CNN Classification Architecture Study for Turbulent Free-Space and Attenuated Underwater Optical OAM Communications
Source abstract
Abstract
Turbulence and attenuation are signal degrading factors that can severely hinder free-space and underwater OAM optical pattern demultiplexing. A variety of state-of-the-art convolutional neural network architectures are explored to identify which, if any, provide optimal performance under these non-ideal environmental conditions. Hyperparameter searches are performed on the architectures to ensure that near-ideal settings are used for training. Architectures are compared in various scenarios and the best performing, with their settings, are provided. We show that from the current state-of-the-art architectures, DenseNet outperforms all others when memory is not a constraint. When memory footprint is a factor, ShuffleNet is shown to performed the best.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Orbital Angular Momentum in Optics · Optical Wireless Communication Technologies · Underwater Vehicles and Communication Systems
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Computer Science, Utah State University, Logan, UT 84322, USA
U.S. Naval Research Laboratory, Washington, DC 20375, USA
DCS Corporation, Alexandria, VA 22310, USA
References (35)
Show source-provided references
1. Communication with spatially modulated light through turbulent air across Vienna10.1088/1367-2630/16/11/113028 ↗
2. Machine learning approach to OAM beam demultiplexing via convolutional neural networks10.1364/AO.56.003386 ↗
3. Lohani, S., and Glasser, R.T. (2019). Robust Free Space OAM Communications with Unsupervised Machine Learning. Frontiers in Optics, Optical Society of America.10.1364/FIO.2019.FTu5B.3 ↗
4. Deep Learning With Persistent Homology for Orbital Angular Momentum (OAM) Decoding10.1109/LCOMM.2019.2954311 ↗
5. Transport-based model for turbulence-corrupted imagery10.1364/AO.57.004524 ↗
6. Experimental measurements of the magnitude and phase response of high-frequency modulated light underwater10.1364/AO.56.004019 ↗
7. Orbital angular momentum of light and the transformation of Laguerre-Gaussian laser modes10.1103/PhysRevA.45.8185 ↗
8. Terabit free-space data transmission employing orbital angular momentum multiplexing10.1038/nphoton.2012.138 ↗
9. Free-space information transfer using light beams carrying orbital angular momentum10.1364/OPEX.12.005448 ↗
10. Entanglement of the orbital angular momentum states of photons10.1038/35085529 ↗
11. Detection of a Spinning Object Using Light’s Orbital Angular Momentum10.1126/science.1239936 ↗
12. Measuring the orbital angular momentum of a single photon10.1103/PhysRevLett.88.257901 ↗
13. Measurement of the light orbital angular momentum spectrum using an optical geometric transformation10.1088/2040-8978/13/6/064006 ↗
14. Topological charge and angular momentum of light beams carrying optical vortices10.1103/PhysRevA.56.4064 ↗
15. Gradient-based learning applied to document recognition10.1109/5.726791 ↗
16. ImageNet Classification with Deep Convolutional Neural Networks10.1145/3065386 ↗
17. Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., and Fei-Fei, L. (2009, January 20–25). ImageNet: A Large-Scale Hierarchical Image Database. Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA.10.1109/CVPR.2009.5206848 ↗
18. Kingma, D.P., and Ba, J. (2015, January 7–9). Adam: A Method for Stochastic Optimization. Proceedings of the 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA.
19. He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27–30). Deep Residual Learning for Image Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.10.1109/CVPR.2016.90 ↗
20. Simonyan, K., and Zisserman, A. (2014). Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv.
21. Xie, S., Girshick, R., Dollár, P., Tu, Z., and He, K. (2017, January 21–26). Aggregated Residual Transformations for Deep Neural Networks. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2017.634 ↗
22. Huang, G., Liu, Z., Maaten, L.V.d., and Weinberger, K.Q. (2017, January 21–26). Densely Connected Convolutional Networks. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2017.243 ↗
23. Iandola, F.N., Moskewicz, M.W., Ashraf, K., Han, S., Dally, W.J., and Keutzer, K. (2016). SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1 MB model size. arXiv.
24. Howard, A.G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., and Adam, H. (2017). MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. arXiv.
25. Zhang, X., Zhou, X., Lin, M., and Sun, J. (2018, January 18–23). ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices. Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00716 ↗
26. Azulay, A., and Weiss, Y. (2018). Why do deep convolutional networks generalize so poorly to small image transformations?. arXiv.
27. Wilson, A.C., Roelofs, R., Stern, M., Srebro, N., and Recht, B. (2017, January 4–9). The Marginal Value of Adaptive Gradient Methods in Machine Learning. Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA.
28. Ruder, S. (2016). An overview of gradient descent optimization algorithms. arXiv.
29. Dozat, T. (2016, January 2). Incorporating Nesterov Momentum into Adam. Proceedings of the ICRL 2016, San Juan, Puerto Rico.
30. Lecture 6.5—RmsProp: Divide the gradient by a running average of its recent magnitude
31. Goodfellow, I., Bengio, Y., and Courville, A. (2017). Deep Learning, The MIT Press.
32. Random Search for Hyper-Parameter Optimization
33. Bergstra, J., Yamins, D., and Cox, D.D. (2013, January 16–21). Making a Science of Model Search: Hyperparameter Optimization in Hundreds of Dimensions for Vision Architectures. Proceedings of the 30th International Conference on International Conference on Machine Learning, Atlanta, GA, USA.
34. Liaw, R., Liang, E., Nishihara, R., Moritz, P., Gonzalez, J.E., and Stoica, I. (2018). Tune: A Research Platform for Distributed Model Selection and Training. arXiv.
35. Smith, L.N. (2017, January 24–31). Cyclical Learning Rates for Training Neural Networks. Proceedings of the 2017 IEEE Winter Conference on Applications of Computer Vision (WACV), Santa Rosa, CA, USA.10.1109/WACV.2017.58 ↗