Call for Papers

Special Issue on Privacy Preserving Analytics forIoT Streaming Systems

Internet of Things (IoT) systems are widely integrated in modern life, from industry applications to public transportation and personal health care. IoT systems continuously monitor the cyber physical and collect vast amount of data. Various analytics are developed to analyze IoT data and discover the business or societal values for their applications, e.g., home surveillance, smart readers, and location services. While we as a society greatly benefit from the utility of IoT data, an alarming concern of privacy breaching arises, i.e., data owners’ social and personal interests are revealed unknowingly. To address the conundrum of extracting utility from IoT data and protecting its privacy, the design of analytics is expected to jointly combine novel solutions in the security and machine learning fields. For example, encrypting generated data and enabling homographic computing can protect the data integrity; obfuscating data and disturbing the learning process with statistical noises are shown effective to guarantee the differential privacy. Moreover, due to the growing complexity and size of IoT systems, the implementation of analytical solutions needs to be scalable and adaptive to the streaming nature of IoT systems, i.e., data are continuously generated.

This special issue aims to gather high quality research papers in the broad area of privacy preserving analytics for IoT streaming systems. The focus of this SI is to address new algorithms, advance software development, novel system architecture, and critical applications that lead to the optimal tradeoff between data utility and privacy for IoT streaming systems.
Topics of interest include, but are not limited to:
- Algorithms: data obfuscation schemes, private machine learning, and homomorphic algorithms for IoT streams
- Architecture: special hardware design, scalable edge/Fog network, and distributed learning systems for IoT streams
- Applications: case studies of specific IoT streaming systems that implement privacy-preserving analytical solutions
- Privacy/Security: novel privacy metrics, data communication protocols, and encryption schemes for IoT streams
- Benchmarking: Innovative IoT performance benchmarking and profiling, and modeling tools

Big Data Mining and Analytics is a young journal from Tsinghua University Press and consciously grows its contributors and readers. It features on technologies to enable and accelerate big data discovery. Submitted articles must not have been previously published or currently submitted for journal publication elsewhere. As an author, you are responsible for understanding and adhering to our submission guidelines. You can access them on the IEEE Xplore at https://ieeexplore.ieee.org/xpl/aboutJournal.jsp?punumber=8254253. Please submit your paper to Manuscript Central at https://mc03.manuscriptcentral.com/bdma.

All papers will be peer-reviewed and selected based on their “originality” and merit, such as relevance to the BDMA themes, as per requirement of BDMA. Once the papers are finalized, the special issue will be published based on the IEEE BDMA publication schedule

Timeline
Submission due: Nov 15, 2018
Reviews notification: Mar 1, 2019
Notification of Acceptance: June 1, 2019
Final Version: June 15, 2019
Publication due: August 1, 2019

Guest Editorial Team
Prof. Lydia Y. Chen, TU Delft, Netherlands, E-mail: lydiaychen@ieee.org
Prof. Xuan (Shawn) Guo, University of North Texas, USA, E-mail: Xuan.Guo@unt.edu
Dr. Robert Birke, ABB Research, Switzerland, E-mail: rober.birke@ch.abb.com
Prof. Laizhong Cui, Shenzhen University, China, E-mail: cuilz@szu.edu.cn