Channel Estimation for Backscatter Relay System With Dynamic Reflection Coefficient

Ambient backscatter communication (AmBC) systems with energy harvesting (EH) can achieve competitive data rates, making them a robust choice for Internet of Things (IoT) networks. In this case, channel characteristics are fundamental to the performance and efficiency of AmBC. However, the existing channel estimation methods are mostly considered in fixed scenarios, resulting in significant performance loss. Thus, in this work, we explore a backscatter relay system comprising a radio frequency (RF) source, mobile RFID tag, and reader. We propose two channel estimation schemes: Dynamic Least Squares (DLS) and Dynamic Minimum Mean Square Error (DMMSE) and derive the closed-form expression for achievable rate. By comparing analytical results for achievable rate and mean squared error (MSE) with the considered channel estimation schemes that incorporate variable input power and frequency, we can better understand the performance improvements and trade-offs. The numerical results show that AmBC using dynamic RC channel estimation schemes have a higher average achievable rate than conventional methods, and the DMMSE scheme performs better than the DLS scheme. Additionally, we achieve the optimal power and frequency corresponding to the optimal RC, which will significantly improve the performance of the AmBC system.