Improved coded/uncoded monobit receiver for transmit-reference UWB communication systems: Performance evaluation and digital circuit design

Abstract

In transmit-reference (TR) receivers, noise in the reference (pilot) pulse causes a noise × noise term in the decision statistic. This term severely degrades receiver performance. We propose a detection algorithm to decrease the noise in the reference pulse in TR receivers. Computer simulations show that at BER = 10−5, the proposed algorithm outperforms the existing algorithm by about 0.8 dB in both coded and uncoded scenarios. In high-rate applications, inter-symbol interference (ISI) drastically impairs the received signal. Monobit quantization intensifies the ISI and hence dramatically degrades the receiver performance. At first glance, channel coding increases ISI and seems useless to combat ISI. However, we propose to employ channel coding to overcome the severe performance loss caused by ISI. Simulation results show that in high SNR values, channel coding tackles the ISI problem especially when convolutional coding is used. Moreover, simulations reveal that convolutional coding is more robust to ISI compared to block coding schemes. Furthermore, the digital circuit design of coded/uncoded traditional and proposed monobit TR receivers is done in 0.18 µm CMOS technology at 2 V supply voltage. Simulations confirm the numerical results obtained in MATLAB. Finally, the power consumption of each receiver is reported.