The authors derive a Bayesian decision feedback equaliser which incorporates co-channel interference compensation. By exploiting the structure of co-channel interfering signals, the proposed Bayesian decision feedback equaliser is able to distinguish an interfering signal from white noise and utilises this information to improve performance. Adaptive implementation of this Bayesian decision feedback equaliser includes identifying the channel model using the least mean square algorithm and estimating the co-channel states by means of an unsupervised clustering scheme. Simulation involving a binary signal constellation is used to compare both the theoretical and adaptive performance of this Bayesian decision feedback equaliser with those of the maximum likelihood sequence estimator. The results obtained indicate that, in the presence of severe co-channel interference, the Bayesian decision feedback equaliser employing the proposed simple scheme to compensate co-channel interference can outperform the maximum likelihood sequence estimator that only treats co-channel interference as an additional coloured noise.