| 王超,杨辉跃,肖玮,何智杰,王范东.基于多尺度残差网络的复值信号频率估计[J].电子测量与仪器学报,2026,40(6):299-307 |
| 基于多尺度残差网络的复值信号频率估计 |
| Frequency estimation of complex-valued signal basedon multi-scale residual network |
| |
| DOI： |
| 中文关键词: 频率估计 深度学习 多尺度残差网络 低信噪比 复值信号 测量仪表 |
| 英文关键词:frequency estimation deep learning multi-scale residual network low SNR complex-valued signal measurement instrument |
| 基金项目:重庆市教委科技项目（KJQN202312903）资助 |
|
|
| Author | Institution |
| Wang Chao | Engineering University of the Joint Logistics Support Force, Chongqing 401331, China |
| Yang Huiyue | Engineering University of the Joint Logistics Support Force, Chongqing 401331, China |
| Xiao Wei | Engineering University of the Joint Logistics Support Force, Chongqing 401331, China |
| He Zhijie | Unit 31680 of the People’s Liberation Army, Chongzhou 611233, China |
| Wang Fandong | Engineering University of the Joint Logistics Support Force, Chongqing 401331, China |
|
| 摘要点击次数: 47 |
| 全文下载次数: 11 |
| 中文摘要: |
| 频率估计是测量仪表领域的关键技术，针对传统周期图法与多信号分类（multiple signal classification，MUSIC）算法在低信噪比环境下性能急剧下降，难以满足仪表测量的精度需求。提出了一种基于多尺度残差网络的频率估计方法（multi-scale frequency representation，MultiFR）。首先，通过并行的3×1和5×1卷积核提取不同尺度特征，增强对局部细节和全局模式的捕获能力。然后，引入残差连接机制克服深层网络梯度消失问题，确保特征有效传播，采用L2范数平方损失函数，实现对网络输出频谱与理想频谱之间偏差的精确度量。最后，使用动态信噪比训练策略，提高模型在不同噪声水平下的适应性，通过高分辨率频率表示输出模块，实现精确的频率峰值定位。实验结果表明，所提方法的频率分辨能力得到提升，在0 dB极低信噪比条件下，MultiFR的漏检率为22.2%，Chamfer距离为0.145，较DeepFreq方法分别相对提升了4.1%和5.8%，并优于PSnet、MUSIC及周期图等主流方法。经过科式流量计信号验证，证明了所提方法在低信噪比下具有一定的可行性、有效性与鲁棒性。 |
| 英文摘要: |
| Frequency estimation is a key technology in the field of measurement instrumentation. Traditional periodogram and multiple signal classification (MUSIC) algorithms suffer from severe performance degradation under low signal-to-noise ratio environments, failing to meet the precision requirements of instrumentation measurements. This paper proposes a frequency estimation method based on multi-scale residual networks (MultiFR). First, parallel 3×1 and 5×1 convolutional kernels extract features at different scales, enhancing the ability to capture both local details and global patterns. Then, residual connections mechanisms are introduced to overcome the gradient vanishing problem in deep networks, ensuring effective feature propagation. The L2 norm squared loss function is adopted to achieve precise measurement of the deviation between the network output spectrum and the ideal spectrum. Finally, a dynamic SNR training strategy is utilized to improve the model’s adaptability under different noise levels, and a high-resolution frequency representation output module is implemented to achieve accurate frequency peak localization. Experimental results demonstrate that the proposed method achieves improved frequency resolution capability. Under extremely low SNR conditions of 0, MultiFR achieves a miss detection rate of 22.2% and a Chamfer distance of 0.145, representing relative improvements of 4.1% and 5.8% respectively compared to the DeepFreq method, and outperforming mainstream methods such as PSnet, MUSIC, and periodogram. Validation with Coriolis flowmeter signals confirms the feasibility, effectiveness, and robustness of the proposed method under low SNR conditions. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
