Mining association rules for electromagnetic spectrum data

In order to make more rational use of spectrum resources and better evaluate various electromagnetic environments, this paper proposes a spectrum data mining scheme based on association rule mining. Firstly, based on the general mining process, the scheme obtains the useful information in spectrum data, including anomaly information, bottom noise information, occupancy information, predetermined time power information and so on. Then, the spectrum information is taken as the association analysis object, the association database and fuzzy set are constructed, and the spectrum information is systematically analyzed based on the fuzzy association rule mining algorithm. In this paper, the traditional operator selection strategy is improved, and the large-scale parameters are used to improve the fuzzy membership function. Through the verification and analysis of the measured data set, the experimental results show that the strong association rules of spectrum information can reflect the implicit association between various information and the frequency of various information. Association rule mining based on spectrum information can effectively simplify spectrum mining. Through the correlation of various information, another spectrum information can be obtained by analyzing a part of spectrum information. At the same time, the association rules of spectrum information can be used to evaluate the electromagnetic radio environment. By selecting the appropriate spectrum information, the scheme can be applied to the evaluation of various electromagnetic environments.