Application of Sparse Bayesian Model in FrequencyHopping
Signal Station Separation

When the electromagnetic environment is more complicated, in existing frequencyhopping (FH) signal separation approaches, such as Kmeans clustering and support vector machine (SVM), these approaches confront with either low separation rate or high computational complexity. To address these two issues, this paper proposes a sparse Bayesian learning (SBL) based method for FH signal separation. After establishing the FH signal separation model, the SBL method is introduced to achieve the separation of different FH stations, furtherly,this paper constructs a multiclassification model of FH stations separation for multiple FH stations existence.The effectiveness of the proposed method is validated by numerical results.