Viterbi decoding of convolutional codes with reliability information for a noncoherent RAKE-receiver in a CDMA-environment

The generation and the use of reliability information for the decoding process of convolutionally encoded data in the context of a CDMA-environment is addressed. Special emphasis lies on the noncoherent detection of M-ary orthogonal symbols with a RAKE-receiver. Although symbol by symbol decisions are carried out, it is possible to characterize the reliability of the individual bits with specific soft information. This can be exploited successfully by a Viterbi decoder. A significant improvement of the bit error performance is achieved, even on low signal to noise conditions. The paper starts with the derivation of the maximum likelihood (ML) criteria for the detection of binary data on the AWGN channel and the binary symmetric channel (BSC).