雷达辐射源信号分选是现代战争中雷达电子侦察的关键环节。然而，随着电磁环境日益复杂，侦察接收机截获的信号往往存在大量脉冲缺失、虚假脉冲和脉冲抖动。复杂电磁环境具有信号长程相关、数据分布多样性、不同时间步信息互补等特点，而现有基于双向长短期记忆网络（Bi-LSTM）的雷达信号分选方法性能不足，主要是因为其全局依赖建模能力不足、存在梯度消失与过拟合风险以及特征交互受限。针对此，提出一种集成注意力的Bi-LSTM方法，通过集成多头自注意力机制、层归一化和残差连接等技术，分别解决上述问题。主要改进工作包括：1）数据预处理：将原始到达时间数据转换为到达时间差数据，以提升输入数据的鲁棒性与精度；2）特征提取与融合：采用Bi-LSTM层提取正向与反向时序信息，并引入多头自注意力机制进行深层次特征融合，兼顾长程依赖的捕捉与局部细节的提取；3）优化训练稳定性：利用残差连接与随机失活正则化，保证梯度传递的畅通性。实验结果表明该方法相较于Bi-LSTM方法准确率有明显提升，结构上的改进有效增强了模型对脉冲缺失、虚假脉冲和脉冲抖动的鲁棒性。

Radar signal sorting is a key aspect of radar electronic reconnaissance in modern warfare. However， as the electromagnetic environment becomes increasingly complex， the signals intercepted by reconnaissance receivers often suffer from a large number of missing pulses， spurious pulses， and pulse timing jitter. The complex electromagnetic environment is characterized by long-range signal correlations， diverse data distributions， and complementary information across different time steps. Existing bidirectional long short-term memory（Bi-LSTM）-based radar signal sorting methods perform inadequately， primarily due to their insufficient ability to model global dependencies， risks of gradient vanishing and overfitting， and limited feature interaction. This paper proposes an integrated attention Bi-LSTM method that addresses these issues through the incorporation of multi-head self-attention mechanisms， layer normalization， and residual connections. The main improvements include： 1） Data preprocessing： Converting the original time of arrival data into time difference of arrival data to enhance the robustness and accuracy of the input data. 2） Feature extraction and fusion： Utilizing Bi-LSTM layers to extract both forward and backward temporal information， while introducing a multi-head self-attention mechanism for deep feature fusion， balancing the capture of long-range dependencies and local detail extraction. 3） Optimizing training stability： Employing residual connections and dropout regularization to ensure smooth gradient propagation. Experimental results indicate that this method achieves significantly improved accuracy compared to traditional Bi-LSTM approaches， and the structural improvements effectively enhance the model’s robustness against pulses loss， spurious pulse， and pulse jitter.