Algorithm for False Alarm Stabilization against the Background of Nonstationary Noise with Trend Estimation

Introduction. A detection algorithm that ensures a constant value of the false alarm rate against the background of nonstationary noise, whose average value varies within a sliding window. The proposed algorithm is based on a linear approximation of the average noise level within a sliding window using the least squares method with subsequent compensation for changes in the average value. The effectiveness of the proposed algorithm was assessed using a simulation method. When working against the background of nonstationary noise, the proposed algorithm reduces the detection threshold compared to an algorithm based on calculating noise dispersion in a sliding window. Aim. Development of a detection algorithm that takes into account the mathematical expectation of noise within a sliding window when calculating the detection threshold. Materials and methods. The research was carried out using the mathematical apparatus of probability theory and estimation theory. The effectiveness of the developed algorithm was assessed by mathematical simulation. Results. A detection algorithm that ensures a constant value of the false alarm value F when detecting a signal against the background of noise was developed. When working against the background of nonstationary noise, the algorithm provides the threshold signal-to-noise ratio of 3.57 dB lower than that provided by an algorithm based on calculating noise dispersion by averaging the elements of a sliding window. Conclusion. This paper proposes an algorithm for stabilizing the false alarm rate based on assessing the noise trend and its subsequent compensation in the subtracting device. The algorithm ensures the constant value of the false alarm rate under changes in the average value of noise within a sliding window.