SenseCrypt: A Security Framework for Mobile Crowd Sensing Applications

Area of Science:

• Computer Science
• Information Security
• Mobile Computing

Background:

• Mobile crowd sensing (MCS) leverages ubiquitous mobile devices for data collection, offering advantages over traditional wireless sensor networks.
• Despite its potential, MCS faces significant security and privacy challenges, particularly concerning the protection of sensitive user location data.
• The dynamic nature of Global Positioning System (GPS) sensor states in smartphones exacerbates these location privacy concerns.

Purpose of the Study:

• To propose a novel framework, SenseCrypt, for securing sensitive location information in mobile crowd sensing.
• To address the critical need for robust privacy-preserving mechanisms in MCS environments.
• To ensure the confidentiality, integrity, and non-repudiation of user data within MCS systems.

Main Methods:

• Developed SenseCrypt, a framework employing the K-means algorithm for data clustering and a certificateless aggregate signcryption (CLASC) scheme.
• Integrated spatial coding for efficient data compression and Message Query Telemetry Transport (MQTT) as the messaging protocol.
• Implemented automated annotation and signcryption of sensitive location data.

Main Results:

• The SenseCrypt framework demonstrated low computational cost and minimal communication overhead.
• Evaluations confirmed the framework's robustness against various security threats, including privileged insider, replay, and forgery attacks.
• The proposed solution effectively provides confidentiality, integrity, and non-repudiation for MCS data.

Conclusions:

• SenseCrypt offers a viable solution for enhancing the security and privacy of location data in mobile crowd sensing.
• The framework's efficiency and resilience make it suitable for practical deployment in MCS applications.
• By addressing key security vulnerabilities, SenseCrypt promotes trustworthy data collection and user privacy in mobile sensing ecosystems.