FM Broadcasting Monitoring Method Based on Time Series Analysis

With the development of modern wireless applications and the popularization of Artificial Intelligence (AI), the intelligentization of radio monitoring system is a development direction. In this context, in order to achieve an intelligent FM broadcasting monitoring, this paper implemented an algorithm which consists three steps. Firstly, occupied channel is searched based on energy detection. Then, Autoregressive Integrated Moving Average (ARIMA) model is adopted to predict the abnormality in the spectrum of FM broadcast. Thirdly, a completely-fully connected neural network is applied to classify the spectrum, to find whether it is authorized spectrum or the frequency bands are occupied by unauthorized users. Results show that the recognition rate for authorized and unauthorized user is 100% and 79%, respectively.