Song Wentao

Genetic algorithm based multiuser detection for CDMA systems

Genetic algorithms (GAs) are applied to suboptimal joint multiuser detection in bit-synchronous code division multiple access (CDMA) systems, whereas GAs have been employed for solving many optimization problems in numerous fields. The major advantage of the proposed GAs multiuser detector is that it provides a good compromise between the performance and computational complexity, that in contrast to the case of the optimal maximum-likelihood multiuser detector has a computational complexity that is exponentially increasing with the number of users. Computer simulation results show that GAs multiuser detector has powerful properties both in bit error probability and near-far resistance comparison of the conventional detector and some other multiuser detector.

Neural networks based parallel Viterbi decoder by hybrid design

A hybrid scheme integrating analog and digital methods is presented to design a Viterbi decoder based on neural networks. Due to its fully parallel architecture, neural networks based Viterbi decoder is significantly faster than the purely digital decoder. The fully parallel structure is obtained by implementing the branch metric calculation and add-compare-select (ACS) using the neural networks while the register exchange using parallel digital circuits. The hybrid Viterbi decoder is more suitable for VLSI implementation.

AGC and IF amplifier circuits design

Effective and fast analog AGC improves receiver operational performance. According to the theory of AGC, this paper develops AGC and IF amplifier circuits that can be applied in a receiver of a base station in 3G (third generation) mobile communications. The paper introduces the structure figure of the design circuit, and the unit circuit design. Experimental results are given. The experimental results have shown that the designed AGC and IF amplifier circuits in the receiver can satisfy the technical specification of the receiver of the base station in 3G mobile communications systems.

VLSI design and implementation of WCDMA channel decoder

We present a memory and driving clock efficient design scheme to achieve WCDMA high-speed channel decoder module (including a de-interleaving and de-multiplexing processing unit, a turbo decoder, a Viterbi decoder, etc.) on a single Xilinx XVC1000E FPGA chip. Using a modified MAP algorithm, say parallel sliding window logarithmic maximum a posteriori (PSW-log-MAP), the on-chip turbo decoder can decode an information bit by only an average of two clocks per iteration. On the other hand, a high-parallel pipeline Viterbi algorithm is adopted to realize the 256-state convolutional code decoding. The final decoder with an 8/sub i/ chip-clock (30.72 MHz) driving can concurrently process a data rate up to 2.5/spl sim/5 Mbps of turbo coded sequences and a data rate over 400 kbps of convolutional codes under a reasonable BER performance. If the de-interleaving and de-multiplexing processing unit is excluded, there is no external memory needed. Test results show that the decoding performance of these two type channel decoders is only 0.2-0.25 dB or less loss compared to float simulations.

3D positioning system using magnetic field

A method to determine the 3D position using harmless magnetic field is proposed as a solution to the problem of positioning the medical instruments such as endoscope inside the patients body. The intension expression of magnetic induction produced by a single loop coil is derived and the mathematical model of 3D positioning system with a mode of magnetic field is established. Then a practical 3D positioning system is developed to implement real-time 6 degrees of freedom measurement by a 3D magnetic sensor attached to the endoscope and 5 degrees measurement by a single magnetic sensor.

A simple channel estimator for space-time coded OFDM systems in rapid fading channels"

A simple channel estimator for space-time coded orthogonal frequency division multiplexing (OFDM) systems in rapid fading channels is proposed. The channels at the training bauds are estimated using the EM (expectation-maximization) algorithm, while the channels at the data bauds are estimated based on the method for modelling the time-varying channel as the linear combination of several time-invariant “Doppler channels”. Computer simulations showed that this estimator outperforms the decision-directed tracking in rapid fading channels and that the performance of this method can be improved by iteration.

Channel estimation for MIMO-OFDM systems in rapid fading channels

A channel estimation approach for orthogonal frequency division multiplexing with multiple-input and multiple-output (MIMO-OFDM) in rapid fading channels is proposed. This approach combines the advantages of an optimal training sequence based least-square (OLS) algorithm and an expectation-maximization (EM) algorithm. The channels at the training blocks are estimated using an estimator based on the OLS algorithm. To compensate for the fast Rayleigh fading at the data blocks, a time domain based Gaussian interpolation filter is presented. Furthermore, an EM algorithm is introduced to improve the performance of channel estimation by a few iterations. Simulations show that this channel estimation approach can effectively track rapid channel variation.

LDPC Based Differential Unitary space-frequency Coding for MIMO-OFDM Systems

This paper proposes a novel LDPC based differential unitary space-frequency coding (DUSFC) scheme for MIMO-OFDM systems when neither the transmitter nor the receiver has access to the channel state information (CSI). The new DUSFC strategy basically consists of coding across transmit antennas and OFDM tones simultaneously as well as differential modulation in the time-domain. It can fully exploit the inherent advantages provided by the multipath fading channels, resulting in a high degree of diversity. The state-of-the-art low-density parity-check (LDPC) codes are concatenated with our DUSFC as channel coding to improve the bit error rate (BER) performance considerably. Owing to the maximum multipath diversity and large coding advantages, LDPC-DUSFC strongly outperforms the differential unitary space-time coded OFDM techniques recently proposed in literature. The corresponding iterative decoding algorithm without channel estimation is finally provided to offer significant performance gain. Simulation results illustrate the merits of the proposed scheme.

An improved channel estimation with multipath search for MIMO-OFDM systems

This paper addresses the problem of channel estimation for broadband MIMO-OFDM systems. An improved channel estimator with multipath time delay detection and channel gain estimation is proposed. In the algorithm, we used the correlation of the channel taps and a well-designed adjustment scheme to increase the accuracy of the time delay detection. The most attractive advantage is that the complicated matrix calculation is replaced by search steps which can acquire the channel order and estimate the channel parameters without significantly increasing the complexity of the system. Computer simulation showed that the proposed algorithm can track the time delays adaptively and, consequently, improve the channel estimation performance.

Efficient coding schemes with power allocation using space-time-frequency spreading

An efficient space-time-frequency (STF) coding strategy for multi-input multi-output orthogonal frequency division multiplexing (MIMO-OFDM) systems is presented for high bit rate data transmission over frequency selective fading channels. The proposed scheme is a new approach to space-time-frequency coded OFDM (COFDM) that combines OFDM with space-time coding, linear precoding and adaptive power allocation to provide higher quality of transmission in terms of the bit error rate performance and power efficiency. In addition to exploiting the maximum diversity gain in frequency, time and space, the proposed scheme enjoys high coding advantages and low-complexity decoding. The significant performance improvement of our design is confirmed by corroborating numerical simulations.