Blockchain Based Privacy Preservation In Edge Computing Paradigms

The development of edge computing has led to the proliferation of emerging M2 M connections such as smart meters and utilities,health applications,GPS in cars,etc.Malicious intruders and honest but curious users can used private information gathered from smart meters,health gadgets,etc.adversely against the user.Intruders can initiate attacks that can lead to privacy leakage,data loss,and illegal operations on the stored data leading to compromises on data integrity,confidentiality and authenticity.End-to-end security therefore has to be integrated into every edge computing platform to guarantee user privacy.Current privacy protection protocols deployed in edge computing environment are pseudo-decentralized.The protocols involve the use of third party elements which may be involved in server or node registration.This inclusion introduces scalability issues and points of failures.This has the potential to affect availability of network.The principles of blockchain technology were integrated into the design of edge computing with the aim of improving upon the security challenges identified.In a framework which involves a lot of different service providers handling user data,it is difficult to predict where an attack can emanate from prompting users to be suspicious in the exchange of information.However,in a network of untrusted parties,blockchain guarantees a secured network suitable for the devices in the environment under discussion.Blockchain eliminates any element of centralization,ensuring scalability within any network it is integrated into.This has a knock-off on further decreasing delay and eliminating any single point of failure.A key principle of blockchain is transparency.Depending on the type of blockchain network deployed,all or some nodes get to have access to all the information in transactions exchanged on the network to assess,audit and verify if these transactions meet the rules governing the rules established in the network.The transparency feature within blockchain coupled with the need to ensure sensitive information is accessed by only the parties involved in the transactional exchange also present a challenge of access to the information by internal attackers.The key contributions of the thesis are as follows:1)First,an assessment of the current models of authentication schemes in edge computing is carried out to determine the drawbacks in the current models of privacy preservation authentication schemes in Chapter 3.This was to improve upon current trends of authentication in which the yet to be proposed schemes may not readily apply.Using the cryptographic primitives of bilinear pairing based cryptography,a scheme is proposed to improve enhance privacy and provide authentication mechanisms in current authentication models for deployment in edge computing platforms.To improve the application of biometric information in cryptosystems,fuzzy extractor bio-cryptosystem is introduced and integrated into this protocol.The proposed Simulations are carried out in an MCC environment.The protocol is further subjected to cryptanalysis using AVISPA to test it vulnerability to known attacks.Results demonstrate that the proposed protocol is robust against known attacks such as masquerading and replay attacks while guaranteeing user anonymity and untraceability.Compared to other benchmark schemes,the proposed scheme is low on computation and provides additional security features.2)Secondly,in Chapter 4,after making proposals to improve current trends of privacy and authentication schemes,a fully decentralized security protocol termed DecChain is introduced to deal primarily with the bottlenecks associated with pseudo-decentralized current cryptosystems.In the absence of any trust third party entities,DecChain eliminates any scalability and points of failure challenges.As in a typical blockchain network,messages are exchanged as transactions with public keys and chosen identities as the main mode of addressing users on the network.To test the efficiency and practicality of the protocol,DecChain is deployed in an MCC environment and its implementation compared to other benchmark schemes.Simulation results show DecChain executes and converges faster while providing data and private user security.DecChain is robust against man-in-the middle and replay attacks while providing mutual authentication and user anonymity.3)Finally,with the aim of improving upon the drawbacks identified in DecChain,a privacy enhancement scheme(PES)fit for blockchain – edge computing environments is presented.PES eliminates any chance internal adversaries can trace public keys on blockchain – edge networks to users.The dynamic generation of public keys and digital signatures for transactions exchanged on the networks ensures user untraceability and anonymity is never compromised.Simulation results indicate the proposed privacy enhancement scheme is lightweight and robust against all forms of insider attacks.The proposed schemes DecChain and PES presented in this thesis will be vital in providing an end-to-end security in any edge computing paradigm.The schemes are designed to be lightweight and suitable for resource constrained devices while providing improved security features needed for guaranteeing mutual authentication,user and data privacy.