A Novel Technique for Blackhole Discernment Employing Big Data Analytics in IoT Networks

Internet of Things is connection of equipment, machine, software and everything around us to make smart things which have unique IP address to distinguish from other things. In Internet of Things everyday objects have capabilities to identify, sense and process the data so they can communicate with each other generating huge volumes of data. But data security is the major issue as when communication takes place the data can be altered through malicious attacks taking place on network while the data is being transmitted from different devices. So, in this paper a novel totalitarian approach is proposed for detection of blackhole attack which drops the packets being transmitted in the network leading to hacking and loss of data. So, in order to eliminate this DoS attack from the network proposed technique can be used where occurrence of malicious infected path results in initiation of request sent by any local node to all of its neighbor’s node for verifying the presence of the malicious path in their routing tables which has a high percentage of usage. We have used the approach of a modified routing table along with analysis of the threshold value generated by nodes which generates huge volume of big data and variation in amount of data generated by malicious nodes we can identify suspicious links before the attack is initiated and starts disturbing within the network. This is done to establish legitimacy of the node and on the corroboration of a malicious attack a blacklisting message is broadcasted to exclude the colluder nodes from the network. Results and analysis of this technique is been done on the basis of packet delivery ratio, throughput, power consumption and congestion control. Along with this comparative analysis of this technique is done with the other existing techniques proposed by various renowned researchers for identifying these blackhole attacks taking place in IoT Networks. The proposed technique makes use of a modified routing table and utilizes Big Data analytics to accurately ascertain the blackhole nodes in the IoT network. The detection and isolation of colluder nodes is being done accurately with minimalistic consumption of resources as opposed to some of the existing techniques, which may require high power consumption, synchronization, as well as other resources thereby making these techniques resource intensive. Since this technique makes use of multivariate parameter, it is more resilient to faults. Our approach also aims to alleviate the security and privacy concerns associated with big data thereby being a potential flag winner for the organizations.