AI/ML in Radar System Design

PhD-level Special Topics Class

[The University of Alabama, Spring 2022]

Recent advances and applications of AI and deep learning as applied to radar system design, especially next generation cognitive radar systems, will be discussed.
1:  Basic principles of radar systems; radar range equation, link budget, FMCW and MIMO radar. Radar data representations (2D,3D,4D)
2: Basic principles of machine learning and deep learning
3: Radar applications (SAR imaging, micro-Doppler, biomedical, human-computer interaction, autonomous vehicles, cyber-physical human systems (CPHS))
4: Sensing specific challenges – training under low sample support, synthetic data generation and transfer learning
5: Physics-aware machine learning and application to radar-based and multi-modal sensing
6: Sequential modeling, online and iterative learning methods
7: Cognitive radar and cognitive process modeling
8: Markov decision processes, reinforcement learning and implementation of a perception-action cycle in fully-adaptive radar 

Pre-Requisites:  B.S. in ECE or CS or related field; Prior experience with radar and AI/ML helpful.

MODULE 1 – INTRODUCTION TO CONCEPT OF COGNITIVE RADAR

MODULE 2 – WAVEFORM DIVERSITY AND BIOMIMETIC RADAR 

Student Presentations:
​

• Inspiration from Dolphins for Waveform Diversity [Presenter: Deepak Elluru]
  • T.G. Leighton, G.H. Chua, P.R. White, K.F. Tong, H.D. Griffiths, and D.J. Daniels, “Radar Clutter Suppression and Target Discrimination Using Twin Inverted Pulses,” in Proc. The Royal Society A 469: 20130512, Sep. 2013.
  • M. Greco and F. Gini, “The Biosonar of the Mediterranean Bottlenose Dolphins:  Analysis and Modeling of Echolocation Signals,” in Biologically-Inspired Radar and Sonar: Lessons from Nature, Ed. A. Balleri, H. Griffiths, and C. Baker, IET, 2017.

​

• Phase-Perturbed Waveform Design for SAR-ECCM [Presenter: Emre Kurtoglu]
  • M. Soumekh, “SAR-ECCM using phase-perturbed LFM chirp phase-perturbed_waveform_design_for_sar-eccm.pdfsignals and DRFM repeat jammer penalization,” in IEEE Transactions on Aerospace and Electronic Systems, vol. 42, no. 1, pp. 191-205, Jan. 2006.
  • M. Soumekh, “SAR-ECCM using phase-perturbed LFM chirp signals and DRFM repeat jammer penalization,” IEEE International Radar Conference, 2005., 2005, pp. 507-512.
  • L. Wei, Y. Lin and G. Xu, “A New Method of Phase-Perturbed LFM Chirp Signals for SAR ECCM,” 2018 China International SAR Symposium (CISS), 2018.

• ​Waveform Diversity with Noise Radar [Presenter: Eddie Hackett]
  • S. D. Blunt et al., “Principles and Applications of Random FM Radar Waveform Design,” in IEEE Aerospace and Electronic Systems Magazine, vol. 35, no. 10, pp. 20-28, 1 Oct. 2020.

• Dual Function Radar Communication Systems [Presenter: Ladi Adeoluwa]
  • ​A. Hassanien, M. G. Amin, E. Aboutanios and B. Himed, “Dual-Function Radar Communication Systems: A Solution to the Spectrum Congestion Problem,” in IEEE Signal Processing Magazine, vol. 36, no. 5, pp. 115-126, Sept. 2019.
  • A. Hassanien, M. G. Amin, Y. D. Zhang and F. Ahmad, “Signaling strategies for dual-function radar communications: an overview,” in IEEE Aerospace and Electronic Systems Magazine, vol. 31, no. 10, pp. 36-45, October 2016.

• Cognitive radar for spectrum sharing [Presenter: Sean Kearney]
  • ​A. F. Martone et al., “Closing the Loop on Cognitive Radar for Spectrum Sharing,” in IEEE Aerospace and Electronic Systems Magazine, vol. 36, no. 9, pp. 44-55, 1 Sept. 2021.
  • P. Stinco, M. Greco, F. Gini and B. Himed, “Cognitive radars in spectrally dense environments,” in IEEE Aerospace and Electronic Systems Magazine, vol. 31, no. 10, pp. 20-27, October 2016.​

MODULE 3 –  MACHINE LEARNING AND DEEP NEURAL NETWORKS

MODULE 4 –  CLASSIFICATION OF RADAR DATASETS WITH DEEP LEARNING

Student Presentations:

• ​SAR Image Classification with Transfer Learning [Presenter: Sean Kearney]
  • ​Z. Huang, Z. Pan and B. Lei, “What, Where, and How to Transfer in SAR Target Recognition Based on Deep CNNs,” in IEEE Transactions on Geoscience and Remote Sensing, vol. 58, no. 4, pp. 2324-2336, April 2020.

• Transfer Learning with Novel Optimization Function [Presenter: Eddie Hackett]
  • Y. Xu and H. Lang, “Ship Classification in SAR Images With Geometric Transfer Metric Learning,” in IEEE Transactions on Geoscience and Remote Sensing, vol. 59, no. 8, pp. 6799-6813, Aug. 2021.

• Transfer Learning in Dynamic Environments [Presenter:  Emre Kurtoglu]
  • C. Yang, Y. -m. Cheung, J. Ding and K. C. Tan, “Concept Drift-Tolerant Transfer Learning in Dynamic Environments,” in IEEE Transactions on Neural Networks and Learning Systems, 2021.

• Data Augmentation and Training SAR Imagery with Limited Data [Presenter: Ladi Adeoluwa]
  • B. Lewis, M. Levy, J. Nehrbass, T. Scarnati, E. Sudkamp, E. Zelnio, “Machine Learning Techniques for SAR Data Augmentation (Preprint),” in Deep Neural Network Design for Radar Applications, Ed. S.Z. Gurbuz, IET, 2020.
  • T. Scarnati and B. Lewis, “A deep learning approach to the Synthetic and Measured Paired and Labeled Experiment (SAMPLE) challenge problem,” Proc. SPIE 10987, Algorithms for Synthetic Aperture Radar Imagery XXVI, 109870G, May 14, 2019.

• Techniques for Exploiting Synthetic Data in Classification of Real Data [Presenter:  Deepak Elluru]
  • N. Inkawhich, M. .J. Inkawhich, E.K. Davis, U.K. Majumder, E. Tripp, C. Capraro, and Y. Chen, “Bridging a gap in SAR-ATR:  Training on Fully Synthetic and Testing on Measured Data,” IEEE Journal of Selected Topics in Applied Earth Observations and Remove Sensing, Vol. 14, 2021.

MODULE 5 –  PHYSICS-AWARE DEEP LEARNING 

Student Presentations:
​

• Graph Neural Networks for Particle Momentum Estimation in the CMS Trigger System [Presenter:  Emre Kurtoglu]
  • Presentation of work conducted as part of Google Summer of Code (2021), Machine Learning for Science (ML4SCI) – Mentor: Dr. Sergei Gleyzer​

• Physics-Based Generative Adversarial Models for Image Restoration   [Presenter: Ladi Adeoluwa]
  • J. Pan, J. Dong, Y. Liu, J. Zhang, J. Ren, J. Tang, Y.-W. Tai and M-H. Yang, “Physics-based generative adversarial models for image restoration and beyond,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021.

• Physics-Aware Machine Learning for Geosciences and Remote Sensing [Presenter: Deepak Elluru] 
  • ​G. Camps-Valls, D. H. Svendsen, J. Cortes-Andres, A. Mareno-Martinez, et. al., “Physics-Aware Machine Learning for Geosciences and Remote Sensing,” IEEE Geoscience and Remote Sensing Symposium, July, 2021.
  • Z. Huang, C. O. Dumitru and J. Ren, “Physics-Aware Feature Learning of Sar Images with Deep Neural Networks: A Case Study,” 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, 2021, pp. 1264-1267.​

• Physics-Aware GAN-based Generation of 3D Data [Presenter: Sean Kearney]
  • Y. Xie, E. Franz, M. Chu, N. Thuerey, “TempoGAN:  A Temporally Coherent, Volumetric GAN for Super-Resolution Fluid Flow,” ACM Transactions on Graphics, Vol. 37, Iss. 4, Aug. 2018.

• Physics-Based Neural Network Architecture Design [Presenter: Eddie Hackett]
  • N. Muralidhar, L. Bu, Z. Cao, et. al., “PhyNet: Physics Guided Neural Networks for Particle Drag Force Prediction in Assembly” in Proc.  SIAM International Conference on Data Mining, 2020.
  • R. Stewart and S. Ermon, “Label-free Supervision of Neural Networks with Physics and Domain Knowledge,” in Proc. of 31st AAAI Conference on Artificial Intelligence, 2017.

MODULE 6 –  SEQUENTIAL MODELING AND CONTINUAL LEARNING

Student Presentations:

• Continual Learning in Radar Micro-Doppler [Presenter: Deepak Elluru]
  • ​D. Lee, H. Park, T. Moon and Y. Kim, “Continual Learning of Micro-Doppler Signature-Based Human Activity Classification,” in IEEE Geoscience and Remote Sensing Letters, vol. 19, pp. 1-5, 2022, Art no. 3502405, doi: 10.1109/LGRS.2020.3046015.
• Continual Learning in GAN-based Image Synthesis [Presenter: Ladi Adeoluwa]
  • M. Zhai, L. Chen, F. Tung, J. He, M. Nawhal and Greg Mori. “Lifelong GAN: Continual Learning for Conditional Image Generation.” 2019 IEEE/CVF International Conference on Computer Vision (ICCV) (2019): 2759-2768.
• Incremental Classification [Presenter: Eddie Hackett]
  • C. Guo, H. Zhou, “Synthetic aperture radar target identification based on incremental kernel extreme learning machine,” Proc. SPIE 10806, Tenth International Conference on Digital Image Processing (ICDIP 2018), 108060L (9 August 2018)
• Iterative Re-Training in SAR Classification [Presenter: Sean Kearney]
  • A. Ahmadibeni, B. Jones, D. Smith, and A. Shirkhodaie, “Dynamic Transfer Learning from Physics-Based Simulated SAR Imagery for Automatic Target Recognition,” In Dynamic Data Driven Applications Systems: Third International Conference, DDDAS 2020, Boston, MA, USA, October 2-4, 2020, Proceedings. Springer-Verlag, Berlin, Heidelberg, 152–159.
• Continual Learning for Activity Recognition [Presenter: Emre Kurtoglu]
  • J. Ye, P. Nakwijit, M. Schiemer, S. Jha, and F. Zambonelli, “Continual Activity Recognition with Generative Adversarial Networks,” ACM Trans. Internet Things 2, 2, Article 9 (March 2021)

MODULE 7 –  COGNITIVE RADAR HARDWARE DESIGN

Student Presentations:

• Reconfigurable RF Transceiver Design [Presenter: Deepak Elluru]
  • ​ A. Egbert, A. Goad, C. Baylis, A. F. Martone, B. H. Kirk and R. J. M. Ii, “Continuous Real-Time Circuit Reconfiguration to Maximize Average Output Power in Cognitive Radar Transmitters,” in IEEE Transactions on Aerospace and Electronic Systems.
  • C. Baylis, R. J. Marks, A. Egbert and C. Latham, “Artificially Intelligent Power Amplifier Array (AIPAA): A New Paradigm in Reconfigurable Radar Transmission,” 2021 IEEE Radar Conference (RadarConf21), 2021.
• Bayesian Framework for Optimal Cognitive Radar Tracking [Presenter: Emre Kurtoglu]
  • ​K. L. Bell, C. J. Baker, G. E. Smith, J. T. Johnson and M. Rangaswamy, “Cognitive Radar Framework for Target Detection and Tracking,” in IEEE Journal of Selected Topics in Signal Processing, vol. 9, no. 8, pp. 1427-1439, Dec. 2015.
  • G. E. Smith et al., “Experiments with cognitive radar,” 2015 IEEE 6th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), 2015.
• Cognitive Radar with Multi-Functional Reconfigurable Antenna Arrays (MRAAs) [Presenter:  Sean Kearney]
  • ​A. C. Gurbuz, R. Mdrafi and B. A. Cetiner, “Cognitive Radar Target Detection and Tracking With Multifunctional Reconfigurable Antennas,” in IEEE Aerospace and Electronic Systems Magazine, vol. 35, no. 6, pp. 64-76, 1 June 2020.
• Cognitive Radar Experiments with CODIR Testbed [Presenter:  Ladi Adeoluwa]
  • ​R. Oechslin, U. Aulenbacher, K. Rech, S. Hinrichsen, S. Wieland and P. Wellig, “Cognitive radar experiments with CODIR,” International Conference on Radar Systems (Radar 2017), 2017.
  • R. Oechslin, P. Wellig, U. Aulenbacher, S. Wieland and S. Hinrichsen, “Cognitive Radar Performance Analysis with Different Types of Targets,” 2019 IEEE Radar Conference (RadarConf), 2019.
• Deployed Canadian Cognitive Radar Example [Presenter: Eddie Hackett]
  • ​Ponsford, A.; McKerracher, R.; Ding, Z.; Moo, P.; Yee, D. “Towards a Cognitive Radar: Canada’s Third-Generation High Frequency Surface Wave Radar (HFSWR) for Surveillance of the 200 Nautical Mile Exclusive Economic Zone.” Sensors 2017, 17, 1588.

MODULE 8 –  COGNITIVE PROCESS MODELING AND IMPLEMENTATION OF A PERCEPTION-ACTION CYCLE

Lecture 10: Cognitive Process Modeling

Student Presentations:

• Levels of Cognition [Presenter: Eddie Hackett]
  • Horne, C., Ritchie, M. and Griffiths, H. (2018), “Proposed ontology for cognitive radar systems.” IET Radar Sonar Navig., 12: 1363-1370.
• Partially Observable Markov Decision Processes (POMDPs) for Cognitive Radar [Presenter:  Sean Kearney]
  • ​A. Charlish and F. Hoffmann, “Anticipation in cognitive radar using stochastic control,” 2015 IEEE Radar Conference, 2015, pp. 1692-1697.
  • A. Charlish, K. Bell and C. Kreucher, “Implementing Perception-Action Cycles using Stochastic Optimization,” 2020 IEEE Radar Conference (RadarConf20), 2020, pp. 1-6.
• Layers of Cognition [Presenter:  Deepak Elluru]
  • S. Brüggenwirth, M. Warnke, C. Bräu, S. SimonWagner, T. Müller, P. Marquardt, and F. Rial, “Sense Smart, Not Hard: A Layered Cognitive Radar Architecture“, in Topics in Radar Signal Processing. London, United Kingdom: IntechOpen, 2018 [Online]. 
• Relationship Between Cognitive Radar and Radar Resource Management [Presenter: Ladi Adeoluwa]
  • A. Charlish, F. Hoffmann, C. Degen and I. Schlangen, “The Development From Adaptive to Cognitive Radar Resource Management,” in IEEE Aerospace and Electronic Systems Magazine, vol. 35, no. 6, pp. 8-19, 1 June 2020.
• Language in Cognitive Radar [Presenter: Emre Kurtoglu]
  • A. Wang and V. Krishnamurthy, “Signal Interpretation of Multifunction Radars: Modeling and Statistical Signal Processing With Stochastic Context Free Grammar,” in IEEE Transactions on Signal Processing, vol. 56, no. 3, pp. 1106-1119, March 2008.