Tian Hongxin

A new method to design CDMA spreading sequences

It is well-known that multi-value CDMA spreading codes can be designed by means of a pair of mirror multi-rate filter banks based on some optimizing criterion. This paper indicates that there exists a theoretical bound in the performance of its circulate correlation property, which is given by an explicit expression. Based on this analysis, a criterion of maximizing entropy is proposed to design such codes. A computer simulation result suggests that the resulting codes outperform the conventional binary balance Gold codes for an asynchronous CDMA system.

A fast codebook design algorithm for vector quantization

This paper presents a fast codebook design algorithm for vector quantization, which eliminates distortion calculations by using inequality between the distortion and the means of two vectors as well as the partial distortion theorem during both the initial codebook splitting and LBG iteration procedures. Moreover, the partition information of vectors in the previous iteration is used in the repartition of the training set to make the employing of the two above eliminating rules more efficiently. This algorithm eliminates about 90% of unnecessary distortion calculations and can produce the same codebook as the LBG algorithm with only a little additional memory, thus outperforming other fast algorithms that always need much more additional memory which is impractical for big codebook cases.

Single tone jamming elimination scheme of multicarrier spread spectrum system

The proposed scheme is based on Discrete Fourier Transform (DFT) domain processing. The key technology of this scheme is jamming parameters’ accurate estimation and jamming reconstruction. Compared with the “threshold exciser” scheme the proposed scheme can eliminate more jamming energy on the whole frequency band with the minimum loss of useful signal energy. As shown in the research and simulation, the proposed scheme is much better than the “threshold exciser” scheme, especially in the case of high power jamming whereas the “threshold exciser” scheme might be invalid.The proposed scheme is based on Discrete Fourier Transform (DFT) domain processing. The key technology of this scheme is jamming parameters’ accurate estimation and jamming reconstruction. Compared with the “threshold exciser” scheme the proposed scheme can eliminate more jamming energy on the whole frequency band with the minimum loss of useful signal energy. As shown in the research and simulation, the proposed scheme is much better than the “threshold exciser” scheme, especially in the case of high power jamming whereas the “threshold exciser” scheme might be invalid.

Demultiplexing and demodulating 32-channel FDM/QPSK signals with a single DSP

This paper proposes a scheme of demultiplexing and demodulating 32-channel FDM/QPSK signals based on a DSP chip of TMS320-C6201. The results of simulating the scheme with a TMS320-C6201 developing kit suggest that the scheme is reasonable and feasible. Implementation of the scheme could provide a low-complexity solution for multi-carrier demodulation in satellite communication.

The study of anm-ary MC-CDMA system based on cyclic spreading codes & pre-equalization

A novel fast despreading scheme for M-ary Multi-Carrier Code-Division Multiple Access (MC-CDMA) system is proposed based on cyclic spreading codes and pre-equalizer. In the transmitter, the M spreading codes of each user are generated by circularly shifting the prototype spreading code. A feedback pre-equalizer is employed to process the M-ary MC-CDMA signal before transmitted. The received signal is multiplied by the Inverse Discrete Fourier Transform (IDFT) result of the mirror image code of the prototype spreading code, and then demodulated by Orthogonal Frequency-Division Multiplexing (OFDM) demodulator. Compared with the conventional M-ary MC-CDMA receiver, the proposed scheme increases bandwidth efficiency, meanwhile, it achieves M-ary despread spectrum and multi-carrier demodulation, which reduces computation complexity remarkably.

A novel statistical language modelling method for continuous Chinese speech recognition

Statistical language models can play an important role in continuous speech recognition, but their performance is often unstable because of the training data sparsity. This paper proposes a statistical language modeling method, where the contribution of the language model is limited by the acoustic matching result and the N-gram probability distribution is modified referring to the length of the silence duration between adjacent syllables. Besides, the paper proposes a powerful single-state hidden Markov model (HMM) to model various kinds of silence segments.