Full MIMO spatial filtering approach for dynamic range reduction in wideband cognitive radios

Wideband cognitive radios (CRs) receive signals from multiple transmitters simultaneously to increase spectrum utilization. Processing a wideband spectrum is challenging due to large dynamic range (DR) of the received signal and required high sampling speed of the ADC. The power consumption of high sampling speed/high-resolution ADCs have been prohibitive for handheld radios. However, in CR applications strong inband signals that pose large DR requirements can be filtered out, since CR needs to detect unused spectrum bands where no signal is present. Spatial domain filtering approaches through use of multiple antennas to reduce DR of the wideband signal are proposed. Algorithms and architectures are developed for vector beamforming (multiple antennas and a single ADC) and full multiple-input multiple-output (MIMO) (multiple antennas with an ADC per antenna) analog spatial filters for adaptive interference suppression. Simulation results indicate that for realistic indoor propagation environments the ADC resolution of an analog beamformer can be reduced by 4 bits when the receiver operates at 2 bits/s/Hz, reducing ADC power consumption by approximately 90%. Moreover, simulations indicate that full MIMO analog spatial filter can reduce ADC resolution with over 3 bits per ADC when the receiver operates at 5 bits/s/Hz, reducing ADC power consumption by approximately 85%.