Health Service Oriented Security Authentication And Privacy Data Share For Wearable Device

Wearable intelligent equipment is integrated into the user's clothes,or attached to the skin surface,or directly implanted in the body,which is one of the most promising future technology leading the development of today's digital tide.Wearable devices are closely integrated with human beings,and provide additional functions,mainly for personal services,and even enhance human instincts.Especially,for the future of wearable technology,some equipment doesn't "wear",but "implant"in the human body to treat the diseases,monitor status,improve the function of the human body,even provide people with the ability they didn't have before.The rapid development of wearable devices for health services has brought unprecedented living convenience to human beings.At the same time,it also infringes personal privacy and even threatens personal safety.In the open architecture of wireless body area network,a security authentication mechanism must be provided between device node and data center.Many authenti-cation methods based on equipment's physical characteristics or user's biological char-acteristics have been proposed.However,the former ignores the uniqueness of user's biological characteristics and fails to resist compromise attack,and the latter fails to against impersonation attack.In this paper,we propose a new two factor authentication protocol based on biological characteristics and physical characteristics of equipment,which utilizes the double uniqueness of interpulse interval(IPI)and physically unclon-able function(PUF).It is more secure to band the equipment to a person as a whole in authentication and improve the output bias caused by the asymmetry of D flip flop.Wearable devices involve spatial data and streaming data,users'location privacy needs to be protected when publishing these aggregate data.Existing differentially pri-vate spatial decomposition based data release methods usually adopt uniform budgeting scheme,which allocating the same privacy budget to each dividing grid and ignoring the influence of actual data query for budget allocation.In this paper,we propose a differentially private data release method based on Fibonacci budgeting scheme,which optimizes the allocation of privacy budget and enhances the utility of sanitized data.Furthermore,data utility has been improved by post-processing and threshold determi-nation.Streaming data of wearable device provides a solid base for policy making,disease prediction in the data mining analysis.However,streaming data publishing also brings users the risk of privacy disclosure.To address this problem,the concept of differential privacy has been applied to streaming data release.The released stream directly affects the results of mining analysis,in order to improve its utility,existing differentially private streaming data publishing methods usually optimize the utility of sanitized data using Kalman filter,which is unsuitable for nonlinear system,so a utility optimization scheme of differentially private streaming data based on unscented Kalman filter is proposed in this paper.The proposed method uses sampling method to approximate nonlinear distribution,and it improves the utility of released streaming data,while guaranteeing a certain level of privacy.The utility and validity of proposed scheme is evaluated by four real-world datasets.For users,health information is relatively sensitive,any inappropriate information exposure may disclose user privacy.The open application environment and the limited resources of equipment have brought great challenges for data security and privacy protection.Aiming at data security,this paper proposes a two factor authentication protocol for wearable devices based on PUF and IPI.In view of the privacy protection of spatial data and streaming data,we propose a differentially private release scheme separately.The systematically research and solve for these problems provide important theoretical support for the security protection mechanism of wearable devices,and have important application value for the popularization of wearable devices.