Blind Equalization Algorithm Of Shortwave Channel Based On Multi-channel Receiving

In shortwave communication system, fading effects of multipath propagation alwaysintroduce Inter-symbol Interference (ISI) in the receiver and affect communication qualityseriously. As we know, diversity is a powerful technology combating fading effects. This thesisfocuses on shortwave channel blind equalization algorithm based on muti-channel receiving,which is under the condition of using existing receiver equipment and receiver position or usingrandomly-distributed receiving array. In non-cooperative conditions, the algorithm employsmulti-channel receiving signals based on active diversity to efficiently combat the multipathfading effects; secondly, it utilizes equalization to overcome ISI, through which source signalscan be perfectly recovered; thirdly, it achieves satisfactory blind equalization performance usingshort data samples for burst signal.The main work and innovative achievements obtained in this thesis are summarized asfollows:1. Targeting at the fact that static diversity combination techniques are ineffective indispersive channels, this thesis proposed diversity-combining blind equalization algorithm basedon MMSE criterion. The principle of the algorithm lies in the idea weighting thereceiver-branches’ and equalizer’s weights simultaneously by the blind equalization algorithm.At the same time, simulation performance shows that the linear and decision feedbackdiversity-combining blind algorithms based on MMSE criterion always perform well when thereceiver-branches are not in phase and the algorithms are robust to receiver-branches’ timedelays.2. Targeting at the fact that blind equalization algorithms of SIMO system based on secondorder statistics suffer from the channel order estimation error, this thesis proposed a novel directblind equalization algorithm independent of channel estimation. The proposed algorithmintroduced a new second order statistics, completely independent of explicit and implicit orderestimation. Research shows that equalization performance is unaffected by channel orderestimation, thus the proposed algorithm can be efficiently applied to several different modulationsignals and can be used in complicated shortwave environment.3. Targeting at the fact that traditional SIMO equalization algorithms need the assumptionthat the parallel channels should be coprime, this thesis proposed a blind equalization algorithmfor SIMO system with common zeros. First, the feasibility report of adopting IIR blindequalization algorithm for SIMO system with common zeros is proposed; then Two-Step IIRblind equalization algorithm based on the feasibility report is proposed for equalization ofseveral different modulation signals in SIMO system with common zeros.4. Semi-definite Programming (SDP) can be used for equalization based blind sequencedetection with short data samples in shortwave burst communication systems. But until today,research on the SDP method for blind equalization is limited and applied only to constantmodulus signals such as BPSK/4QAM signals. Through blind sequence detection of BPSK/4QAM, this thesis summarizes the general methods and steps, according to which twoschemes for blind sequence detection of multi-modulus signals such as MPAM/MQAM signalsare proposed by introducing medium matrix and semi-definite relaxation appropriately.Moreover, using the SDP solver toolbox cvx which is easy to use and has simple interface, wecan solve the SDP problem efficiently.