Research On The Key Combining Techniques Based On Collaborative Reference For Multi-antenna Signals Reception

Due to the development of the wireless communication, great importance has been attached to the reception of low signal-to-noise ratio(SNR) signals. As the effective reception of weak signal can’t be conducted by only on antenna, one of the solutions is the multi-antenna combining technique, which improves the signal-to-noise ratio(SNR) of the received signal. In the traditional scheme in which one signal is taken as a fixed reference, if the reference signal has lower SNR or experiences a more complex channel, the estimation will deteriorate and thus decreases the combination performance. Therefore, it is necessary to optimize the reference signal for the combination performance improvement.This dissertation focuses on optimization of the reference signal and researches on the alignment technique of multi-antenna signal time-delay under additive white Gaussian noise(AWGN) channels and the combination technique of multi-antenna signal under frequency selective fading channels. The contributions obtained in this dissertation can be summarized as follows.1. Aiming at the optimization of the reference signal in the alignment of multi-antenna signal time-delay under AWGN channels, this paper firstly derives the explicit formulation of the combining performance affected by the time-delay alignment error by using the matched filter bound for an antenna array. Analysis shows that the larger number of the antennas, the higher performance of the time-delay alignment is required in order to cut down the combined output SNR gain loss. Then, this thesis introduces a collaborative reference signal which is the combined output rather than one antenna signal, and proposes a partial compensation based time-delay alignment algorithm(partial compensation based algorithm) based on the analysis of the convergence of the output combined signal referred algorithm. The constant convergence of the algorithm is proved. Theoretical analysis and simulations show that the proposed algorithm can avoid time-delay alternating variation phenomenon in the output combined signal referred algorithm and exhibit stronger robustness.2. Aiming at the optimization of the compensation factors in a non-uniform antenna array, this paper optimizes the objective function of convergence speed and sets up a mathematical model based on the further research of the partial compensation based algorithm. The explicit formulations of the optimal compensation factors and time-delay center are both derived. Theoretical derivation indicates that the optimal compensation factor of one signal is determined by the SNRs of all the signals, when as the SNR of one signal increases, the optimal compensation factor decreases. The time-delay center is the weighted average of the initial time-delay of each signal, where the weight is the SNR. Experiment results show that the proposed technique outperforms the combined output referred technique in time-delay alignment accuracy and convergence speed.3. Aiming at the optimization of the reference signal in the combination of multi-antenna signals under frequency selective fading channels, this paper firstly compares the space diversity combining and equalizer(SDE) with the frequency domain max ratio combine(FMRC) algorithm and proves the equivalence in the equalizer output mean square error(MSE) performance of the two algorithms. Then, the effect of the phase frequency response of the fixed reference channel on the blind equalizer performance of the combined output signal is analyzed. This paper avoids the selection of fixed reference signal and selects the average of all the channels’ phase frequency responses as the reference in order to avoid the performance loss of blind equalizer. A two-signal-combination based and average phase referred multi-antenna combination algorithm is proposed by taking the advantage of the fact that no phase ambiguity exists between channels in two signals combination. The complexity of the algorithm is proportional to the number of the antenna and the effectiveness is testified through experiments.4. An 8-antenna signal combination system is designed and implemented and its high performance is proved by practical testing results. Firstly, an overall plan of 8-antenna signal combination system is given according to the technical requirements. Then, software and hardware implementation is conducted aiming at the application. It is mainly composed of the implement of combiner control software, the implement of signal processing algorithms on DSP such as partial compensation based algorithm and the implement of the interface between DSP and FPGA. Finally, the high performance of the combination system is proved by practical testing results.