To address the performance degradation caused by power fluctuations of tones, the multi-frame coherent integration-based track-before-detect (TBD) method is proposed. In the proposed method, the high gain advantage of coherent integration and the information accumulation ability of multi-frame TBD method based on batch processing technique are combined to improve the detection and tracking performance of fluctuating tones. Firstly, a new measurement model is established with batch processing technology and coherent integration across multiple data frames is achieved utilizing the tonal coherence, thus the SNR in measurement is greatly improved. Then, a new likelihood ratio function is derived to match the established measurement model, ensuring that TBD processing can be performed correctly. Finally, the effects of the newly established likelihood ratio and the batch length on the detection performance are analyzed in detail through theoretical derivations. The superiorities and robustness of the proposed method are demonstrated through simulation analysis and processing results for sea experimental data.

中文翻译：

---

被动声纳弱音多帧相干检测前跟踪方法

为了解决音调功率波动引起的性能下降问题，提出了基于多帧相干积分的检测前跟踪（TBD）方法。该方法结合了相干积分的高增益优势和基于批处理技术的多帧TBD方法的信息积累能力，提高了波动音调的检测和跟踪性能。首先，利用批处理技术建立了新的测量模型，并利用音调相干性实现了多个数据帧的相干集成，从而大大提高了测量中的信噪比。然后，推导新的似然比函数来匹配已建立的测量模型，确保TBD处理能够正确进行。最后，通过理论推导详细分析了新建立的似然比和批次长度对检测性能的影响。通过海上实验数据的仿真分析和处理结果证明了该方法的优越性和鲁棒性。