A Methodology for Data Sustainability in Smart City - Dissertation Technologie

Different sources generate a lot of data of different types, data is sent by the sensors collected, parsed and normalized by centralized in the data collectors. The data of a smart city is generated by the sensors and then they are processed using the mechanisms to be used by end users. But the problem here is the lack of a security tool to secure the smart city against attacks, so we must propose a solution that will fit with all services in the smart city. Concerning the architecture, we can divide it at four levels (data collection level, data processing level, data integration and reasoning level, data control and alerts levels).

Level 1: Data collection:In this level, information collected from sensors and stored for processing. the formats heterogeneous data that are collected are csv, tweets, database schemas, and text messages. The collected formats are then processed using semantic web technologies in order to convert them into a common format. Level 2: Data processing:The main role of this level is to convert the collected heterogeneous information into a common format, RDF is the way to exchange information over the web and it facilitates heterogeneous data sharing and integration for different Smart City services, Different applications use RDF data for intelligent reasoning operations. The information generated at this level will be exploited using semantic knowledge and reasoning rules at the next level.

Level 3: Data integration and Reasoning: The Dempster-Shafer approach is used to combine data from different sensors in smart cities, This approach will help to learn new knowledge.Level 4: Data control and alerts:In this level, we can find all things about alerts warnings Messages.This architecture concerns the process of generating data from sensors but the problem here is the lack of security layer to protect and improve the data that is generated by the smart city services, thats why we must choose a smart security system, but provided that the solution is adaptable with a smart in order to secure the services offered by smart city (smart transport, smart industry, smart environment, smart home …….).Before moving to the presentation of a smart security system based on intrusion detection system as a solution we must classify the attacks that can be found in a smart city Classification of attacks in the smart city according to these four layers (physical layer, Data Link layer, Transport layer, Application layer).

style=”font-weight: 400;”>Figure 2. Data processing in the smart city.3.2 Attacks classifications in the smart city:The abilities of WSNs nodes concerning processing power and battery are the main obstacle in deploying protection mechanisms; WSN is usually considered to be easy to attack. Authors list the most popular attacks on WSNs, and they also describe several countermeasures to prevent them. In this section we list the most popular attacks in the smart city : Attacks in the physical layer: Data tampering, Node tampering, Node replication.Attacks in the Data link layer: sybil, collision, Exhaustion, interrogation, unfairness .Attacks in the transport layer: De-synchrosination,Flooding .Attacks in the Application layer :Deluge , Re-play, overwhelme, Eavesdropping.

3.3 Smart security System as a solution (Intrusion detection system) :Among the characteristics of a smart city, we find system efficiency, developed infrastructure, and others. But the main one is actually its safety. In order to secure services offered by smart city (smart home, smart grid, smart healthcare ) we must propose a smart solution to defend against attacks, so it is necessary to develop a smart security system based on intrusion detection systems.Let\’s start by defining what is an intrusion detection system (IDS),it is a mechanism to identify abnormal or suspicious activity on the target being scanned (a network or a host).

In general, an IDS is a engine that analyzes traffic according to the rules, which describe the traffic to report. IDS are able to detect malware (viruses, worm …), analyze or detect networks, detect attacks by DOS or DDOS.

Accéder à l'échantillon complet