Abstract: A method is described for convolutionally encoding and decoding data (voice coded data), organized into (35-bit, 20 msec) frames, where the data is encoded by a finite-state data encoder (RSC coder 216) for transmission over a data channel (18). The RSC encoder avoids the need for termination bits associated with each N-bit frame of data, so that the number of bits associated with each encoded frame is reduced, and the throughput of the channel can be increased. The method according to the invention includes storing the first M bits of each frame. Once they are stored, they are loaded in parallel into the M stages of the encoder, thereby deleting residual states from the preceding frame. The remaining (N−M) bits are then applied to the encoder, causing it to produce the convolutional code. After the last of the (N−M) bits are encoded, the M stored bits are summed with feedback from the encoder, and reapplied to the encoder. The ending state of the encoder is thus equal to the starting state.

Abstract: A method for decoding (voice)data where the data is encoded by a finite-state data encoder (216) for transmission over a data channel (18), includes the step of procuring a table of joint statistics representing the probability of occurrence, in a frame of source data of each of the bits of the frame. The joint statistics may be determined ahead of time. The method includes the step of calculating intermediate gamma signals in response to the decoded data of the preceding frame and the joint statistics. The joint statistics source distribution signals represent the likelihood that, for a given logic level of the preceding bit, the “current” bit takes on a particular state; for uncorrelated bits, this value is 0.5. State probability signals and transition probability signals are generated from the gamma signals.

Abstract: A feedback control for a turbo decoder controls the feedback between component decoders by modifying updated a priori probabilities calculated by one component decoder and used as inputs to another component decoder during the decoding process, resulting in a significant performance advantage. A feedback control switch selects either previously estimated a posteriori probabilities, modifications of these values, or neutral values as a priori probabilities utilized by the next component decoder.

Abstract: A decoder of digital data signals for a message sequence of N symbols from a data source, encoded with a tail-biting convolutional code, and for correcting all combinations of up to a preselected number e of bit errors, operates upon the sequentially-received collection of consecutive-encoded tail-biting data symbols and selects an unmerged decision depth LU(e), a forward decision depth LF(e) and then a decoder decision depth L, before executing in a processor of the Viterbi type, which has a decision depth of L trellis levels, a decoding algorithm for a total of (N+LF(e)+LU(e)) trellis stages to produce a decoded data bit output sequence representative of the N transmitted source digital symbols.

Abstract: A circular MAP decoder for error-correcting trellis codes that employ tail biting produces soft-decision outputs provides an estimate of the probabilities of the states in the first stage of the trellis, which probabilities replace the a priori knowledge of the starting state in a conventional MAP decoder. The circular MAP decoder provides the initial state probability distribution in either of two ways. The first involves a solution to an eigenvalue problem for which the resulting eigenvector is the desired initial state probability distribution; with knowledge of the starting state, the circular MAP decoder performs the rest of the decoding according to the MAP decoding algorithm. The second is based on a recursion for which the iterations converge to a starting state distribution. After sufficient iterations, a state on the circular sequence of states is known with high probability, and the circular MAP decoder performs the rest of the decoding according to the MAP decoding algorithm.

Abstract: A parallel concatenated convolutional coding scheme utilizes tail-biting nonrecursive systematic convolutional codes. The associated decoder iteratively utilizes circular maximum a posteriori decoding to produce hard and soft decision outputs. This encoding/decoding system results in improved error-correction performance for short messages.