The Recognition Of Radio Frequency Band Abnormal Signal

Broadcasting is a media of the distance of people’s daily lives. Regardless of theaudience age, educational level, broadcast is suitable for all people. Because of its’ mobilityand portability, people can easily get the latest information from the radio at anytime,anywhere. Due to the rapid development of society, the pace of people’s lives is faster.Broadcast become an important media for people to obtain information. People canunderstand the weather conditions, road traffic conditions, entertainment and informationthrough the radio. In major disasters, it can also serve as information distribution platform.The broadcast band which is not standardized or illegal used direct impact on people’snormal life and information which they want to get. If the information blocked or unclear atan important moment, it may even cause a significant loss of property and life.From the perspective of monitoring, we can able to find the abnormal signal about thenon-standard or illegal using of the frequency. In this paper, we introduce some existidentification methods about abnormal signal from the experience. However, the majority ofthe existing methods about discovering and recognizing of abnormal signal are by simplestatistics and four operations. Monitoring results and the actual situation will differ widely. Inthis paper, we propose a new method to indentify the abnormal signal. The method uses thefuzzy C-means or K-means to deal with the data. Properties characteristic of the signal isextracted and converted to the same measure. Using ordered weighted aggregation operatorsto polymerize and using the polymerization results and related databases to identify. The mainresearch works are as follows:(i) According to the description from the staff and monitoring data, the paper abstract out theformula to calculate the characteristics of radio signals.(ii) According to the characteristics of the radio signal, we propose an identification of theabnormal signal of the broadcast band. In this method, we normalize the character dataand use the ordered weighted aggregation operators to aggregate the data.(iii)During the polymerization, the weight of the choice directly impact on the result ofpolymerization. We propose a method to calculate the weight. The weight vector isapplied to the Fuzzy C-Means as the variables. Stop the iteration when the fuzzyclassification matrix and the center vector are in a certain error range.Using the Above method to test measured data of the broadcast band and compare severalexisting identification methods in the experiment.