| In the Evolved Universal Terrestrial Radio Access (E-UTRA) downlink, OFDM based radio access was adopted because of its inherent immunity to multipath interference, which is generated in a frequency-selective fading channel. In OFDM-based radio access for the E-UTRA downlink, effective channel encoding is essential to decrease the required received signal-to-interference plus noise power ratio associated with interleaving in the frequency domain. The current working assumption in the study item phase is Turbo coding following a proven practice for high data rate provisioning in W-CDMA and High-Speed Downlink Packet Access. However, its high decoding complexity hold large part in baseband signal processing at a receiver. Thus, employing a low-density parity check (LDPC) code has drawn attention particularly for high data-rate applications because it has a decreased level of decoding complexity compared to the Turbo decoder. Many options for LDPC codes and decoding schemes were proposed and investigated. However, most of the reports assumed an Additive White Gaussian Noise channel with perfect diversity techniques. As a result, it is very difficult to answer the question whether the LDPC code has a distinct advantage over the Turbo code in the Evolved UTRA downlink using OFDM, from the viewpoints of the achievable performance gain and decoding complexity.
This paper presents comprehensive comparisons of the achievable packet error rate (PER) performance and computational complexity for a decoder employing the Turbo code and that employing the LDPC code considering essential key techniques such as hybrid ARQ with packet combining, channel-dependent scheduling, and antenna diversity reception for various modulation schemes. For the LDPC codes, the Rate compatible /Quasi Cyclic LDPC code, Concatenated ZigZag code, and Turbo Single Parity Check code are used, which have high affinity to the effective Incremental redundancy scheme, which takes advantage of the puncturing of a low coding rate. More specifically, we investigate the PER and throughput performance levels employing Incremental redundancy as a function of the coding rate, the information length, and the number of decoding iterations. Furthermore, we evaluate the computational complexity of a decoder on the condition that the performance is almost saturated in order to establish the optimum channel code for a shared data channel in the Evolved UTRA downlink.
|
| Naoto Ohkubo received the B.E. and M.E. degrees from Tokyo University of Science, Chiba, Japan, in 2002 and 2004, respectively. In 2004, he joined NTT DoCoMo, Inc.
Since joining NTT DoCoMo, Inc., he has been engaged in the research of radio transmission technologies for the third and fourth generation mobile communication systems. |