Kongu Engineering College

About: Kongu Engineering College is a based out in . It is known for research contribution in the topics: Computer science & Cluster analysis. The organization has 2001 authors who have published 1978 publications receiving 16923 citations.

A hybrid scheme using packet marking and logging for IP traceback

Abstract: IP traceback is a mechanism for tracing IP packets back to their sources. Tracing mechanisms include packet marking and logging. Log based traceback has the ability to backtrack a single packet by logging each packet at intermediate nodes in the networks. Marking based traceback helps to embed the path information of the intermediate nodes in the packets and the embedded information is used by a victim to reconstruct the attack path. Recent researches show that the performance of hybrid methods comprising logging and marking are appreciable as they help to traceback a single attack packet with less storage overhead on routers. In this study, we use a hybrid approach based on marking and logging to traceback single attack packet with less storage and traceback overhead on routers. We show this through a mathematical analysis. We also evaluate the traceback accuracy of our system and other hybrid approaches. Additionally, the simulation results are also presented to verify the effectiveness of the proposed system.

A Theoretical Implementation: Agriculture-Food Supply Chain Management using Blockchain Technology

Abstract: The traceability of Agriculture food supply chain management is important to ensure the food safety. It also increases the customer satisfaction and peer-to-peer productivity. The centralized data storage makes it more difficult to assure quality, rate and origin of the products. So we are in need of a decentralized system where transparency is available which makes people from the producers to consumers satisfaction. Blockchain technology, which is a digital technology that allows us to acquire traceability and transparency in the supply chain. Making use of this technology actually improves the community between different stakeholders and farmers. The properties of blockchain essentially provides increased capacity, better security, immutability, minting, faster settlement and full traceability of stored transactions records. This paper presents a fully decentralized blockchain based traceability that enables to build blocks for agriculture that continuously integrate with IoT devices from provider to consumer. To implement, we introduced “Provider-Consumer Network” -a theoretical end to end food traceability application. The objective is to create distributed ledger that is accessible by all users in the network that in turn brings transparency.

Development of solar photovoltaic inverter with reduced harmonic distortions suitable for Indian sub-continent

Abstract: In this paper, the power quality improvement techniques with respect to various modulation algorithms for the solar Photovoltaic (PV) inverter are reviewed. In power quality, harmonics are an important concern in all the utility sectors. The factors that contribute the harmonic distortion on distribution systems include increased application of capacitors and non-linear devices. In order to improve the power quality and also to maintain stable power supply performance, an inverter topology with harmonic reduction techniques is required. The design and development of solar Photovoltaic inverter suitable for the Indian sub-continent is proposed and reviewed in terms of survey, simulation and experimental results. The proposed multi-stage inverter provides the advantage of reduced harmonic distortions and suitable for standalone and grid connected systems. The reduction of harmonics is governed by proper switching sequences required for the inverter switches. Modified multicarrier modulation techniques are developed in a single chip controller. A 3 kWp solar PV plant with multistage inverter system is implemented and as per the results, the quality of power is increased and achieved the desired output voltage inspite of variations in the solar PV.

Intrusion detection of distributed denial of service attack in cloud

Abstract: Security issue in cloud environment is one of the major obstacle in cloud implementation. Network attacks make use of the vulnerability in the network and the protocol to damage the data and application. Cloud follows distributed technology; hence it is vulnerable for intrusions by malicious entities. Intrusion detection systems (IDS) has become a basic component in network protection infrastructure and a necessary method to defend systems from various attacks. Distributed denial of service (DDoS) attacks are a great problem for a user of computers linked to the Internet. Data mining techniques are widely used in IDS to identify attacks using the network traffic. This paper presents and evaluates a Radial basis function neural network (RBF-NN) detector to identify DDoS attacks. Many of the training algorithms for RBF-NNs start with a predetermined structure of the network that is selected either by means of a priori knowledge or depending on prior experience. The resultant network is frequently inadequate or needlessly intricate and a suitable network structure could be configured only by trial and error method. This paper proposes Bat algorithm (BA) to configure RBF-NN automatically. Simulation results demonstrate the effectiveness of the proposed method.