Radio Frequency Fingerprinting on the Edge

Deep learning methods have been very successful at radio frequency fingerprinting tasks, predicting the identity of transmitting devices with high accuracy. We study radio frequency fingerprinting deployments at resource-constrained edge devices. We use structured pruning to jointly train and sparsify neural networks tailored to edge hardware implementations. We compress convolutional layers by a <inline-formula><tex-math notation="LaTeX">$27.2\times$</tex-math><alternatives><mml:math><mml:mrow><mml:mn>27</mml:mn><mml:mo>.</mml:mo><mml:mn>2</mml:mn><mml:mo>×</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="jian-ieq1-3064466.gif"/></alternatives></inline-formula> factor while incurring a negligible prediction accuracy decrease (less than 1 percent). We demonstrate the efficacy of our approach over multiple edge hardware platforms, including a Samsung Gallaxy S10 phone and a Xilinx-ZCU104 FPGA. Our method yields significant inference speedups, <inline-formula><tex-math notation="LaTeX">$11.5\times$</tex-math><alternatives><mml:math><mml:mrow><mml:mn>11</mml:mn><mml:mo>.</mml:mo><mml:mn>5</mml:mn><mml:mo>×</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="jian-ieq2-3064466.gif"/></alternatives></inline-formula> on the FPGA and <inline-formula><tex-math notation="LaTeX">$3\times$</tex-math><alternatives><mml:math><mml:mrow><mml:mn>3</mml:mn><mml:mo>×</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="jian-ieq3-3064466.gif"/></alternatives></inline-formula> on the smartphone, as well as high efficiency: the FPGA processing time is <inline-formula><tex-math notation="LaTeX">$17\times$</tex-math><alternatives><mml:math><mml:mrow><mml:mn>17</mml:mn><mml:mo>×</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="jian-ieq4-3064466.gif"/></alternatives></inline-formula> smaller than in a V100 GPU. To the best of our knowledge, we are the first to explore the possibility of compressing networks for radio frequency fingerprinting; as such, our experiments can be seen as a means of characterizing the informational capacity associated with this specific learning task.

[1]  Stratis Ioannidis,et al.  Open-World Class Discovery with Kernel Networks , 2020, 2020 IEEE International Conference on Data Mining (ICDM).

[2]  Stratis Ioannidis,et al.  Learn-Prune-Share for Lifelong Learning , 2020, 2020 IEEE International Conference on Data Mining (ICDM).

[3]  Stratis Ioannidis,et al.  More Is Better: Data Augmentation for Channel-Resilient RF Fingerprinting , 2020, IEEE Communications Magazine.

[4]  Stratis Ioannidis,et al.  Exposing the Fingerprint: Dissecting the Impact of the Wireless Channel on Radio Fingerprinting , 2020, IEEE INFOCOM 2020 - IEEE Conference on Computer Communications.

[5]  Stratis Ioannidis,et al.  Deep Learning for RF Fingerprinting: A Massive Experimental Study , 2020, IEEE Internet of Things Magazine.

[6]  Yanzhi Wang,et al.  PatDNN: Achieving Real-Time DNN Execution on Mobile Devices with Pattern-based Weight Pruning , 2020, ASPLOS.

[7]  Niraj K. Jha,et al.  Grow and Prune Compact, Fast, and Accurate LSTMs , 2018, IEEE Transactions on Computers.

[8]  Enrico Mattei,et al.  Feature Learning for Enhanced Security in the Internet of Things , 2019, 2019 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[9]  Stratis Ioannidis,et al.  MAC ID Spoofing-Resistant Radio Fingerprinting , 2019, 2019 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[10]  Stratis Ioannidis,et al.  Spectrum Awareness at the Edge: Modulation Classification using Smartphones , 2019, 2019 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN).

[11]  Stratis Ioannidis,et al.  Finding a ‘New’ Needle in the Haystack: Unseen Radio Detection in Large Populations Using Deep Learning , 2019, 2019 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN).

[12]  James Stankowicz,et al.  Complex Neural Networks for Radio Frequency Fingerprinting , 2019, 2019 IEEE Western New York Image and Signal Processing Workshop (WNYISPW).

[13]  Yasamin Mostofi,et al.  XModal-ID: Using WiFi for Through-Wall Person Identification from Candidate Video Footage , 2019, MobiCom.

[14]  Jieping Ye,et al.  AutoSlim: An Automatic DNN Structured Pruning Framework for Ultra-High Compression Rates , 2019, ArXiv.

[15]  Hao Xu,et al.  Real-Time Single Object Detection on The UAV , 2019, 2019 International Conference on Unmanned Aircraft Systems (ICUAS).

[16]  Bingbing Ni,et al.  Variational Convolutional Neural Network Pruning , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  R. Venkatesh Babu,et al.  Zero-Shot Knowledge Distillation in Deep Networks , 2019, ICML.

[18]  Hao Wang,et al.  Acceleration of LSTM With Structured Pruning Method on FPGA , 2019, IEEE Access.

[19]  Aiqun Hu,et al.  A Robust RF Fingerprinting Approach Using Multisampling Convolutional Neural Network , 2019, IEEE Internet of Things Journal.

[20]  Stratis Ioannidis,et al.  DeepRadioID: Real-Time Channel-Resilient Optimization of Deep Learning-based Radio Fingerprinting Algorithms , 2019, MobiHoc.

[21]  Ali Kara,et al.  Assessment of Features and Classifiers for Bluetooth RF Fingerprinting , 2019, IEEE Access.

[22]  Toon Goedemé,et al.  Super accurate low latency object detection on a surveillance UAV , 2019, 2019 16th International Conference on Machine Vision Applications (MVA).

[23]  Jiashi Feng,et al.  Partial Order Pruning: For Best Speed/Accuracy Trade-Off in Neural Architecture Search , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[24]  Gary Steri,et al.  Comparison of techniques for radiometric identification based on deep convolutional neural networks , 2019, Electronics Letters.

[25]  Jiayu Li,et al.  ADMM-NN: An Algorithm-Hardware Co-Design Framework of DNNs Using Alternating Direction Methods of Multipliers , 2018, ASPLOS.

[26]  Stratis Ioannidis,et al.  ORACLE: Optimized Radio clAssification through Convolutional neuraL nEtworks , 2018, IEEE INFOCOM 2019 - IEEE Conference on Computer Communications.

[27]  Mingjie Sun,et al.  Rethinking the Value of Network Pruning , 2018, ICLR.

[28]  Jiayu Li,et al.  A Unified Framework of DNN Weight Pruning and Weight Clustering/Quantization Using ADMM , 2018, ArXiv.

[29]  Jing Liu,et al.  Discrimination-aware Channel Pruning for Deep Neural Networks , 2018, NeurIPS.

[30]  Sparsh Mittal,et al.  A survey of FPGA-based accelerators for convolutional neural networks , 2018, Neural Computing and Applications.

[31]  Hossein Jafari,et al.  IoT Devices Fingerprinting Using Deep Learning , 2018, MILCOM 2018 - 2018 IEEE Military Communications Conference (MILCOM).

[32]  Jiayu Li,et al.  Progressive Weight Pruning of Deep Neural Networks using ADMM , 2018, ArXiv.

[33]  Andreas S. Tolias,et al.  Generative replay with feedback connections as a general strategy for continual learning , 2018, ArXiv.

[34]  Stratis Ioannidis,et al.  Deep Learning Convolutional Neural Networks for Radio Identification , 2018, IEEE Communications Magazine.

[35]  Jiayu Li,et al.  ADAM-ADMM: A Unified, Systematic Framework of Structured Weight Pruning for DNNs , 2018, ArXiv.

[36]  Houqiang Li,et al.  Improving Deep Neural Network Sparsity through Decorrelation Regularization , 2018, IJCAI.

[37]  Hui Liu,et al.  On-Demand Deep Model Compression for Mobile Devices: A Usage-Driven Model Selection Framework , 2018, MobiSys.

[38]  Yanzhi Wang,et al.  A Systematic DNN Weight Pruning Framework using Alternating Direction Method of Multipliers , 2018, ECCV.

[39]  Vincent C. Gaudet,et al.  Efficient Hardware Realization of Convolutional Neural Networks Using Intra-Kernel Regular Pruning , 2018, 2018 IEEE 48th International Symposium on Multiple-Valued Logic (ISMVL).

[40]  Haichen Shen,et al.  TVM: An Automated End-to-End Optimizing Compiler for Deep Learning , 2018, OSDI.

[41]  Shauna Revay,et al.  Deep Learning for RF Device Fingerprinting in Cognitive Communication Networks , 2018, IEEE Journal of Selected Topics in Signal Processing.

[42]  Sung Ju Hwang,et al.  Lifelong Learning with Dynamically Expandable Networks , 2017, ICLR.

[43]  Yangqing Jia,et al.  High performance ultra-low-precision convolutions on mobile devices , 2017, ArXiv.

[44]  Tony X. Han,et al.  Learning Efficient Object Detection Models with Knowledge Distillation , 2017, NIPS.

[45]  Scott A. Mahlke,et al.  DeftNN: Addressing Bottlenecks for DNN Execution on GPUs via Synapse Vector Elimination and Near-compute Data Fission , 2017, 2017 50th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO).

[46]  Zhiqiang Shen,et al.  Learning Efficient Convolutional Networks through Network Slimming , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[47]  Jianxin Wu,et al.  ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[48]  Xiangyu Zhang,et al.  Channel Pruning for Accelerating Very Deep Neural Networks , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[49]  Scott A. Mahlke,et al.  Scalpel: Customizing DNN pruning to the underlying hardware parallelism , 2017, 2017 ACM/IEEE 44th Annual International Symposium on Computer Architecture (ISCA).

[50]  Rajesh Krishna Balan,et al.  DeepMon: Mobile GPU-based Deep Learning Framework for Continuous Vision Applications , 2017, MobiSys.

[51]  William J. Dally,et al.  SCNN: An accelerator for compressed-sparse convolutional neural networks , 2017, 2017 ACM/IEEE 44th Annual International Symposium on Computer Architecture (ISCA).

[52]  Xin Dong,et al.  Learning to Prune Deep Neural Networks via Layer-wise Optimal Brain Surgeon , 2017, NIPS.

[53]  Cong Xu,et al.  Coordinating Filters for Faster Deep Neural Networks , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[54]  Surya Ganguli,et al.  Continual Learning Through Synaptic Intelligence , 2017, ICML.

[55]  Bernt Schiele,et al.  Zero-Shot Learning — The Good, the Bad and the Ugly , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[56]  Hanan Samet,et al.  Pruning Filters for Efficient ConvNets , 2016, ICLR.

[57]  Zhongfei Zhang,et al.  Doubly Convolutional Neural Networks , 2016, NIPS.

[58]  Yiran Chen,et al.  Learning Structured Sparsity in Deep Neural Networks , 2016, NIPS.

[59]  Tien Dang Vo-Huu,et al.  Fingerprinting Wi-Fi Devices Using Software Defined Radios , 2016, WISEC.

[60]  Alec Wolman,et al.  MCDNN: An Approximation-Based Execution Framework for Deep Stream Processing Under Resource Constraints , 2016, MobiSys.

[61]  Koray Kavukcuoglu,et al.  Exploiting Cyclic Symmetry in Convolutional Neural Networks , 2016, ICML.

[62]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[63]  Srdjan Capkun,et al.  Types and Origins of Fingerprints , 2016, Digital Fingerprinting.

[64]  Song Han,et al.  Learning both Weights and Connections for Efficient Neural Network , 2015, NIPS.

[65]  Hassan Foroosh,et al.  Sparse Convolutional Neural Networks , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[66]  Geoffrey E. Hinton,et al.  Distilling the Knowledge in a Neural Network , 2015, ArXiv.

[67]  Jason Cong,et al.  Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks , 2015, FPGA.

[68]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[69]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[70]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[71]  Benjamin W. P. Ramsey,et al.  Wireless infrastructure protection using low-cost radio frequency fingerprinting receivers , 2015, Int. J. Crit. Infrastructure Prot..

[72]  David A. Clifton,et al.  A review of novelty detection , 2014, Signal Process..

[73]  Kevin W. Sowerby,et al.  Analysis of impersonation attacks on systems using RF fingerprinting and low-end receivers , 2014, J. Comput. Syst. Sci..

[74]  Jizhong Zhao,et al.  GenePrint: Generic and accurate physical-layer identification for UHF RFID tags , 2013, 2013 21st IEEE International Conference on Network Protocols (ICNP).

[75]  Hui Zheng,et al.  Radio frequency fingerprinting based on the constellation errors , 2012, 2012 18th Asia-Pacific Conference on Communications (APCC).

[76]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[77]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..

[78]  Darko Kirovski,et al.  RF Fingerprinting Physical Objects for Anticounterfeiting Applications , 2011, IEEE Transactions on Microwave Theory and Techniques.

[79]  Robert F. Mills,et al.  Radio frequency fingerprinting commercial communication devices to enhance electronic security , 2008, Int. J. Electron. Secur. Digit. Forensics.

[80]  Marco Gruteser,et al.  Wireless device identification with radiometric signatures , 2008, MobiCom '08.

[81]  Robert A. van de Geijn,et al.  Anatomy of high-performance matrix multiplication , 2008, TOMS.

[82]  Oktay Ureten,et al.  Wireless security through RF fingerprinting , 2007, Canadian Journal of Electrical and Computer Engineering.

[83]  Aamer Jaleel,et al.  DRAMsim: a memory system simulator , 2005, CARN.

[84]  D. McNeill,et al.  Frequency offset estimation and correction in the IEEE 802.11a WLAN , 2004, IEEE 60th Vehicular Technology Conference, 2004. VTC2004-Fall. 2004.

[85]  Don Coppersmith,et al.  Matrix multiplication via arithmetic progressions , 1987, STOC.