Nonlinear filtering algorithms of acoustic signals in oil producing wells

The rsearch describes the algorithms for filtering signals in the signal transmission conditions of the oil-well acoustic channel. The complexity of such systems development is that the acoustic signal transmitter is located deeper than submersible equipment and has a self-contained power supply. In addition, the electrical submersible pump creates a strong acoustic noise. To solve this problem, the neural network filter based on the Haar wavelet decomposition is proposed. The experiment has shown that the proposed filter is able to restore the signal with a signal-to-noise ratio to -15 dB unlike other filtering algorithms considered. The neural network basis for wavelet decomposition allows one to build an effective program and to implement a filter on the microprocessor. For these purposes, the Texas Instruments SM320F28335-HT signal microprocessor has been selected as it has all the necessary characteristics for the filter work.