Abstract
The increasing popularity of wearable consumer products can play a significant role in the healthcare sector. The recognition of human activities from IoT is an important building block in this context. While the analysis of the generated datastream can have many benefits from a health point of view, it can also lead to privacy threats by exposing highly sensitive information. In this article, we propose a framework that relies on machine learning to efficiently recognise the user activity, useful for personal healthcare monitoring, while limiting the risk of users re-identification from biometric patterns characterizing each individual. To achieve that, we show that features in temporal domain are useful to discriminate user activity while features in frequency domain lead to distinguish the user identity. We then design a novel protection mechanism processing the raw signal on the user’s smartphone to select relevant features for activity recognition and normalise features sensitive to re-identification. These unlinkable features are then transferred to the application server. We extensively evaluate our framework with reference datasets: Results show an accurate activity recognition (87%) while limiting the re-identification rate (33%). This represents a slight decrease of utility (9%) against a large privacy improvement (53%) compared to state-of-the-art baselines.
- [n.d.]. Amazon Elastic Compute Cloud (Amazon EC2). Retrieved from http://aws.amazon.com/ec2.Google Scholar
- [n.d.]. Homomorphic Encryption for Arithmetic of Approximate Numbers. Retrieved from https://github.com/snucrypto/HEAAN.Google Scholar
- [n.d.]. TFHE: Fast Fully Homomorphic Encryption over the Torus. Retrieved from https://tfhe.github.io/tfhe/.Google Scholar
- G. Acs and C. Castelluccia. 2014. A case study: Privacy preserving release of spatio-temporal density in paris. In KDD. 1679--1688.Google Scholar
- D. Anguita, A. Ghio, L. Oneto, X. Parra, and J. L. Reyes-Ortiz. 2013. A public domain dataset for human activity recognition using smartphones. In ESANN.Google Scholar
- D. Aranki and R. Bajcsy. 2015. Private disclosure of information in health tele-monitoring. CoRR abs/1504.07313. https://arxiv.org/abs/1504.07313.Google Scholar
- R. Assam, M. Hassani, and T. Seidl. 2013. Differential private trajectory obfuscation. In MOBIQUITOUS. 139--151.Google Scholar
- E. Ayday and M. Humbert. 2017. Inference attacks against kin genomic privacy. IEEE Secur. Privacy 15, 5 (2017), 29--37.Google Scholar
Digital Library
- C. BenAbdelkader, R. Cutler, and L. Davis. 2002. Stride and cadence as a biometric in automatic person identification and verification. In FG. 372--377.Google Scholar
- S. D. Bersch, D. Azzi, R. Khusainov, I. E. Achumba, and J. Ries. 2014. Sensor data acquisition and processing parameters for human activity classification. Sensors 14, 3 (2014), 4239--4270.Google Scholar
Cross Ref
- Raphael Bost, Raluca Ada Popa, Stephen Tu, and Shafi Goldwasser. 2014. Machine learning classification over encrypted data. IACR Cryptology Eprint Archive 2014 (2014), 331. https://eprint.iacr.org/eprint-bin/cite.pl?entry=2014/331.Google Scholar
- A. Boutet, S. Ben Mokhtar, and V. Primault. 2016. Uniqueness Assessment of Human Mobility on Multi-Sensor Datasets. Research Report. LIRIS UMR CNRS 5205. Retrieved from https://hal.archives-ouvertes.fr/hal-01381986.Google Scholar
- A. Boutet, D. Frey, R. Guerraoui, A.-M. Kermarrec, and R. Patra. 2014. HyRec: Leveraging browsers for scalable recommenders. In Middleware. 85--96.Google Scholar
- L. Breiman. 2001. Random forests. Mach. Learn. 45, 1 (2001), 5--32.Google Scholar
Digital Library
- J. T. Bushberg, J. A. Seibert, E. M. Leidholdt, and J. M. Boone. 2011. The Essential Physics of Medical Imaging. Wolters Kluwer Health. 280 pages.Google Scholar
- I. Y. Cheong, S. Y. An, W. C. Cha, M. Y. Rha, S. T. Kim, D. K Chang, and J. H. Hwang. 2018. Efficacy of mobile health care application and wearable device in improvement of physical performance in colorectal cancer patients undergoing chemotherapy. Clin. Colorect. Cancer 17, 2 (2018), e353--e362.Google Scholar
- Ilaria Chillotti, Nicolas Gama, Mariya Georgieva, and Malika Izabachène. 2016. Faster fully homomorphic encryption: Bootstrapping in less than 0.1 seconds. In ASIACRYPT. 3--33.Google Scholar
- Josep Domingo-Ferrer, David Sánchez, and Jordi Soria-Comas. 2016. Database Anonymization: Privacy Models, Data Utility, and Microaggregation-based Inter-model Connections. Morgan 8 Claypool Publishers.Google Scholar
- Léo Ducas and Daniele Micciancio. 2015. FHEW: Bootstrapping homomorphic encryption in less than a second. In EUROCRYPT. 617--640.Google Scholar
- Cynthia Dwork. 2006. Differential privacy. In Automata, Languages and Programming. Vol. 4052. 1--12.Google Scholar
Digital Library
- P. Eckersley. 2010. How unique is your web browser? In PETS’10. 1--18.Google Scholar
- C. Frindel and D. Rousseau. 2017. How accurate are smartphone accelerometers to identify intermittent claudication? In HealthyIoT. 19--25.Google Scholar
- P. Gard, L. Lalanne, A. Ambourg, D. Rousseau, F. Lesueur, and C. Frindel. 2018. A secured smartphone-based architecture for prolonged monitoring of neurological gait. In HealthyIoT. 3--9.Google Scholar
- Johannes Gehrke, Edward Lui, and Rafael Pass. 2011. Towards privacy for social networks: A zero-knowledge based definition of privacy. In TCC. 432--449.Google Scholar
- Craig Gentry, Amit Sahai, and Brent Waters. 2013. Homomorphic encryption from learning with errors: Conceptually-simpler, asymptotically-faster, attribute-based. In CRYPTO. 75--92.Google Scholar
- R. C. Geyer, T. Klein, and M. Nabi. 2017. Differentially private federated learning: A client level perspective. CoRR abs/1712.07557 (2017). https://arxiv.org/abs/1712.07557.Google Scholar
- O. Goldreich. 2003. Cryptography and cryptographic protocols. Distrib. Comput. 16, 2--3 (2003), 177--199.Google Scholar
- M. Gramaglia and M. Fiore. 2015. Hiding mobile traffic fingerprints with GLOVE. In CoNEXT. 26:1--26:13.Google Scholar
- B. Gregorutti, B. Michel, and P. Saint-Pierre. 2017. Correlation and variable importance in random forests. Stat. Comput. 27, 3 (2017), 659--678.Google Scholar
Digital Library
- T. Gu, L. Wang, H. Chen, X. Tao, and J. Lu. 2011. Recognizing multiuser activities using wireless body sensor networks. IEEE Trans. Mobile Comput. 10, 11 (Nov 2011), 1618--1631.Google Scholar
- S. Guha, M. Jain, and V. N. Padmanabhan. [n.d.]. Koi: A location-privacy platform for smartphone apps. In NSDI. 183--196.Google Scholar
- A. Gupta, T. Stewart, N. Bhulani, Y. Dong, Z. Rahimi, K. Crane, C. Rethorst, and M. S. Beg. 2018. Feasibility of wearable physical activity monitors in patients with cancer. JCO Clinical Cancer Informatics 2 (2018), 1--10.Google Scholar
- M. Haghi, K. Thurow, and R. Stoll. 2017. Wearable devices in medical internet of things: Scientific research and commercially available devices. Healthcare Informatics Research 23, 1 (2017), 4--15.Google Scholar
Cross Ref
- J. Han, J. Pei, and M. Kamber. 2011. Data Mining: Concepts and Techniques. Elsevier.Google Scholar
Digital Library
- Jane Henriksen-Bulmer and Sheridan Jeary. 2016. Re-identification attacks--A systematic literature review. Int. J. Inf. Manage. 36, 6, Part B (2016), 1184--1192.Google Scholar
- Ehsan Hesamifard, Hassan Takabi, and Mehdi Ghasemi. 2017. Cryptodl: Deep neural networks over encrypted data. arXiv:1711.05189. Retrieved from https://arxiv.org/abs/1711.05189.Google Scholar
- Ehsan Hesamifard, Hassan Takabi, and Mehdi Ghasemi. 2019. Deep neural networks classification over encrypted data. InCODASPY. 97--108.Google Scholar
- G. James, D. Witten, T. Hastie, and R. Tibshirani. 2013. An Introduction to Statistical Learning. Vol. 112. Springer.Google Scholar
- D. M. Karantonis, M. R. Narayanan, M. Mathie, N. H. Lovell, and B. G. Celler. 2006. Implementation of a real-time human movement classifier using a triaxial accelerometer for ambulatory monitoring. Trans. Inf. Technol. Biomed. 10, 1 (2006), 156--167.Google Scholar
Digital Library
- Panagiotis Kasnesis, Charalampos Patrikakis, and Iakovos Venieris. 2019. PerceptionNet: A deep convolutional neural network for late sensor fusion. In IntelliSys, 101--119.Google Scholar
- J. Konecný, H. Brendan McMahan, D. Ramage, and P. Richtárik. 2016. Federated optimization: Distributed machine learning for on-device intelligence. CoRR abs/1610.02527 (2016). https://arxiv.org/abs/1610.02527.Google Scholar
- J. Konecný, H. Brendan McMahan, F X. Yu, P. Richtárik, A. Theertha Suresh, and D. Bacon. 2016. Federated learning: Strategies for improving communication efficiency. CoRR abs/1610.05492 (2016). https://arxiv.org/abs/1610.05492.Google Scholar
- Dhanya R. Krishnan, Do Le Quoc, Pramod Bhatotia, Christof Fetzer, and Rodrigo Rodrigues. 2016. IncApprox: A data analytics system for incremental approximate computing. In WWW. 1133--1144.Google Scholar
- Lamberg L. 2001. Confidentiality and privacy of electronic medical records. J. Am. Med. Assoc. 285, 24 (2001), 3075--3076.Google Scholar
Cross Ref
- Ninghui Li, Tiancheng Li, and S. Venkatasubramanian. 2007. t-closeness: Privacy beyond k-anonymity and l-diversity. In ICDE. 106--115.Google Scholar
- Ashwin Machanavajjhala, Daniel Kifer, Johannes Gehrke, and Muthuramakrishnan Venkitasubramaniam. 2007. l-diversity: Privacy beyond k-anonymity. ACM Trans. Knowl. Discov. Data 1, 1, Article 3 (2007).Google Scholar
Digital Library
- Mohammad Malekzadeh, Richard G. Clegg, Andrea Cavallaro, and Hamed Haddadi. 2018. Protecting sensory data against sensitive inferences. In W-P2DS. 2:1--2:6.Google Scholar
- D. Manousakas, C. Mascolo, A. R. Beresford, D. Chan, and N. Sharma. 2018. Quantifying privacy loss of human mobility graph topology. In PETS’18, 5--21.Google Scholar
- R. Masood, B. Zi Hao Zhao, H. J. Asghar, and M. A. Kâafar. 2018. Touch and you’re trapp(ck)ed: Quantifying the uniqueness of touch gestures for tracking. PoPETs’18 2018, 122--142.Google Scholar
- S. Mehrang, J. Pietilä, and I. Korhonen. 2018. An activity recognition framework deploying the random forest classifier and a single optical heart rate monitoring and triaxial accelerometer wrist-band. Sensors 18, 2 (2018), 613.Google Scholar
Cross Ref
- F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, et al. 2011. Scikit-learn: Machine learning in python. Journal of Machine Learning Research 12 (2011), 2825--2830.Google Scholar
Digital Library
- A. PETIT, T. Cerqueus, S. Ben Mokhtar, L. Brunie, and H. Kosch. 2015. PEAS: Private, efficient and accurate web search. In TrustCom.Google Scholar
Digital Library
- Albin Petit, Thomas Cerqueus, Antoine Boutet, Sonia Ben Mokhtar, David Coquil, Lionel Brunie, and Harald Kosch. 2016. SimAttack: Private web search under fire. J. Internet Serv. Appl. 7, 1 (2016), 1--17.Google Scholar
- L. T. Phong, Y. Aono, T. Hayashi, L. Wang, and S. Moriai. 2018. Privacy-preserving deep learning via additively homomorphic encryption. IEEE Trans. Inf. Forens. Secur. 13, 5 (2018), 1333--1345.Google Scholar
Digital Library
- I. M. Pires, N. M. Garcia, N. Pombo, and F. Flórez-Revuelta. 2016. From data acquisition to data fusion: A comprehensive review and a roadmap for the identification of activities of daily living using mobile devices. Sensors 16, 2 (2016), 184.Google Scholar
Cross Ref
- S. J. Preece, J. Y. Goulermas, L. PJ Kenney, D. Howard, K. Meijer, and R. Crompton. 2009. Activity identification using body-mounted sensors--A review of classification techniques. Physiol. Meas. 30, 4 (2009), R1.Google Scholar
Cross Ref
- J. L. Reyes-Ortiz. 2015. Smartphone-based Human Activity Recognition. Springer.Google Scholar
- M. Rushanan, A. D. Rubin, D. F. Kune, and C. M. Swanson. 2014. SoK: Security and privacy in implantable medical devices and body area networks. In S8P. 524--539.Google Scholar
- S. Scalvini, D. Baratti, G. Assoni, M. Zanardini, L. Comini, and P. Bernocchi. 2014. Information and communication technology in chronic diseases: A patient’s opportunity.Google Scholar
- J. Schrack, G. Gresham, and A. Wanigatunga. 2017. Understanding physical activity in cancer patients and survivors: New methodology, new challenges, and new opportunities. Molec. Case Stud. 3, 04 (2017), mcs.a001933. DOI:https://doi.org/10.1101/mcs.a001933Google Scholar
- B. Seref and E. Bostanci. 2016. Opportunities, threats and future directions in big data for medical wearables. In BDAW. 15:1--15:5.Google Scholar
- Muhammad Shoaib, Stephan Bosch, Ozlem Incel, Hans Scholten, and Paul Havinga. 2014. Fusion of smartphone motion sensors for physical activity recognition. Sensors 14, 6 (2014), 10146--10176. DOI:https://doi.org/10.3390/s140610146Google Scholar
- S. Sprager and M. B. Juric. 2015. Inertial sensor-based gait recognition: A review. Sensors 15, 9 (2015), 22089--22127.Google Scholar
Cross Ref
- Latanya Sweeney. 2002. k-anonymity: A model for protecting privacy. Int. J. Uncert. Fuzz. Knowl.-Based Syst. 10, 5 (2002), 557--570.Google Scholar
Digital Library
- Y. Tang and C. Ono. 2016. Detecting activities of daily living from low frequency power consumption data. In MOBIQUITOUS. 38--46.Google Scholar
- F. Tramèr, Z. Huang, J.-P. Hubaux, and E. Ayday. 2015. Differential privacy with bounded priors: Reconciling utility and privacy in genome-wide association studies. In CCS. 1286--1297.Google Scholar
- J. B. Wang, L. A. Cadmus-Bertram, L. Natarajan, M. M. White, H. Madanat, J. F. Nichols, G. X. Ayala, and J. P. Pierce. 2015. Wearable sensor/device (fitbit one) and SMS text-messaging prompts to increase physical activity in overweight and obese adults: A randomized controlled trial. Telemed. E-Health 21, 10 (2015), 782--792.Google Scholar
Cross Ref
- Yue Wang, Xintao Wu, and Donghui Hu. 2016. Using randomized response for differential privacy preserving data collection. In EDBT.Google Scholar
- H. Watanabe, T. Terada, and M. Tsukamoto. 2016. Gesture recognition method based on ultrasound propagation in body. In MOBIQUITOUS. 288--289.Google Scholar
- D. Wood, N. Apthorpe, and N. Feamster. 2017. Cleartext data transmissions in consumer IoT medical devices. In IoT S8P. 7--12.Google Scholar
- Lina Yao, Feiping Nie, Quan Z. Sheng, Tao Gu, Xue Li, and Sen Wang. 2016. Learning from less for better: Semi-supervised activity recognition via shared structure discovery. In Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing. 13--24.Google Scholar
Digital Library
- Z.-K. Zhang, M. C. Y. Cho, C.-W. Wang, Ch.-W. Hsu, C.-K. Chen, and S. Shieh. 2014. IoT security: Ongoing challenges and research opportunities. In SOCA. 230--234.Google Scholar
Index Terms
Privacy-preserving IoT Framework for Activity Recognition in Personal Healthcare Monitoring
Recommendations
Toward privacy in IoT mobile devices for activity recognition
MobiQuitous '18: Proceedings of the 15th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and ServicesRecent advances in wireless sensors for personal healthcare allow to recognise human real-time activities with mobile devices. While the analysis of those datastream can have many benefits from a health point of view, it can also lead to privacy threats ...
Mitigating Bystander Privacy Concerns in Egocentric Activity Recognition with Deep Learning and Intentional Image Degradation
Recent advances in wearable camera technology and computer vision algorithms have greatly enhanced the automatic capture and recognition of human activities in real-world settings. While the appeal and utility of wearable camera devices for human-...
The purpose driven privacy preservation for accelerometer-based activity recognition
Accelerometer-based activity recognition (AAR) attracted a lot of attentions due to the wide spread of smartphones with energy-efficiency. However, since accelerometer data contains individual characteristics; AAR might raise privacy concerns. Although ...






Comments