<p>This article tackles the distributed filtering problem over wireless sensor networks for a class of time-varying systems with state saturation and deception attacks. To ensure more reliable and secure data transmission, a novel dynamic decode-and-forward (DDaF) relay scheme is proposed, implemented through coordinated encoders deployed at the sensors’ side and decoders deployed at the relay nodes’ side. The network-based deception attacks occur probabilistically across the communication channels and are characterized by independent Bernoulli random sequences. This study is primarily intended to design distributed filters that effectively estimate the system states in the presence of state saturation, DDaF relays and deception attacks. By employing the stochastic analysis and matrix theories, upper bounds on the second-moment matrices of the filtering errors are derived for the considered system, and subsequently such bounds are minimized with designed filter gains. Furthermore, rigorous mathematical analysis is conducted to examine the uniform boundedness of the upper bounds and their monotonicity with respect to the deception attacks. Finally, simulation results demonstrate effectiveness of the proposed distributed filtering strategy.</p>