Chirp Pulse Compression in Non-Coherent Impulse-Radio Ultra-Wideband Communications

The paper investigates chirp pulse compression as a signal dimension reduction technique in non-coherent impulse-radio ultra-wideband (IR-UWB) communications. Two common types of non-coherent IR-UWB detection are considered, energy detection of IR-UWB pulse position modulated (PPM) symbols and sample-wise differential detection of differential phase shift keying (DPSK) IR-UWB symbols, as well as a newly introduced sample-wise differential detection of IR-UWB PPM symbols. A common problem with these low-complexity IR-UWB detection schemes is poor performance when the dimension of the symbol in detection is high, which is usually the case at low data rates and is especially pronounced in interference environments. Chirp pulse compression mitigates this problem by reducing the dimension of the symbol in detection along with very small fading factor, which also lowers the computational complexity in the case of digital implementation. In the analytic part of the paper, we develop a closed-form expression for the bandwidth and dimension of the signal after the chirp pulse compression, which was lacking before. Furthermore, a closed-form expression of the bit error probability in white noise with the above detection schemes is also given. In the numerical part of the paper, we compare the performance with and without chirp pulse compression of the above modulation/detection pairs in noise and interference environments.