The Recognition Method Of Ground-To-Air Communication Interference Signals Based On Support Vector Machine

With the increasing complexity of electromagnetic environment, the event of radio communication interference is becoming more seriously and affecting the safety of people’s life, property. At present, air ground communication abnormal signal search is often perform ed by monitoring staff experience combined with the application of monitoring equipment. It is need more work to do. It is becoming an important target of radio monitoring work and has a high theoretical value, that how to accurately, efficiently identify air ground communication abnormal signal.The business of ground to air communication is generally the speech communication.Although it has the accidental appear and lowly probability characteristics, the damage is very strong. Therefore in the ground air communication abnormal signal recognition processing, it is very important that how to effectively use the voice information intuitive, choosing the appr opriate classifier be accurate fast, efficient, and automatically identify.K- means clustering algorithm has been widely used in signal processing and signal recognition feature. But the classifier K- mean clustering algorithm exists the problem of the initial clustering center, so the recognition rate is not the stationary in different data sets. While Support vector machine(SVM) is good at solving the classification problem of signal complex and is widely used in the field of image processing, medical research. This paper will be doing more research in SVM classifier, giving a new classifier based on the artificial intelligence optimistic. And they would be applied to construct ground to air communications audio signal characteristics and its recognition. The main research contents are followed as:1.Explore the radio ground-to-air communications audio signal as radio identification on the basis of the feasibility of ground-to-air communications abnormal signal. K- means clustering algorithms are used to construct sets of its characteristics and to do the discriminate experiment by Euclidean distance method.2.We propose that using gravitational search algorithm to optimize the selection of kernel function parameter and penalty factor of support vector machine(GSA-SVMC)。3.The GSA-SVMC is applied to identify the ground to air communication of audio signal.Through the contrast experiment shows it is a new method for recognition of air ground communication interference of audio signals.