Broadband PLC-channel equalisation in the frequency domain based on complementary sequences

Abstract

Multipath channels, like power lines, have a periodic time-variant response that impacts on data transmission. Power Line Communications (PLC) performance, in the context of a cyclic-prefix single carrier modulation scheme, can benefit from frequency domain equalisation techniques. This study proposes a frequency-domain dynamic characterisation and equalisation algorithm based on the properties of complementary sequences (CSs). This proposal takes advantage of CS properties to reduce the complexity of the algorithm by performing all the operations in the frequency domain, and without the necessity of noise/variance estimators. The proposal is compared to some wellknown methods like zero forcing and minimum mean-square error (MMSE) through the transmission of data under different PLC channels. Bit error rate (BER) is also measured using Middleton Class A impulse noise (IN) model and the performance of all methods is finally evaluated under a time-variant PLC channel model showing the importance of dynamic equalisation on PLC systems. Reported computational resources and simulations show that the proposal is four time faster than MMSE and improves the BER performance by up to 4 dB. In addition, the proposed identification algorithms shows to work properly even in PLC channels with attenuations higher than 40 dB and severe IN scenarios.