Sharing of Network Flow Data across Organizations using Searchable Encryption

• Published in: 2021 23rd International Conference on Control Systems and Computer Science (CSCS) pp. 189 - 196
• Main Authors: Florea, Iulia Maria, Ghinita, Gabriel, Rughinis, Razvan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-05-2021
• Subjects: Encoding; Hidden Vector Encryption; Inner Product Encryption; Medical conditions; Organizations; Pandemics; Predicate Based Encryption; Privacy; Scalability; Searchable Encryption; Taxonomy
• ISSN: 2379-0482
• DOI: 10.1109/CSCS52396.2021.00038

Given that an increasingly larger part of an organization's activity is taking place online, especially in the current situation caused by the COVID-19 pandemic, network log data collected by organizations contain an accurate image of daily activity patterns. In some scenarios, it may be useful to share such data with other parties in order to improve collaboration, or to address situations such as cyber-security incidents that may affect multiple organizations. However, in doing so, serious privacy concerns emerge. One can uncover a lot of sensitive information when analyzing an organization's network logs, ranging from confidential business interests to personal details of individual employees (e.g., medical conditions, political orientation, etc). Our objective is to enable organizations to share information about their network logs, while at the same time preserving data privacy. Specifically, we focus on enabling encrypted search at network flow granularity. We consider several state-of-the-art searchable encryption flavors for this purpose (including hidden vector encryption and inner product encryption), and we propose several customized encoding techniques for network flow information in order to reduce the overhead of applying state-of-the-art searchable encryption techniques, which are notoriously expensive.