An architecture for miniaturised low power convolutional decoders

Convolutional decoders have long been important in applications where very noisy channels are encountered. The technique has become more significant with the advent of new application areas such as mobile radio systems of various types and digital magnetic recording systems. The most commonly employed decoding technique is the Viterbi algorithm (VA). This provides an optimum method for realising a convolutional decoder, but requires a large amount of digital memory, typically up to 50% of the total chip area. In this paper we describe an alternative method which we call the modified feedback decoding algorithm (MFDA). This requires no digital path memory and so can be realised entirely using analogue components. Although strictly sub-optimal, the loss of coding gain compared to the VA is only about 0.4 dB for a K=5 code.