Multi-agent systems (MASs) have received tremendous attention from scholars in different disciplines, including computer science and civil engineering, as a means to solve complex problems by subdividing them into smaller tasks. The individual tasks are allocated to autonomous entities, known as agents. Each agent decides on a proper action to solve the task using multiple inputs, e.g., history of actions, interactions with its neighboring agents, and its goal. The MAS has found multiple applications, including modeling complex systems, smart grids, and computer networks. Despite their wide applicability, there are still a number of challenges faced by MAS, including coordination between agents, security, and task allocation. This survey provides a comprehensive discussion of all aspects of MAS, starting from definitions, features, applications, challenges, and communications to evaluation. A classification on MAS applications and challenges is provided along with references for further studies. We expect this paper to serve as an insightful and comprehensive resource on the MAS for researchers and practitioners in the area.

Blockchain offers immutability, transparency, and security in a decentralised way for many applications, including finance, supply chain, and the Internet of Things (IoT). Due to its popularity and widespread adoption, it has started to process an enormous number of transactions, placing an ever-growing demand for storage. As the technology gains more popularity, the storage requirements of blockchain will increase, necessitating storage optimisation solutions. Proposed solutions for blockchain storage efficiency range from reducing the degree of data replication to redacting or compressing data. Each of these storage optimisation categories involves a complex interplay with the timing of blockchain data processing and mining, yet no existing survey analyses these dimensions. This article surveys the state-of-the-art blockchain storage optimisations and categorises them into replication-based, redaction-based, and content-based optimisations. Replication-based optimisations focus on reducing duplication of blockchain data shared among participants after committing data on the blockchain ledger. Redaction-based optimisations allow users to modify or delete data already committed on the ledger in various ways, while content-based optimisations compress data before or after committing it to the ledger. We analyse and evaluate these solutions in the aspects of security, decentralisation, and scalability. We present the advantages and disadvantages of the existing blockchain storage optimisations and comprehensively compare them. Additionally, we discuss the opportunities and challenges for future work to optimise blockchain storage.

In cryptocurrencies, privacy of users is preserved using pseudonymity. However, it has been shown that pseudonymity does not result in anonymity if a users transactions are linkable. This makes cryptocurrencies vulnerable to deanonymization attacks. The current solutions proposed in the literature suffer from at least one of the following issues: (1) requiring a trusted thirdparty entity, (2) poor performance, and (3) incompatible with the standard structure of cryptocurrencies. In this paper, we propose Unlinkable Coin (UCoin), a secure mixbased approach to address these issues. In UCoin, the link between the input (payer) and output (payee) addresses in a transaction is broken. This is done by mixing the transactions of multiple users into a single aggregated transaction in which the output addresses have been secretly shuffled. In our protocol design, we first develop HDCnet, a secure shuffling protocol that enables a group of users to anonymously publish their data. Then, we deploy the proposed HDCnet protocol in the UCoin architecture (as a mixing unit) to generate the aggregate transactions. We show that UCoin (1) does not rely on a trusted thirdparty, (2) can mix 50 transactions in 6.3 seconds that is 18% faster than the current solutions, and (3) is fully compatible with the architecture of cryptocurrencies.

Distribution, security, and immutability have led to the great success of blockchain in many applications, while contributing to major increases in ledger size. The storage challenge is one of the major barriers to the adoption of blockchain in the Internet of Things (IoT), which consists of many resource constrained devices. In this paper, we propose Multi-Level Distributed Caching (MLDC) for blockchain storage optimisation which reduces data replication based on data access patterns in a decentralised manner. For storage optimisation of data-centric blockchains, MLDC introduces a hierarchical storage class (SC), in which every node is assigned to an SC with its own Access Frequency (AF) threshold based on node availability. To reduce the number of replications shared among participant nodes, each node in a SC continues to remove unaccessed data from local storage based on a threshold time determined by the AF threshold of the SC, while maintaining all block hashes for consistency. Eventually, all nodes in MLDC store the most frequently accessed data in their local storage, so MLDC effectively reduces the storage and query costs while minimising network overhead. We also analyse the security of MLDC and quantitatively evaluate its performance for both the uniform access and exponentially decaying access patterns. The evaluation was carried out on a representative blockchain simulator with 15 storage nodes. Our results from 11 hours of experiments producing 6667 blocks and 39997 transactions show good performance for MLDC. The results of the experimentation for the exponentially decaying assess pattern show that MLDC can reduce the total storage cost by 83% compared to conventional blockchain systems, while maintaining blockchain consistency and data availability with a slight increase in network overhead and query cost.