Comprehensive Effects of Imperfect Synchronization and Channel Estimation on Known Interference Cancellation

Abstract

Known interference (KI) is commonly used in many scenarios, where the original interference data is known to the receiver in advance. Unfortunately, due to the impacts of hardware imperfections and propagation environment, the received KI may suffer frequency offset, phase noise, and multi-path propagation with different time delays and channel gains. However, perfect time-frequency synchronization and channel estimation are hard to implement in practice, making it challenging to cancel the KI completely. In this paper, the comprehensive effects of synchronization and channel impairments on KI cancellation are investigated. First, the normalized time synchronization error is modeled as the inter-symbol interference (ISI), and the frequency synchronization error and phase noise are modeled as the inter-frequency interference (IFI). Subsequently, the closed-form expression of KI cancellation ratio (KICR) is derived for the multi-path block fading channel, by jointly considering imperfect time-frequency synchronization and channel estimation, as well as phase noise. Furthermore, assuming the channel and synchronization parameters estimation reach their corresponding Cramer-Rao bounds (CRBs), the upper bound of KICR and the lower bound of channel capacity loss are analyzed for the single-path channel. Simulation results confirm the theoretical analysis that the KICR is more sensitive to phase noise compared to time-frequency synchronization and channel estimation errors. In addition, it is also proved that lower KI power is less sensitive to the implementation imperfections.