1st International ICST Workshop on Microwaves and Photonics in Access Networks

Research Article

OFDM Impairment Mitigation Techniques

  • @INPROCEEDINGS{10.1109/ACCESSNETS.2007.4447139,
        author={Bharath Umasankar and Fernando Xavier},
        title={OFDM Impairment Mitigation Techniques},
        proceedings={1st International ICST Workshop on Microwaves and Photonics in Access Networks},
        publisher={IEEE},
        proceedings_a={MICPAN},
        year={2008},
        month={2},
        keywords={Attenuation  Bit error rate  Data communication  Fading  Frequency dependence  Frequency diversity  Intersymbol interference  Maximum likelihood detection  Narrowband  OFDM},
        doi={10.1109/ACCESSNETS.2007.4447139}
    }
    
  • Bharath Umasankar
    Fernando Xavier
    Year: 2008
    OFDM Impairment Mitigation Techniques
    MICPAN
    IEEE
    DOI: 10.1109/ACCESSNETS.2007.4447139
Bharath Umasankar1,*, Fernando Xavier1,*
  • 1: Dept. Electrical and Computer Engineering Ryerson University,Toronto
*Contact email: bumasank@ee.ryerson.ca, fernando@ee.ryerson.ca

Abstract

OFDM is an attractive option for high speed wired/wireless data transmission due to its inherent resistance to inter symbol interference (ISI). However, lack of frequency diversity makes it vulnerable to frequency selective impairments such as narrowband interference, frequency selective fading, frequency dependent attenuations etc. In this work, we describe a novel Adaptive Weighting (AW) technique as an efficient method to improve the BER performance of grouped linear constellation precoding (GLCP) OFDM which is used to achieve frequency diversity. In this method during maximum likelihood sequence detection (MLSD), the distance of the possible constellation points from the received symbols are weighted according to the SNR of each subcarrier before the sequence with the minimum distance is chosen. The proposed technique is shown to outperform previous GLCP-OFDM and GLCP-ACS (Adaptive Subcarrier Selection) OFDM over all SNR ranges at the expense of slightly increased complexity.