| Interferences exist in wireless communications widely, such as celluar mobile communications, wireless local area networks, and cognitive radio. With the development of wireless communications, the problem of interference becomes more important, and it is one of the key obstacles for improving wireless spectral efficiency.The existing transformation domain interference suppression techniques generally separate the desired signal and interference in a fixed transformation domain. But the fixed transformation domain interference suppressions only exploit the common feature of a kind of interference, and they cannot improve performance by the statistical information of a specific interference. In this thesis, interference suppression algorithms which explore the statistical information of narrowband interference(NBI) are investigated. The main contributions of the thesis are listed as follows:First, a variable window eigen domain interference suppression algorithm in single input single output(SISO) systems is proposed. The NBI and the desired signal have different power distribution when transforming the direct sequence spread spectrum(DSSS) signal into the eigen domain. The proposed algorithm rejects the interference through weighting on each eigen domain element. Besides, the study discusses the impacts of processing window length on performance, convergence speed, and computational complexity. Because of its simple receiver structure and blind adaptive implementation, the algorithm is practical. Simulations show, when the processing gain is 4, the interference is single tone, the interference to signal ratio is 50 d B, and the BER is 0.01, the proposed algorithm outperforms the code-aided technique by about 3 d B.Second, an eigen transformation domain interference rejection combining algorithm in multiple input multiple output(MIMO) systems is proposed. For received signals from multiple antennas, the NBI and the desired signal have different power distribution in the eigen domain and different channel matrices in the spatial domain. The eigen domain interference rejection combining can rejects the interference in the elements with heavy interference and exploits the diversity gain in the elements with no interference. Compared with the traditional interference rejection combining, the eigen domain interference rejection combining can achieve higher SINR and diversity orders. Simulations show, in the 1×2 SIMO systems, when the normalized bandwidth of the NBI is 1/8, the interference to signal ratio is 10 d B, and the BER is 0.01, the proposed algorithm outperforms the interference rejection combining by about 3 d B.Third, the transformation domain interference suppression with precode is proposed. The study assumes the statistical information of the NBI is known at the transmitter, through joint processing of the transmitter and receiver, the channel is divided into multiple parallel subchannels. As the power of NBI is not uniform distributed in eigen domain, the interference level of the subchannels are different from each other, thus the transmitter can align different space time code on each subchannel to increase the capacity or enhance the reliablity. Analysis points out that the statistical information of NBI at the transmitter can increase the capacity. Simulations show, in the 4×4 MIMO systems, when the normalized bandwidth of the NBI is 1/8, the interference to signal ratio is 10 d B, and the capacity is 30 bits/s/Hz, the proposed algorithm outperforms the interference rejection combining by about 4 d B.Finally, eigen domain interference suppression for time synchronization is proposed. The proposed algorithm exploits the power distriburion of the NBI, and meets the special requirement of the time synchronization. Simulations and analysis show, for both single antenna and multiple antennas, the interference is suppressed by the proposed algorithm successfully, the output SINR is increased, and the performance of probability of detection(PD) is improved. For instance, with single receive antenna, when the normalized bandwidth of the NBI is 1/8, the interference to signal ratio is 10 d B, and the PD is 0.9, the proposed algorithm outperforms the frequency domain interference suppression by about 2 d B, and the performance gap increases with the increase of the interference to signal ratio.This thesis investigates the transformation domain interference suppression algorithms which exploit the statistical information of the NBI. The research results in this thesis can be applied in the systems such as wireless sensor networks, wireless local area networks(W-LAN), and next-generation celluar mobile communications.
|
PDF Full Text Request