采用小波包变换和隐马尔科夫模型结合的方法,提出了机载通信电台的故障诊断模型。通过评价函数的建立,得到最优测试点集合;利用小波包变换提取测试点的能量特征值作为模型的输入,利用历史数据对隐马尔科夫模型进行训练,得到各个状态下的HMM参数,通过模式识别,实现电台的故障诊断;试验研究结果验证了所提出的方法的可行性和有效性。 Using the combination of wavelet packet transform and hidden Markov model, the fault diagnosis model of airborne communication station is proposed. Through the establishment of the evaluation function, the optimal set of test points is obtained. Using the wavelet packet transform to extract the energy eigenvalues ​​of the test points as the input of the model, the Hidden Markov Model is trained by historical data to obtain the HMM parameters in each state. Pattern recognition, to achieve the radio fault diagnosis; experimental results verify the feasibility and effectiveness of the proposed method.