Wavelet-based dynamic and privacy-preserving similitude data models for edge computing
- PDF / 1,176,199 Bytes
- 16 Pages / 595.276 x 790.866 pts Page_size
- 50 Downloads / 147 Views
(0123456789().,-volV)(0123456789(). ,- volV)
Wavelet-based dynamic and privacy-preserving similitude data models for edge computing Philip Derbeko1
•
Shlomi Dolev1 • Ehud Gudes1
Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract The privacy-preserving data release is an increasingly important problem in today’s computing. As the end devices collect more and more data, reducing the amount of published data saves considerable network, CPU and storage resources. The savings are especially important for constrained end devices that collect and send large amounts of data, especially over wireless networks. We propose the use of query-independent, similitude models for privacy-preserving data release on the end devices. The conducted experiments validate that the wavelet-based similitude model maintains an accuracy compared to other state-of-the-art methods while compressing the model. Expanding on our previous work (Derbeko et al. in: Cyber security cryptography and machine learning-second international symposium, CSCML 2018, Beer Sheva, Israel, 2018) we show how wavelet-based similitude models can be combined and ‘‘subtracted’’ when new end devices appear or leave the system. Experiments show that accuracy is the same or improved with a model composition. This data-oriented approach allows further processing near the end devices in a fog or a similar edge computing concept. Keywords Datasets Neural networks Gaze detection Text tagging Wavelet
1 Introduction We consider typical end devices that generate data such as sensors, IoT devices, vehicles, smartphones, etc. Usually, the end devices collect data and transmit it to the cloud servers [2–4]. In many cases, the latency and the bandwidth of this connection are limiting. Edge computing goal is to solve the latency and bandwidth issue by bringing the processing and storage closer to the end devices [5, 6]. Edge computing does not define the specifics of the edge devices, only that they should be close to the end devices. An interesting implementation of edge computing is fog computing, where the processing and storage are performed in a cloud computing that is close to the edge & Philip Derbeko [email protected] Shlomi Dolev [email protected] Ehud Gudes [email protected] 1
Computer Science Department, Ben-Gurion University of Negev, Beer-Sheva, Israel
device, thus the name ‘‘fog’’ (low cloud) as opposite to ‘‘cloud’’ [4, 7]. Data release Release of the previously collected data is called data release. In many cases the collected data on the end devices contains private information, which should not be shared outside the device. To ensure that, a plethora of privacy preserving methods were developed [8–12]. Privacy of the data release The privacy-preserving data release always presented several challenges: a choice of a privacy providing method, deciding on what is Personally Identifiable Information (PII), the right balance between privacy and utilization, etc. With the growth of end devices, the scale of devices has be
Data Loading...
