A UAV Detection Algorithm Based on an Artificial Neural Network

This paper presents an artificial neural network (ANN)-based detection algorithm for an unmanned aerial vehicle (UAV). The slope, kurtosis, and skewness of the signal received from the UAV are employed in this algorithm. The training of the three corresponding feature matrices is done using UAV, and non-UAV signals can be classified effectively for the UAV sensor network based on ANN. Outdoor data over a bridge in the Jimo District, Qingdao, and indoor data from a research laboratory are used for system training and evaluation. The results obtained show that the proposed detection algorithm based on an ANN outperforms methods based on the slope, kurtosis, and skewness of the received signal in terms of the error rate. The recognition rate with the proposed algorithm is greater than 82% within a distance of 3 km, which is better than other UAV detection methods such as active radar, acoustic, and visual recognition.