Backscatter communication is a key enabling technology for ultra-low-power Internet of Things (IoT) systems; however, the inherently weak reflected signals make the communication links highly vulnerable to performance degradation under active jamming. This paper proposes an adaptive reflection coefficient control mechanism to enhance anti-jamming capability in dynamically interfered environments. The proposed scheme allows the backscatter device to adjust its reflection coefficient based on estimated interference levels and instantaneous channel conditions, instead of relying on a fixed reflection configuration. A multi-state jammer model with time-varying channel occupation probability is developed, from which analytical expressions for the signal-to-interference-plus-noise ratio (SINR), outage probability, and
expected throughput are derived. The reflection selection algorithm is designed with low computational complexity, making it suitable for resourceconstrained backscatter devices. MATLAB simulations with 105 iterations demonstrate that the proposed method reduces the average outage probability by approximately 27.4%, increases effective throughput by 18.6%–31.2%, and improves SINR stability by about 22% compared with fixed-reflection
schemes. In addition, energy efficiency is improved by nearly 19% under highpower jamming scenarios. The results confirm the feasibility and effectiveness of the adaptive mechanism for anti-jamming backscatter systems