S-Parameters Based Channel Characterization Using Ultra-Wideband Pulse Signal

Ultra-Wideband has recently become a powerful technology that involves wireless communication, digital signal processing, machine learning, and even hardware design for ubiquitous sensing applications. This paper presents a straight forward signal processing method to characterize the wireless communication channel using the UWB pulse signal. This proposed method uses a narrow width based modified Hermitian pulse (MHP) for calculating the output signal at the receiver end from measured network parameters. Magnitude attenuation, wave shape distortion, and phase shift of the output signal indicate the impact of the communication channel on the traveling wave. A neural network-based deep learning algorithm uses to classify the obstacle types from the measured output signal using the MATLAB tool. Measurement results demonstrate the real scenario of channel effects on the high-frequency UWB signal and select a suitable modulation scheme for data encoding. It also helps design the receiver compensation, reconstruct the distorted signal, and channel access to use the MHP pulse signal for high-density data communication. This method can also be used other sensing applications such as damage localization, detection, displacement measurement, material identification, etc.