Detection of tampering attacks and parameter identification for FIR systems with binary-valued observations: An input design approach

Abstract

This paper investigates the issues of data tampering attacks detection and system parameter identification in finite impulse response (FIR) systems with binary-valued observations. Without the need to acquire system-related prior information or perform operations such as adding watermarks or encryption to the data, a detection and identification joint algorithm is proposed using input design. This algorithm can detect potential data tampering attacks in the system while achieving consistent identification of system parameters. Subsequently, a pair of metrics for evaluating the detection performance of the algorithm, namely the missing detection rate and false detection rate, are introduced, and approximate expressions for both are provided, followed by a discussion on the impact of data length and detection threshold on these metrics. Finally, numerical simulations are conducted to validate the conclusions obtained and the results of the discussions.