Privacy-Preserving Indoor Trajectory Matching with IoT Devices.

Area of Science:

• Computer Science
• Cybersecurity
• Data Science

Background:

• Internet of Things (IoT) devices widely use Wi-Fi for trajectory acquisition.
• Indoor trajectory matching for people monitoring and analysis faces privacy challenges due to cloud computation reliance.
• Existing methods lack robust privacy-preserving mechanisms for sensitive trajectory data.

Purpose of the Study:

• To propose a novel trajectory-matching calculation method supporting ciphertext operations for enhanced privacy.
• To address data integrity issues caused by missing values in indoor trajectory datasets.
• To evaluate the feasibility and effectiveness of the proposed privacy-preserving method.

Main Methods:

• Utilized hash algorithms and homomorphic encryption for secure private data handling.
• Employed correlation coefficients to determine actual trajectory similarity on ciphertexts.
• Implemented mean, linear regression, and K-Nearest Neighbors (KNN) algorithms for imputing missing ciphertext values.

Main Results:

• The proposed method enables secure indoor trajectory matching through ciphertext operations.
• Data imputation algorithms achieved over 97% accuracy in complementing missing values on ciphertexts.
• Demonstrated high feasibility and effectiveness in practical applications with minimal calculation time and accuracy loss.

Conclusions:

• The developed trajectory-matching method effectively balances privacy preservation with analytical accuracy.
• Ciphertext-based data imputation is a viable solution for handling incomplete indoor trajectory data.
• The approach offers a practical and secure solution for indoor trajectory analysis in IoT environments.