date:
2007-03-12T22:05:45Z
Full metadata record
For the 4 wire-loop telephony circuit echo paths [1], mobile radio channels [2] and other channels, the impulse response of the channel has sparsely active" characteristics. We consider a discrete NLMS adaptive IIR equalizer connected in cascade with the channel to compensate for the system degradation due to Intersymbol Interference (ISI) [3]. The conventional NLMS adaptive IIR equalizer adapts each and every tap at each sample interval. However, this approach suffers from the slow convergence problem. Motivated by the sparse channel characteristics, we investigate the use of active parameter detection technique within the NLMS adapted IIR equalizer. The proposed detection technique is based on that employed for channel estimation applications in [4] [5]. The aim is to adapt only the active (or significant) taps of the equalizer. Improved convergence rates are achieved. The decision feedback technique is also incorporated in the IIR equalizer to cancel ISI [6]. Simulations of this newly proposed active-parameter detection guided IIR equalizer show the favourable accelerated convergence"
Please use this identifier to cite or link to this item: