Since the introduction of blockchain by Satoshi Nakamoto, its applications have rapidly expanded across various sectors such as finance, healthcare, supply chain management, and IoT. Blockchain’s decentralized, immutable, secure, and transparent structure enables efficient and trustworthy data exchange without relying on centralized authorities. Conventional systems like Electronic Health Records (EHR) and Public Key Infrastructure (PKI) suffer from single-point failures, limited scalability, and weak transparency, which blockchain can effectively overcome.

This thesis is divided into three parts. The first part develops a blockchain-based decentralized application (dApp) for EHR, enabling secure medical data sharing between patients and doctors using MetaMask wallets and Ethereum-based payments. The second part proposes a blockchain-based PKI (BC-PKI) employing a lightweight smart contract and Delegated Proof of Stake (DPoS) consensus to mitigate attacks such as DoS, DDoS, and Man-in-the-Middle. The third part enhances BC-PKI efficiency by introducing a clustering and trust-based Certificate Authority (CA) selection mechanism to reduce computational overhead. Nodes are evaluated based on trust, validation, and response times, ensuring efficient and secure communication. The implementation uses the Ethereum platform, Truffle Suite, Remix IDE, and Solidity for smart contract development and deployment.

The emergence of Blockchain-based dApps has addressed two fundamental issues present in centralized application systems: single-point failure and security risks using inherent features. The features of decentralization, immutability, and transparency in dApps enable them to effectively address the aforementioned issues. However, blockchain-based applications also face some challenging issues such as security concerns and computation overhead, which need to be effectively addressed. The objective of the current thesis is to address the issues present in both centralized application systems and blockchain-based applications. The three primary contributions of the thesis are stated in Chapters 3, 4, and 5 respectively. Chapter 3 presents a basic blockchain-based decentralized application for storing and exchanging data of EHR among doctors and patients. This is a basic implementation to understand the working principle of blockchain-based applications. This dApp helps in removing the single-point failure of the centralized application system. However, the dApp does not provide any effective way to deal with cyber attacks present in blockchain-based applications. In Chapter 4, a security solution for blockchain-based PKI is developed. This BC-PKI prevents many popular threats like DoS, DDoS, MITM, 51%, Injection, Routing, and Eclipse attacks. Unlike conventional PKI, the developed PKI provides an effective way to identify malicious Certificate Authorities using its smart contract. In addition, this provides an equal opportunity to all available nodes to get Certificate Authority status. This is achieved by setting the threshold value of each node in order to become the Certificate Authority. If a node exceeds the predefined threshold, it will no longer be allowed to become a Certificate Authority. The DPoS consensus algorithm utilized in our PKI reduces timing complexity by avoiding excessive computational capacity at the nodes’ end. The consensus mechanism and the adopted smart contract make the developed PKI affordable for lightweight applications. However, the developed PKI does not focus on reducing the computational overhead related to the Certificate Authority selection process. To address this issue, Chapter 5 proposes a PKI that utilizes clustering approaches based on validation time, response time, and trust. The developed PKI searches for Certificate Authorities on the nodes of the chosen cluster, rather than searching on all participant nodes, thereby reducing the search space for the Certificate Authority selection process. This work also focuses on the trust calculation of every participating node. The node having a higher trust value, lower validation time, and lower response time has a higher probability of becoming the Certificate Authority for a transaction. For every successful transaction, the trustworthiness of the Certificate Authority is increased, while the trust value is decreased for every unsuccessful transaction. All the three works of this thesis are implemented in the Ethereum blockchain environment (GETH) in association with Ganache Truffle Suite. The author has noticed that GETH has a scalability issue in terms of the number of transactions and the number of nodes. An increase in transactions and nodes may affect the efficiency of the developed PKIs. However, the works mentioned above have certain limitations stated as follows: (1) The developed PKI calculates trust based on successful and unsuccessful transactions. Considering communication quality factors such as data transmission rate and data delay rate in trust calculation can make the proposed consensus algorithm more efficient. (2) The implementations have been tested up to 100 nodes due to commercial constraints of the Ethereum platform. Future work will study larger network traffic to identify effectiveness. (3) Network energy consumption and computation effort were not meticulously studied and can be addressed in future works. (4) All designs are implemented in the Ethereum platform; future work should explore other blockchain platforms to study comparability, scalability, and effectiveness. In the future, the author intends to develop a browser plug-in for implementing BC-PKI as mentioned in this thesis to identify malicious Certificate Authorities and will also address all aforementioned limitations with feasible solutions.