Secure Cloud Storage and Retrieval of Personal Health Data From Smart Wearable Devices With Privacy-Preserving Techniques

Introduction

With the increasing public health awareness, personal health data management is becoming an important component in people's lives (Zeng et al., 2015). Smart wearable devices (SWDs), such as smartwatches, smart bracelets, etc., can collect motion data (such as motion trajectory and status, etc.) and physiological data (such as heart rate and blood pressure, etc.) of the SWD wearer (SWDW) anytime and anywhere. In a personal health data management, the data collected by SWDs are typically organized into structured records. For example, data collected by SWDs show that on June 16, 2023, at 8:15 a.m., the SWDW's heart rate was 80 while located at 31°N, 114°E. This information can be represented as a record <2023, 6, 16, 08, 15, 1, 114, 0, 31, 80>, where 1 represents east longitude and 0 represents north latitude. However, due to the small storage space of SWDs and the risk of accidental damage or loss, the collected data are often automatically transmitted to the paired smartphone through a Bluetooth connection and then uploaded to the cloud to obtain unlimited storage space and use the cloud's data backup and disaster recovery mechanisms to ensure that the data are permanently available. In addition, when the data collected by SWDs are uploaded to the cloud as part of electronic health records, medical institutions, insurance companies, or other health management institutions can access and use the data to provide more personalized health management services (Zeng et al., 2018). However, the data collected by SWDs contain sensitive information about the SWDWs. Once this sensitive information is leaked, it may affect personal image, property safety, and even life safety. Therefore, ensuring the security and privacy of sensitive information collected by SWDs and outsourced to the cloud is very important. Encryption is an effective solution to protect the data collected by SWDs. However, traditional encryption methods (e.g., block encryption) cannot support the most common data operations in the cloud, such as ciphertext retrieval (Cui et al., 2023). Although new ciphertext retrieval schemes have been proposed, they have limitations when applied to personal health data management.