Improved Multilevel Coding Schemes

Download or read book Improved Multilevel Coding Schemes written by Muhammad Ahsan Naim and published by . This book was released on 2015 with total page 138 pages. Available in PDF, EPUB and Kindle. Book excerpt: Multilevel coding (MLC) is a powerful coded modulation technique for band limited channels. Various types of error correction codes can be employed at different coding levels of MLC schemes. In general, MLC schemes can be decoded either by a low complexity multistage decoding (MSD) or with complex soft iterative decoding which is based on Maximum-a-Posteriori (MAP) decoding. This dissertation makes the following improvements in MLC: (a) introduces a novel improved hard iterative decoding (IHID) technique for MLC schemes to improve upon the existing MSD decoding, (b) extends the IHID technique for soft decoding by combining with List Viterbi Decoding (LVD) to introduce a novel LV-IHID decoding technique for MLC schemes, and (c) presents a novel building block approach for systematically constructing constellations for MLC schemes. In addition, a novel list sequence MAP iterative decoding (LS-MAP-ID) algorithm is developed for coded schemes that employ hybrid automatic repeat request with forward error correction (ARQ-FEC). It is known that MSD suffers from performance degradation especially at low signal to noise ratio (SNR) due to multiple representations of signal points at lower coding levels which leads to error propagation to the higher coding levels in presence of channel variations. In this study, a novel improved hard iterative decoding algorithm (IHID) that mitigates the problems associated with MSD at comparable complexity is presented. Performance of IHID algorithm is further improved at low to moderate SNR by employing the LVD and a new decoding algorithm List Viterbi¿IHID (LV-IHID) is developed. A novel list sequence MAP iterative decoding (LS-MAP-ID) algorithm is developed in this study that incorporates hybrid automatic repeat request with forward error correction (ARQ-FEC) which improves the decoding performance of soft iterative decoding schemes by employing List decoding on soft iterative decoding of MLC schemes and ARQ criterion when frames cannot be resolved by the decoder. Also, a novel encoder design for MLC schemes based on building block approach is developed that not only improves power efficiency but also presents an elegant design strategy especially for higher order constellation design for MLC.