The effect of the phase noise on the performance of bandwidth-efficient coded modulation is studied. To this end, the average mutual information (AMI) for specific constellations such as 8-phase-shift keying and 16-quadrature amplitude modulation is calculated in the presence of carrier phase error caused by imperfect carrier tracking over an additive white Gaussian noise channel. The AMI not only quantifies the effect of the phase noise from an information-theoretic viewpoint, but also serves as an estimate for a permissible amount of the phase noise for a given signal-to-noise ratio. The bit-error rate (BER) performance of a near-optimal turbo trellis-coded modulation scheme is then investigated over such a channel. For this purpose, an optimal branch metric which best fits the channel characteristics is derived. Furthermore, simple branch metrics (referred to as suboptimal, simplified, and Gaussian metrics) are derived, which may offer the tradeoff between BER performance and computational complexity. Numerical analysis shows that a near-optimal coded-modulation scheme renders a transmission system more robust against phase noise than is the case with a conventional trellis-coded modulation scheme.

In this paper we present a new iterative decoding algorithm of the high-rate turbo codes which employ high-rate convolutional codes as component codes. This algorithm constructs a syndrome trellis using the parity check matrix of the code, and produces soft-output of each bit of a codeword. The algorithm is named "turbo syndrome decoding" (TSD). In the trellis diagram, only two paths merge into a node. It therefore leads to the significant reduction in both computational complexity of add-compare-select (ACS) operations and storage requirement for the soft-output to the subsequent decoder as compared with the classical decoding of the high-rate turbo codes. Nevertheless, the TSD algorithm has the same capability as the classical one in a sense that the most probable path is selected.

This paper presents a new reduction method of both the computational complexity and storage requirements in the nonpunctured turbo decoding. The idea of the proposed algorithm is based on the computational complexity reduction, which is attendant upon the process of add-compare-select in each stage of iterative decoding. Therefore, the proposed method is completely different from all the other reduction algorithms presented before now. We compare the nonpunctured turbo code using the proposed reduction method with the punctured turbo code in terms of performance-complexity tradeoff. Simulation results show that the proposed approach provides an alternative encoding/decoding scheme of the classical punctured turbo coding for the application to high-data-rate wireless communications system.