Improving frame synchronization when using convolutional codes

Presents a method of improving the performance of a frame synchronizer when the data frames are convolutionally encoded and terminated. Termination of the data helps to ensure correct decoding and reduces the effective rate of the code, especially for short data sequences. The synchronizer makes use of the symbols transmitted for termination in order to increase the probability of correct frame synchronization. The author derives an optimum decision rule and several suboptimal algorithms that considerably reduce computation and complexity. Improvement of several dB in frame sync performance was found by simulations. The technique can be applied to "one shot" (e.g. packet) or continuous transmission applications. Furthermore, no changes need to be made to the frame structure, only modifications of the receiver.<<ETX>>