An Upper Bound on Frame Error Rate for Generalized Concatenated Convolutional Codes

Introduction In order to achieve highly reliable communication over an additive white Gaussian noise (AWGN) channel under a restriction of computational complexity of decoding , one of the promising approaches is concatenated coding. The concatenated code consists of two codes: an inner and an outer codes combined in cascaded manner. A short block code or trellis-code with a small number of states is used for the inner code and an algebraic block code, such as Reed-Solomon(RS) code, is used for the outer code. Forney[1] first proposed the concatenated code and presented the existence of concatenated codes of rate less than the capacity which give an arbitrary small error probability. His work were motivated from a purely theoretical point of view, however, today's state-of-theart technology enables us to exploit the concatenated code for practical use. There are some decoding strategis for such a concatenated coded system. The combination of the softoutput Vite