Opt Lett. 2019 Dec 1;44(23):5723-5726. doi: 10.1364/OL.44.005723.
The Brillouin instantaneous frequency measurement (B-IFM) is used to measure instantaneous frequencies of an arbitrary signal with high frequency and broad bandwidth. However, the instantaneous frequencies measured using the B-IFM system always suffer from errors, due to system defects. To address this, we adopt a convolutional neural network (CNN) that establishes a function mapping between the measured and nominal instantaneous frequencies to obtain a more accurate instantaneous frequency, thus improving the frequency resolution, system sensitivity, and dynamic range of the B-IFM. Using the proposed CNN-optimized B-IFM system, the average maximum and root mean square errors between the optimized and nominal instantaneous frequencies are less than 26.3 and 15.5 MHz, which is reduced from up to 105.8 and 57.0 MHz. The system sensitivity is increased from 12.1 to 7.8 dBm for the 100 MHz frequency error, and the dynamic range is larger.
布里渊瞬时频率测量（B-IFM）用于测量具有高频率和宽带宽的任意信号的瞬时频率。然而，由于系统缺陷，B-IFM 系统测量的瞬时频率总是存在误差。为了解决这个问题，我们采用了卷积神经网络（CNN），它建立了测量和标称瞬时频率之间的函数映射，以获得更准确的瞬时频率，从而提高了 B-IFM 的频率分辨率、系统灵敏度和动态范围。使用所提出的 CNN 优化的 B-IFM 系统，优化后的和标称瞬时频率之间的平均最大和均方根误差小于 26.3 和 15.5 MHz，比高达 105.8 和 57.0 MHz 有所降低。对于 100 MHz 的频率误差，系统灵敏度从 12.1 dBm 增加到 7.8 dBm，动态范围更大。
