• PDF / 958,443 Bytes
• 4 Pages / 595.276 x 790.866 pts Page_size
• 36 Downloads / 192 Views

DOWNLOAD

REPORT

---

Kaunas University of Technology/Biomedical engineering institute, Kaunas, Lithuania 2 School of Medicine, Democritus University of Thrace, Alexandroupoli, Greece

Abstract— Biosignals recorded using personal health devices and stored in General Data Format (GDF) are vulnerable when the data is transferred, processed and stored to the external servers. The aforementioned vulnerabilities influence data security and user’s privacy. In this paper, we propose modifications of GDF format that enables the encryption both - personal data and biosignals. These modifications do not corrupt the intrinsic structure of the GDF format and allow to encrypt independently the header with personal data and the section of biosignals. The proposed modifications were implemented, embedded and tested in a personal health device – multiparametric scale. The header data and biosignals are encrypted directly in the scale, and saved in the micro-SD card using our modified GDF format. Finally, we present the required resources needed for encryption process. Keywords— Biomedical signals, General Data Format, data security and privacy.

I. INTRODUCTION

We are witnessing the era of personal and wearable health devices, such as smart wristwatches, physical activity meters/pedometers, weight and body composition scales, biopatch monitors, etc. Most of these devices are able to gather and send biosignals together with personal data to cloud services for storage and processing. The data can be potentially intercepted during the transmission and storage. Therefore, security and privacy of such devices is a sensitive issue, restricting the wide use of smart technologies for personal and professional use. Recent vulnerability analysis of wearable devices have shown that 70% of wristwatch firmware data was transmitted without encryption, only 50% of tested devices offered the ability to implement a screen lock [1]. Personal health devices, including wireless body area networks and personal e-Health systems provide benefit to people but there are many security and privacy issues that must be solved [2, 3]. Manufacturers usually provide apps for mobile phones to connect the devices and transfer the collected data to cloud services. Most users’ privacy is considered “safe” by agreeing to terms of usage between service provider and the user. On the change of the agreement, e.g. sharing data to third parties, user must accept or stop using services. When a new sensor is developed, there is a possibility to create a new proprietary file format for storing the data, or to © Springer Nature Singapore Pte Ltd. 2018 H. Eskola et al. (eds.), EMBEC & NBC 2017, IFMBE Proceedings 65, DOI: 10.1007/978-981-10-5122-7_183

choose from well-documented existing file formats. Since usually there is a need to store multimodal signals with multiple sampling rates, the list of available data formats considerably shrinks. An overview of data formats for biomedical signals [4] shows that the best candidate is General Data Format (GDF), described as a superset of best features