Communications phase synchronization using the Adaptive Network Fuzzy Inference System (ANFIS)

The use of soft computing (SC) techniques in communications phase synchronization provides an alternative to analytical or hard computing methods. A review of current literature indicates that the application of soft computing to communications phase synchronization is in its infancy. In this work we derive new lower bounds on synchronizer performance, and demonstrate the use of a fuzzy logic-based approach in phase synchronization for a multiple-phase-shift-keying (M-PSK) communications system.; Maximum Likelihood (ML) Estimation theory provides a basis for the most widely used phase estimation schemes. In this context, a review of ML statistical communication phase estimation theory and ad-hoc techniques for M-PSK modulation is presented first. New lower bounds on phase estimator performance are derived. These new Cramer-Rao Bounds (CRBs) provide the lower bound on which to judge the performance of any estimator for M-PSK modulation with a random phase model. Computational complexities and non-linear systems that result from statistical estimation theory for M-PSK systems are also addressed.; Next the fundamentals of the principle constituents of soft computing, namely fuzzy logic and neural networks, are reviewed. The fundamentals of a hybrid soft computing technique known as Adaptive Neural Fuzzy Inference System (ANFIS) are discussed. Recent advances in communications that utilize soft computing approaches to phase synchronization are reviewed.; Finally, the ANFIS approach is applied to phase synchronization for M-PSK. It is shown that the ANFIS approach overcomes many of the limitations of the Mamdani and statistical approaches. A comparison of statistical and ANFIS estimator performance is presented.