Improved low complexity hybrid turbo codes and union bound analysis
ABSTRACT Turbo convolutional codes (TCC) are excellent error correcting codes for wireless channels. However, TCC decoders require large decoding complexity. Moreover, complexity of TCC decoder does not reduce even if puncturing is used to change the coding rate. Modified turbo codes require lower decoding complexity than TCC as they use multiple concatenations of simple block codes and convolutional codes. Recently, a class of modified turbo codes called low complexity hybrid turbo codes (LCHTC) and improved low complexity hybrid turbo codes (ILCHTC) have been proposed. It has been shown that LCHTC and ILCHTC achieve bit error rate (BER) which is comparable to TCC and have much lower decoding complexity. Simulation results show that BER performance of ILCHTC is better than that of LCHTC. Rate-1/3 ILCHTC achieve BER of 10-5 at bit energy-to-noise ratio ( E b/ N 0) of 1.9 dB, which is 0.4 dB higher than E b/ N 0 for TCC adopted by third generation partnership project (3GPP). Moreover, ILCHTC and LCHTC decoders require half the number of computations as compared to those required for TCC decoder. In this study, union-bound analysis of ILCHTC is presented to investigate BER performance<;10-6. For large interleaver lengths, analysis of theoretical union bound requires numerous computations. Therefore approximate analysis of union bound is derived from theoretical union bound. It is shown that the analysis of approximate union bound achieves reasonable accuracy. Moreover, approximate union bound can be evaluated with significantly less computational complexity than the theoretical union bound.
- SourceAvailable from: edu.hk[show abstract] [hide abstract]
ABSTRACT: This letter presents an improved two-state turbo single-parity-check code for applications with rates around 1/3. The new code is compared with the (15,13)<sub>8</sub> turbo code used in the Third Generation Partnership Project. Simulation results demonstrate that the proposed code can achieve similar performance at reduced decoding complexity.IEEE Transactions on Communications 09/2004; · 1.75 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: This paper is concerned with a family of concatenated tree (CT) codes. CT codes are special low-density parity check (LDPC) codes consisting of several trees with large spans. They can also be regarded as special turbo codes with hybrid recursive/nonrecursive parts and multiple constituent codes. CT codes are decodable by the belief-propagation algorithm. They combine many advantages of LDPC and turbo codes, such as low decoding cost, fast convergence speed, and good performanceIEEE Transactions on Information Theory 03/2001; · 2.62 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: This paper provides a description of the turbo code used by the UMTS third-generation cellular standard, as standardized by the Third-Generation Partnership Project (3GPP), and proposes an efficient decoder suitable for insertion into software-defined radio architectures or for use in computer simulations. Because the decoder is implemented in software, rather than hardware, single-precision floating-point arithmetic is assumed and a variable number of decoder iterations is not only possible but desirable. Three twists on the well-known log-MAP decoding algorithm are proposed: (1) a linear approximation of the correction function used by the max* operator, which reduces complexity with only a negligible loss in BER performance; (2) a method for normalizing the backward recursion that yields a 12.5% savings in memory usage; and (3) a simple method for halting the decoder iterations based only on the log-likelihood ratios.International Journal of Wireless Information Networks 09/2001; 8(4):203-215.