2-Wire 4-Duplex Transmission Signal Denoising Using Nonlinear Wavelet Thresholding

High-speed data transmission is expected in communication and control systems. Conventionally, in 2-wire half-duplex transmission system, data can be transmitted in both directions but not at the same time. In order to increase transfer performance, clock frequency must be increased. However, if transmission frequency exceeds 100 kHz, transmission environment will become selective. What we call 2-wire 4-duplex transmission system is treated in this paper. A low-frequency high-speed data transmission is available using the 2-wire 4-duplex transmission system. However there is a big problem if transmission signal is contaminated by noise. It will become difficult to recognize components of the signal from noisy signal. In this research, nonlinear wavelet thresholding is applied to 2-wire 4-duplex transmission signal in order to remove noise. We propose a modified universal threshold and a noise estimation method instead of median absolute deviation (MAD) method. Recognition of basic signals from denoised transmission signal is shown. We have evaluated the proposed methods on 2-wire 4-duplex transmission signal