Application Of Wavelet Denoising Combined With Bispectrum Analysis In Carrier Communication

The correlation properties of the medium-voltage power line carrier have endured detailed study and large amount of test by the related domestic and international research institutions. Because of too much branch and distribution transformers on the medium-voltage power line, the wave trapper are not easy to be used and exist noise on actual information channels at the same time, the medium-voltage power line communication has obvious different with the other communication modes. The medium-voltage power line communication applied widely and involved in many fields. With the features of bigger signal noise, stronger disturbance and more complex environment appeared on the carrier communication channel of the medium-voltage power line, the communication sign would be reduced and the stability would be affected. What method adopted to solve these problems above has been the core content to realize the medium-voltage power line carrier communication. The main research contents in this article are as follows:(1)The medium-voltage carrier signal were detected by associated equipment, and analyzed to filter the mixed noise and disturbance, in order that the carrier could be transformed into the other styles which transferred, processed, studied and identified more easily. Lots of information in the whole process was analyzed and a clear judgment was put forward based on the research on the state of power carrier signal.(2)Aim at there exist noising distribution on the medium-voltage power line communication, the wavelet de-noising combined with bi-spectrum analysis method was put forward. Mainly based on the signal gathered from medium-voltage broadband power line communication, the wavelet de-noising was set in the time-frequency domains, the AR-model was successfully modeled, and bi-spectrum calculation of signals was given after wavelet transform. In the aspects of processing signals and extracting signal features, the characteristic of wavelet and bi-spectrum were considered, and the effect of wavelet de-noising combined with bi-spectrum calculation was better than traditional bi-spectrum analysis. For this reason the wavelet transform domain bi-spectrum analysis was adopted in this article and the extracting of medium-voltage broadband power line communication signals became more obvious.(3)The Matlab simulation was given and the bi-spectrum images of medium-voltage carrier signal were got. A kind of phenomenon was found that the wave crest on the bi-spectrum images increased gradually and the energy distribution on the contour line also became more diffuse from the bi-spectrum images of several environments without wavelet de-noising. On the contrary, these kinds of signal feature became more similar, the spectral peak became more smoothly, and the energy distribution on the contour line also became more concentrated under different environments for the main disturbance had been removed, there were the changes appeared on the bi-spectrum images after wavelet de-noising.