Yin Li, Qiaosong Cai, Jie Yang, Tong Zhou, Yuanxi Peng, Tian Jiang, "Adaptive microwave photonic angle-of-arrival estimation based on BiGRU-CNN [Invited]," Chin. Opt. Lett. 21, 090001 (2023)
- Chinese Optics Letters
- Vol. 21, Issue 9, 090001 (2023)
Fig. 1. Schematic diagram of the proposed adaptive microwave photonic AOA estimation system using BiGRU-CNN. V, envelope voltage vector; R, correlation matrix.
Fig. 2. (a) RF signal incident on two array elements of the DDMZM-i; (b) simulation prediction of the relationship between the output envelope voltage and AOA under different signal frequencies: di = 1.875 cm, ΔL = 0 cm; (c) model of the BiGRU-CNN.
Fig. 3. Experimental setup of three-DDMZM-based AOA estimation system.
Fig. 4. (a) Normalized amplitude response for AOA at signal frequency 13 GHz; (b) normalized amplitude response for frequency at AOA 30°.
Fig. 5. (a) Training loss and (b) validation loss of different neural network architectures during training.
Fig. 6. Experimental results of actual AOA and estimated AOA (blue circles) and the corresponding errors at different frequencies (3, 8, and 13 GHz) over a −80° to 80° measurement range. The blue solid line represents the ideal curve for actual AOA and estimated AOA.
Fig. 7. Real-time IFM with 3 GHz pulse signal (width 1 µs, period 2 µs) at AOA of 30°; the frequency of 0 GHz means no signal is incident.
|
Table 1. Parameters of the Optimized BiGRU-CNN
Set citation alerts for the article
Please enter your email address
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20