Abstract
Blockchain technology has been applied to various applications (e.g., smart buildings and smart cities) that typically run in an environment of smart devices, known as Internet-of-Things (IoT). To support these applications, different blockchain architectures, data structures and consensus algorithms have been proposed, tailored to IoT. One such proposal, appendable-block blockchain, is a promising blockchain framework for use in IoT environments. It provides a scalable data structure that allows parallel insertions between independent nodes. However, it has some limitations, in particular related to the possible eclipse attack by malicious gateways and the lack of consensus for transactions insertion. To solve these issues, we propose a new consensus mechanism for appendable-block blockchains, called context-based consensus. Using context-based consensus, information can be inserted in parallel across devices (called context) while ensuring that light-weight consensus is performed to guarantee that a transaction is well-formed and it is placed in the correct order. We implemented context-based consensus and show that using multiple contexts reduces latency and increases the throughput of transaction insertions when compared to consensus without contexts or using single transaction consensus.