Improved Gentry–Halevi's fully homomorphic encryption‐based lightweight privacy preserving scheme for securing medical Internet of Things

Abstract

Internet of Medical Things (IoMT) solutions have proliferated rapidly in the COVID-19 pandemic era. The smart medical sensors capture real-time data from remote patients and communicate it to medical servers in a secure and privacy-preserving manner. It is a herculean challenge to guarantee security and privacy in Medical IoT applications. Hence, an improved Gentry-Halevi's fully homomorphic encryption-based (IGHFHE) lightweight privacy preserving user authentication scheme is proposed in this work. The scheme is proposed with an integer matrix computation strategy for securing data computation with privacy protection. It adopts the translation process of Gentry-Halevi's fully homomorphic encryption process for performing homomorphic addition and multiplication, then encrypt an integer matrix modulo that represents a positive integer. Extensive informal investigation and simulation of the proposed IGHFHE scheme shows that it is more resistant to well-known attacks for preventing authentication breaches. Also, the proposed IGHFHE scheme reduced computational and storage overhead by 4.98% and 5.78% respectively on average in comparison to other prevailing schemes. INTRODUCTIONThe pervasiveness of artificial intelligence, 5G/6G wireless communication, and smart sensors is enabling the Internet of Things (IoT) in healthcare to become a new dimension in the field of emerging technology. 1 As in Figure 1, the remote patient monitoring (RPM) applications of IoMT platform are layered as perception, network, data and the application. The perception/sensor layer medical devices such as insulin pumps, cardiac activity monitoring devices, continuous positive airway pressure (CPAP) devices, and other associated equipment's plays an indispensable role in the lifestyle of the patients by connecting in a remote manner. 2 These sensors sense the patient's condition instantly and communicate it to the medical professionals through the network layer components. Then the RPM system in the application layer can able to obtain the real time-data at any time without demanding them to visit a healthcare facility or a hospital at any time. It greatly paves the path for reducing hospital capacity overflow and reduces hospital stays, particularly during pandemics such as COVID. 3 The sensors attached with the patients transmit the physiological signs such as respiration rate, blood pressure, blood glucose level, heart rate variation, blood oxygen saturation level, and an electrocardiogram (ECG) patterns using the networking components such as gateway Ramalingam Praveen and Parameswaran Pabitha are equally contributing authors.

A Survey on Key Agreement and Authentication Protocol for Internet of Things Application

“…In order to provide secure and privacy-preserving communication in remote patient data monitoring, which has become imperative in the COVID-19 pandemic era, an improved lightweight privacy-preserving authentication scheme was proposed. The scheme is proven resistant to attacks and reduces computational and storage overhead [4].…”

“…State of the art for authentication in recent research includes but is not limited to: Lightweight Authentication Protocols. Due to the limited resources of IoT devices, various fields such as medical IoT [4], [5], [6], 5G-enabled IoT [7], 6G-based cellular networks [8], and industrial IoT [9] need lightweight authentication and key agreement (AKA) protocols. Current AKA protocols provide lightweight authentication by minimizing the computational and communication overhead [10], [11], [12].…”

Hybrid homomorphic‐asymmetric lightweight cryptosystem for securing smart devices: A review

“…18,19 In addition, the key size is 126 bits with poor hash function support. 20 Further, modified HE models, a hybrid homomorphic scheme based on Rivest, Shamir, and Adleman (RSA), 21 enhanced Gentry-Helevi's fully homomorphic encryption (FHE), 22 federated learning with FHE, 23 improved Martino HE 24 and so forth are introduced by incorporating different techniques or doing alternations on the standard models, however, their computational time was more due to complicated calculations.…”