基于深度限制波尔兹曼机的辐射源信号识别

针对电子侦察中使用常规参数难以有效识别复杂体制雷达信号的问题,提出利用深度限制波尔兹曼机对辐射源识别的模型。模型由多个限制波尔兹曼机组成,通过逐层自底向上无监督学习获得初始参数,并用后向传播算法对整个模型进行有监督的参数微调,利用Softmax进行分类识别。通过仿真实验表明该模型能对辐射源进行有效的特征提取和分类识别,具有较高的识别精度和较强的鲁棒性。

Emitter Recognition Method based on Deep Restricted Boltzmann Machine

To deal with the problem of radar emitter recognition caused by parameter complexity and agility of muti-function radars in electronic intelligence reconnaissance field, a new recognition model which is based on Deep Restricted Boltzmann Machine is proposed in this paper. The model is composed of multiple restricted boltzman machine. A bottom-up hierarchical unsupervised learning is used to obtain the initial parameters, and then the traditional back propagation algorithm is conducted to fine-tune the network parameters, Softmax is used to classify the results at last. Simulation and comparison experiment shows that the proposed method has the ability of extracting the parameter features and recognizing the radar emitters, and it has strong robustness as well as high recognition rate.