Research On Identity Authentication And Access Control Of Wireless Medical Monitoring Network

The aging population has been the basic national conditions of China in the 21 st century.It is very important to establish a perfect medical and health system for the elderly.At present,the wireless medical monitoring network based on the combination of wireless body area network,mobile communication technology and data cloud system,is an important way to effectively solve the problem of healthcare for the elderly.In the wireless medical monitoring network,the physiological data of patients monitored by various sensors is forwarded to the cloud server through the base station or gateway.By reading the patient’s information in the cloud,medical staff can diagnose and treat the health status of patients in time.However,wireless medical monitoring network transmits sensitive data over an open channel,while storing data in an incompletely trusted cloud.In the process of data transmission and access,there are security risks such as data intercepted,illegally accessed,and even tampered by attackers.If those are not properly handled,it may lead to the leakage of users’ privacy information,and even seriously threaten the lives of patients.Therefore,it is especially necessary to design a mechanism to meet the data security and privacy protection requirements of the wireless medical monitoring network.In this regard,this paper proposes corresponding solutions from the identity authentication and access control of the wireless medical monitoring network.The main research contents and contributions of the paper are as follows:In wireless medical monitoring network,the legitimate identity of both the user terminal and the cloud server needs to be ensured the legality of the identity,so that attackers cannot impersonate the user terminal or access the data illegally.This is to ensure the safety of medical data of patients in the medical monitoring network.The traditional identity authentication protocols are not suitable for the wireless medical monitoringnetwork with limited resources because they have large overhead in computing and storage.The ECG signal has individual uniqueness in the wireless medical monitoring network and is a conventional human physiological detection data,and it does not require additional overhead.Therefore,combining the ECG features with the traditional cryptography-based identity authentication scheme,we designed a lightweight mutual identity authentication protocol between user terminal and the cloud server in wireless medical monitoring network.In the protocol,the set encoding is used the to extract the ECG signal,and the ECG feature is used to realize the identity authentication of the cloud server by the user terminal.While the GPS identity authentication protocol is used to complete the challenge response authentication of the user terminal by the cloud server.The protocol consists of three phases: initialization,registration,and authentication.The analysis results of the correctness,security and accuracy of the protocol shows that the protocol meets the identity authentication requirements of the wireless medical monitoring network.Because the original ECG signal collected is often accompanied by various interference noises,it needs to be denoised by complex pre-processing.At present,there is no effective pre-processing method in the wireless medical monitoring network with limited resources,which can ensure both the accuracy of ECG signals and low time complexity and space-complexity.In this regard,combining the convolutional neural network in artificial intelligence technology and data signature technology,we designed a new identity authentication protocol based on ECG signal.The identity authentication protocol including the convolutional neural network can simultaneously extract and classify the original ECG signals,which can eliminate complicated preprocessing work.The integrity of the information transmitted is guaranteed by digital signatures.The key of the protocol is that the collected ECG signal is transformed into two-dimensional gray ECG by cloud server,which is input into the designed convolutional neural network to learn ECG features for registration.The authentication process is to compare the learning results of the newly collected ECG signals with the registered data through the convolutional neural network.The experimental results show that the scheme improves the accuracy of identity authentication comparing to several other authentication schemes.The medical data collected will eventually be stored in the cloud for access by different entities such as doctors,patients’ relatives,researchers,etc.In order to prevent attackers from illegally accessing the patient’s physiological data or even tampering with it,and realize fine-grained access control among different entities by user terminals,we proposed an access control scheme that combines the attribute encryption algorithm with the control of the operation authority.In the protocol,different access strategies and operation privileges are designed for data requesters with different attributes and attribute revocation with timestamps is implemented.The access control and operation control of cloud data by user terminals are realized.Through the analysis of the correctness,security and performance of the scheme,the scheme meets the security requirements of data sharing in the wireless medical monitoring network.