Optimal weighting for ML decoding of convolution code in COFDM systems

When convolution code is used in wireless coded orthogonal frequency division multiplexing (COFDM) systems, effective maximum likelihood (ML) decoding requires information on the frequency response of each subcarrier. This translates into increased hardware complexity, for instance, in the form of an additional deinterleaving block. We propose an optimal weighting scheme that combines the metrics for soft decision and channel influence into a single metric, can be performed before the deinterleaving block, and makes ML decoding possible. It shows comparable performance to ML decoding and better performance over suboptimal decoders