Adaptive Pre-Compensation of Transmitter Phase Noise for User Coexistence in Cognitive Radios

Secondary wide-band users have to guarantee they will not disturb the in-band narrow-band primary licensed users. There are techniques proposed to limit the interference by secondary wide-band users in the specified in-band narrow bands. Active Interference Cancelation (AIC) can flexibly solve this problem in the digital domain and hence remove the need for extra expensive analog hardware. The main challenge is the degradation in the final depth of a nulled band by radio impairments such as PA non-linearity, phase noise, and I/Q imbalance. In this paper, we analyze the effect of phase noise on the performance of active interference cancelation techniques and then propose a mitigation technique. Our analysis shows that phase noise can degrade the nulling depth performance by 11.4-19.6dB at the edge of nulled band. We propose an adaptive pre-filtering technique to improve the nulling performance by ~3-11dB depending on system configuration. The proposed scheme utilizes the existing analog-to-digital converter in the device's receiver path for providing feedback when in transmit mode. This can leads to higher power consumption in transmit mode due to this additional analog-to-digital conversion in the feedback path. To address this limitation, we propose an adaptive tracking filter with low duty-cycle on/off periods. We show that sacrificing 0.5dB in the null-depth can reduce the ADC power consumption by 95%.