Compressed sensing of IR-UWB signals with waveform sparsity

Extremely high sampling rate is required for digital processing in impulse radio Ultra-Wideband (IR-UWB) system, and it is difficult to meet this requirement under the current level of the hardware chips. To reduce this sampling rate requirement, a waveform signal recovery method is proposed based on compressed sensing (CS) theory in this paper. The CS model for IR-UWB signal is developed based on measurement matrix whose entries satisfy with the logarithm normal distribution. In order to ensure the success of the CS recovery, the waveform domain transform of IRUWB signal is developed to improve the sparsity of IR-UWB signal. Simulation results demonstrate that the sparse representation of the IR-UWB signal in waveform domain has better sparsity than the time-domain signal, and 1/15–1/20 of the Nyquist sampling rate will be sufficient to efficiently recover the IR-UWB signals.