An autonomous detection and robust estimation method of spectrum anomaly based on time series analysis

How to precisely detect electromagnetic spectrum anomaly is a major challenge for radio monitoring and electromagnetic environment evaluation, especially in the condition of complex electromagnetic environment and lack of pre-knowledge information about frequency use. Based on time series analysis theory, a timing model which presents the correlation between the previous and the following sequence of spectrum occupancy, is built to help us realize the autonomous identification and robust estimation of typical spectrum anomaly. The analysis results indicate that, without actually requiring pre-knowledge of frequency database and radio monitoring historical data support, this method can effectively identify the types of spectrum anomaly, occurrence time, anomaly effect intension and other relative information. Furthermore, through the robust estimation of spectrum occupancy model, we can significantly improve the model's fitting performance and raise the adaptability and robustness of the model to external interferences.