The Internet-of-Things (IoT) produces and transmits enormous amounts of data. Extracting valuable information from this enormous volume of data has become an important consideration for businesses and research. However, extracting information from this data without providing privacy protection puts individuals at risk. Data has to be sanitized before use, and anonymization provides solution to this problem. Since, IoT is a collection of numerous different devices, data streams from these devices tend to vary over time thus creating varied data streams. However, implementing traditional data stream anonymization approaches only provide privacy protection for data streams that have predefined and fixed attributes. Therefore, conventional methods cannot directly work on varied data streams. In this work, we propose K-VARP (K-anonymity for VARied data stream via Partitioning) to publish varied data streams. K-VARP reads the tuple and assigns them to partitions based on description, and all tuples must be anonymized before expiring. It tries to anonymize expiring tuple within a partition if its partition is eligible to produce a K-anonymous cluster. Otherwise, partition merging is applied. In K-VARP we propose a new merging criterion called R-likeness to measure similarity distance between tuple and partitions. Moreover, flexible re-using and imputation free-publication is implied in K-VARP to achieve better anonymization quality and performance. Our experiments on a real datasets show that K-VARP is efficient and effective compared to existing algorithms. K-VARP demonstrated approximately three to nine and ten to twenty percent less information loss on two real datasets, while forming a similar number of clusters within a comparable computation time.

IoT devices are capable of capturing physiological measures, location and activity information, hence sharing sensed data can lead to privacy implications. Data anonymization provides solution to this problem; however, traditional anonymization approaches only provide privacy protection for data stream generated from a single entity. Since, a single entity can make use of multiple IoT devices at an instance, IoT data streams are not fixed in nature. As conventional data stream anonymization algorithms only work on fixed width data stream they cannot be applied to IoT. In this work, we propose an anonymization algorithm for publishing IoT data streams. Our approach anonymizes tuples with similar description in a single cluster under time based sliding window. It considers similarity of tuple when clustering, and provides solution to anonymize tuples with missing value using representative values. Our experiment on real dataset shows that the proposed algorithm publishes data with less information loss and runs faster compared to conventional anonymization approaches modified to run for IoT data streams.

The Internet-of-Things (IoT) has formed a whole new layer of the world built on internet, reaching every connected devices, actuators and sensors. Many organizations utilize IoT data streams for research and development purposes. To make value out of these data streams, the data handling party must ensure the privacy of the individuals. The most common approach to provide privacy preservation is anonymization. IoT data provides varied data streams due to the nature of the individual's preference and versatile devices pool. The conventional single tuple expiration driven sliding window method is not adequate to provide efficient anonymization. Furthermore, minimization of missingness has to be considered for the varied data stream anonymization. Therefore, we propose X-BAND algorithm that utilizes the new expiration-band mechanism for handling varied data streams to achieve efficient anonymization, and we introduce weighted distance function for X-BAND to reduce missingness of published data. Our experiment on real datasets shows that X-BAND is effective and efficient compared to famous conventional anonymization algorithm FADS. X-BAND demonstrated 5% to 11% and 1% to 3% less information loss on real dataset Adult and PM2.5 respectively while performing similar on clustering, comparable to re-using suppression and runtime. Also, the new weighted distance function is effective for reducing missingness for anonymization.

No abstract

This paper designs a scheme, which can quantize the personal privacy based on profile and context, using entropy. It aims to configure the persons privacy in numerical form that can be dynamically adjusted instead of using the traditional simple rule engine . In addition, we design the scenario to design a adjust algorithm to modify the privacy based on numerical values.