Dynamic Self-Interference Cancellation for Mitigating PLL Non-Ideality in Backscatter Communications

Backscatter communication (BC) has emerged as a promising paradigm for enabling energy-efficient and low-cost Internet of Things (IoT) systems. One practical challenge for BC implementation is phase-locked loop (PLL) non-ideality, which mainly includes phase noise and spurs. These non-ideal factors can introduce time-varying interference, yielding wavy backscatter signals and striped-shape constellation clusters that can substantially degrade the system performance. In this paper, we investigate the PLL non-ideality of the BC system, and propose a method to suppress its resulted time-varying interference. Specifically, we first establish the BC system model with phase noise and spurs, mathematically showing how these factors can generate time-varying interference and result in wavy signals and striped-shape constellations. We then introduce Dynamic Self-Interference Cancellation (DSIC) algorithms to mitigate the time-varying interference and implement it on a practical backscatter platform. Finally, our experimental measurements show that DSIC can reduce the bit error rates (BER) by half, demonstrating its effectiveness in mitigating the effects of PLL non-ideality and improving the system performance.