An anomaly can be defined as a data or unwanted observations which can also be an outlier in a dataset. As IoT gather a huge volume of data, there could be many outliers, and it will be a difficult task to classify an outlier among a huge set of data observations. There are many Anomaly detection techniques existing in the current literature, but many of them require massive processing and iterative training, so that an anomaly can be detected. In recent years, a distributed<br>ledger which is maintained by all the peers within a peer-to-peer network, called Blockchain, has been proposed to detect anomalies in IoT networks.
The core idea behind this Blockchain-based anomaly detection is to provide a new de-centralised system which is a trust-based and immutable. Each transaction is represented by a block, and since each block is built on top of the previous block, the immutability has been achieved. Here immutability means it is very difficult to fake/alter a block and very easy to detect any tampering. Using the fork mechanism in the Blockchain technology, anomaly detection in IoT networks can be achieved. As part of Blockchain-based anomaly detection, we propose to collect, share, and enrich the information with other peers in the networks. In this context, a peer can be a satellite or an IoT or a ground station.
Professsor Naveen Chilamkurti, La Trobe University