Feng Guang-zeng

PERFORMANCE ANALYSIS OF CHANNEL ESTIMATION FOR LDPC-CODED OFDM SYSTEM IN MULTIPATH FADING CHANNEL

In this paper, the channel estimation techniques for Orthogonal Frequency Division Multiplexing (OFDM) systems based on pilot arrangement are studied and we apply Low Density Parity Check (LDPC) codes to the system of IEEE 802.16a with OFDM modulation. First investigated is the influence of channel estimation schemes on LDPC-code based OFDM system in static and multipath fading channels. According to the different propagation environments in 802.16a system, a dynamic channel estimation scheme is proposed. A good irregular LDPC code is designed with code rate of 1/2 and code length of 1200. Simulation results show that the performance of LDPC coded OFDM system proposed in this paper is better than that of the convolution Turbo coded OFDM system proposed in IEEE standard 802.16a.

A subspace constrained LSCM algorithm

In this paper, a subspace constrained least-squares constant modulus algorithm (LSCMA) is proposed for blind adaptive multiuser detection. The new algorithm has been derived by integrating projection approximate subspace tracking with deflation (PASTd) algorithm, singular value decomposition (SVD) technique and LSCMA. This proposed algorithm reduces computational complexity remarkably compared with the traditional eigenvalue decomposition (ED) subspace algorithm. Simulation results show that the proposed algorithm outperforms LSCMA in terms of bit error rate (BER) and convergence rate, especially when the signal-to-noise ratio (SNR) is low. Simulation results also show that the proposed algorithm achieves a comparable performance as the traditional ED subspace algorithm based LSCMA in terms of convergence rate, tracking ability and BER performance, but with a much lower complexity.

A Novel STBC Scheme in MIMO CDMA System

A novel STBC scheme in MIMO CDMA system is presented. In the proposed scheme, the input symbols are split in blocks, each block has M symbols (M is the number of transmitting antennas), the M symbols are circularly encoded into M groups (each group has M cyclic symbols), the encoded M × M symbols are transmitted in Msymbol durations through Mantennas by multiplying M spread-codes respectively. The proposed STBC scheme has full coding rate (rate 1) and full diversity order but the decoding method is as simple as the conventional STBC introduced by Alamouti [1].

A Novel Beam-forming Algorithm in Antenna Array MC-CDMA System

A novel beam-forming algorithm in antenna array MC-CDMA system is proposed in this paper. First, the beam-formed signals on every sub-carrier are combined by using ORC, EGC or MRC, and then the decision of the combined signal and the estimated channel parameters are used to reconstruct the signal on every sub-carrier, finally the difference between the reconstructed signal and the output signal of the beam-former is used to control the coefficients of the beam-former. The bit error probability of the proposed algorithm is also analyzed in this paper

Turbo trellis code modulation using MQAM

The paper presents a kind of reasonable structure for implementing MQAM T-TCM based on the principles of turbo codes and TCM for the first time. It can also be expanded to PSK T-TCM system, and the corresponding decoding algorithm is derived. By computer simulation, its performance is analyzed. The results show that T-TCM takes the advantages of turbo codes and TCM technology, and is a kind of bandwidth-efficient coded-modulation technique obtaining high coding gain. So, in the future, T-TCM would be applied in many fields.

A multisolution learning algorithm for fuzzy rules

This paper presents a novel learning algorithm for the approximate reasoning-based Fuzzy Adaptive Mapping Network (FAMN). In the proposed algorithm, we use a serial genetic algorithm to adjust the membership function of the rules. In particular, unlike other adaptive methods, the learning is achieved by incorporating the idea of multiple resolution. The tuning is first implemented with few rules at the coarsest resolution of input/output variables, then with many rules at the higher resolution until the training precision required is obtained. The 2D sine function net is constructed as an illustrative example.