Wireless tomography, Part III: Compressed sensing for ultra-wideband signals

This is Part III of the wireless tomography three-paper series. Wireless tomography is related to wireless communications in that it requires the channel recovery between different waveforms at transmit and receive, as well as multiple-input-multiple-output (MIMO) communication system. According to the pulse propagation mechanisms of reflection and diffraction, ultra-wideband (UWB) waveforms suffer pulse distortion. Distorted pulses will overlap and therefore increases the sampling rate for accurate UWB channel recovery. Thanks to the recent progresses in sampling theory and radio propagation theory, we are able to propose a compressed sensing (CS) based UWB channel recovery method considering pulse distortion. The concept has been demonstrated through simulations. The sampling rate is as low as 2 Gsps, compared with the Nyquist rate of 50 Gsps. A CS based 2×2 MIMO communication system is also proposed and simulated. The communication problem is modeled as CS problem, and further reduce sampling rate required at the receiver.