e-space
Manchester Metropolitan University's Research Repository

    Performance analysis of adaptive hybrid nonlinear preprocessors for impulsive noise mitigation over power-line channels

    Rabie, KM and Alsusa, E (2015) Performance analysis of adaptive hybrid nonlinear preprocessors for impulsive noise mitigation over power-line channels. In: 2015 IEEE International Conference on Communications (ICC), 08 June 2015 - 12 June 2015, London, UK.

    [img]
    Preview

    Available under License In Copyright.

    Download (6MB) | Preview

    Abstract

    © 2015 IEEE. Impulsive noise (IN) over power-lines can significantly corrupt communication signals. To diminish its effect, a nonlinear preprocessor is usually applied at the receiver's frond-end to blank or clip the incoming signal when it exceeds a certain threshold. Applying a combination of blanking and clipping in a hybrid fashion is characterized by two thresholds T1 and T2 (T2 = αT1), where α is a scaling factor. Previous studies assumed a fixed value for the scaling factor and found that optimizing the threshold T1 is the key to improve performance. In contrast to the existing work, in this paper we show that the performance of the hybrid technique is sensitive not only to the threshold but also to the scaling factor, and in light of this we propose to enhance the capability of this technique by optimizing the two parameters. System Performance is evaluated mathematically in terms of the probability of missed blanking/clipping (Pm), probability of IN identification (Pi) and the symbol error rate (SER) performance. In all our investigations, simulation results are provided to validate the analysis. Results reveal that the proposed scheme is superior in terms of minimizing Pm and maximizing Pi which consequently results in improving SER performance.

    Impact and Reach

    Statistics

    Activity Overview
    6 month trend
    282Downloads
    6 month trend
    408Hits

    Additional statistics for this dataset are available via IRStats2.

    Altmetric

    Repository staff only

    Edit record Edit record