Design of Turbo Trellis Coding Modulation Scheme of Rate 4/9 for Rician Fading Channel

When the fading channels are encountered during data communication, errors are likely to occur at the receiving end due to multipath propagation. Researchers have been consistently striving to develop Error Correction Schemes that can effectively handle these errors and ensure error-free data reception at the receiver end. Of particular interest are the Forward Error Correction Schemes that can be implemented at the transmitter end itself. However, the implementation of error correction coding through these schemes incurs additional costs in terms of bandwidth expansion, as extra bits need to be added to facilitate error correction. Fortunately, there exists one coding scheme called Trellis Coded Modulation (TCM), which addresses this issue. TCM selects a modulation scheme based on the rate of the convolutional coding scheme. However, this coding technique has limitations in correcting the number of errors, leading to the development of Turbo Coding. This scheme utilizes two coders at the transmitter, arranged in either serial or parallel configuration, and a suitable decoder at the receiver. A design of Turbo Coding scheme has been presented in this paper, that employs convolutional coders having rate 2/3, in a serially concatenated configuration, providing an effective rate of 4/9. This turbo coding scheme is then applied to TCM scheme in order to preserve the bandwidth. Therefore, if using the convolutional coding scheme of rate 2/3, the modulation scheme is 8-QAM and in order to preserve bandwidth after coding, using the Turbo coding scheme of rate 4/9, then the modulation scheme will be 512-QAM. The simulations have been conducted in MATLAB and the error correcting capabilities of the designed scheme in comparison with convolutional coding scheme using the constituent convolutional encoder have also been compared. It has been observed that in the Rician fading channel conditions, the Turbo Trellis Coding Modulation Scheme provides approximately 5 dB gain compared to the convolutional coding scheme.