Using Blockchain and Low Power in Smart Cities to internet of thigs applications : a Fog Computing approach.

Abstract

With the advent and popularization of Internet of Things (IoT) devices, new possibilities for applications that use data extracted from the things we use in everyday life arise. Cars, wearables, health sensors, and home appliances will generate unprecedented amounts of data and bring insights that will revolutionize our daily routines. A potential scenario significantly impacted is Smart Cities (SC), which uses devices spread out on a large scale in an urban environment to extract traffic, weather, and equipment maintenance data to obtain insights acting on city management and disaster prevention. The network infrastructure currently available for these network applications uses proprietary communication technologies and is dependent on mobile phone companies. Their systems are proprietary, centralized, isolated from other databases, and constantly exposed to Single Point of Failure (SPOF). IoT applications are still primarily embryonic and do not provide reliable verification of the data source at the edge, as in the case of IoT devices, often with outdated firmware. Our work investigates the use in SC of a composition of Low Power Wide Area Networks (LPWAN) and the popular Personal Area Networks (PAN), independence of mobile network providers, and Low Power consumption. For this, we used development kits with LoRa and BLE to verify the feasibility and possible problems in this integration, and we evaluated the scalability of LoRa using a simulator. Security gaps in IoT Apps in Smart Cities mainly come from the difficulty of knowing and trusting edge devices. The problem of standardizing and updating these devices during their lifetime justifies our search for using tools that support transparency, scalability, reliability, resilience, and implicit requirements of decentralized Blockchain networks that support Smart Contracts. For this, we present a network architecture using Fog Computing and Smart Contracts Blockchain, which, through API gateways, authorizes and authenticates edge communication from IoT devices previously known by their metadata and firmware. To provide standard and link data from Blockchain with existing Web datasets, we use and add new components to ontologies that model Ethereum entities. This approach allows us to use the semantic web for data consumption and linking, which exposes data from Ethereum networks in soft-realtime through middleware. This work investigates the potential use of Fog Computing in SC in Low Power networks, strategies to identify and authenticate IoT devices at the edges using Blockchain and Smart Contract, and consumption and data link of Blockchain with the current web using the Semantic web. The set of these resources used in Fog computing allows searching for a composition of independent SC network infrastructures, Low Power, with reliable information coming from the edges and integrable with other pre-existing data sets. As the main results, we show the limits of the LoRa network, using a simulator in single-gateway and multi-gateway scenarios. We present scenarios of mixed use of traditional using Blockchain as authentication and validation background, by API gateway in Fog Computing architecture, and we present the times in transactions per second of this approach considering signatures and validation of payloads using Ethereum Blockchain. We present a middleware to expose Ethereum data in soft-realtime using ontologies that model Ethereum in the literature and extended by our EthExtras ontology, providing classes and properties for links and queries.The main advances of this work are the models using the Fog Computing paradigm for Smart Cities, where we present its use as a mixing point of LoRa and BLE and the Blockchain API Gateway to validate data from IoT devices. In addition to our Middleware for extracting and consuming Ethereum data in soft real-time using our EthExtras and EthOn vocabulary.