Showing papers by "Mayank Dave published in 2020"

SAPDA: Secure Authentication with Protected Data Aggregation Scheme for Improving QoS in Scalable and Survivable UWSNs

Abstract: Security is one of the main objectives while designing protocols for underwater wireless sensor networks (UWSN), since the sensors in UWSN are vulnerable to malicious attack. So it becomes easy for opponents to manipulate the communication channel of UWSN and its nodes. Authentication and data integrity play important roles in the context of security to make network scalable and survivable. Hence in this paper, a secure authentication and protected data aggregation method for the cluster based structure of UWSN is proposed as because cluster based arrangement produces a concise and stable network. In this technique, the cluster head in each cluster is authenticated by the gateway to ensure that all the clusters are being handled by valid nodes. Also, the data being communicated in the network will be securely handled to ensure that it will not get compromised during network operations. In this way, the security of all the nodes is ensured to maintain safe network communication. The proposed technique improves the data reliability in the network by reducing the energy consumption and delay. Here, the proposed method is moreover compared with the state of the art techniques to prove the validity and effectiveness.

Voting‐based intrusion detection framework for securing software‐defined networks

Abstract: Software-defined networking (SDN) is an emerging paradigm in enterprise networks because of its flexible and cost-effective nature. By decoupling control and data plane, SDN can provide various defense solutions for securing futuristic networks. However, the architectural design and characteristics of SDN attract several severe attacks. Distributed Denial of Service (DDoS) is considered as a major destructive cyber attack that makes the services of controller unavailable for its legitimate users. In this research paper, an intrusion detection framework is proposed to detect DDoS attacks against SDN. The proposed framework relies on voting based ensemble model for the attack detection. Ensemble model is a combination of multiple machine learning classifiers for prediction of final results. In this research paper, we propose and analyze three ensemble models named as Voting-CMN, Voting-RKM, and Voting-CKM particularly to benchmarking datasets like UNSW-NB15, CICIDS2017, and NSL-KDD, respectively. For validation of the proposed models, a cross validation technique is used with the prediction algorithms. The effectiveness of proposed models is evaluated in terms of prominent metrics (accuracy, precision, recall, and F measure). Experimental results indicate that the proposedmodels achieve better performance in terms of accuracy as compared to other existing models.

DDoS Attacks and Defense Mechanisms Using Machine Learning Techniques for SDN

Abstract: Distributed denial of service (DDoS) attack is one of the most disastrous attacks that compromises the resources and services of the server. DDoS attack makes the services unavailable for its legitimate users by flooding the network with illegitimate traffic. Most commonly, it targets the bandwidth and resources of the server. This chapter discusses various types of DDoS attacks with their behavior. It describes the state-of-the-art of DDoS attacks. An emerging technology named “Software-defined networking” (SDN) has been developed for new generation networks. It has become a trending way of networking. Due to the centralized networking technology, SDN suffers from DDoS attacks. SDN controller manages the functionality of the complete network. Therefore, it is the most vulnerable target of the attackers to be attacked. This work illustrates how DDoS attacks affect the whole working of SDN. The objective of this chapter is also to provide a better understanding of DDoS attacks and how machine learning approaches may be used for detecting DDoS attacks.

Morse-Comm: Mobile Application Environment for Visually-Impaired and Neuro-Muscular Disabled People Using Morse Code Conversion

Abstract: In a world where the next great invention is expected to appear on mobile phone screens, visually impaired people and people with Neuro-Muscular disabilities have been left behind. This paper describes a new Android application, "Morse-Comm" that improves the mobile phone experience for the visually impaired people or people with Neuro-Muscular disabilities who can barely speak by presenting a slate on which they can draw sequences of dots and dashes i.e. Morse code through "gestures" and use the application for intercommunication. The application is also used as Morse keyboard for writing documents on a personal computer. "Morse-Comm" provides better, more private and more efficient inter-communication mechanism for the blind as well as the people with Neuro-Muscular disorders.

Privacy Preserving Data Aggregation in Wireless Body Sensor Network

Abstract: With the expeditious expansion of the Internet of Things (IoT), an individual device’s privacy has become a huge concern for the industry, especially if medically sensitive information is being transferred. Safe and secure passage of this information to the intended destination is one of the main features of IoT. Currently, many schemes have been proposed to guarantee safe passage and secure Data Aggregation of individual device’s message(s). This work investigates present privacy concerns and data aggregation issues in IoT applications such as Wireless Body Sensor Network (WBSN) and goes through different Privacy-preserving Data Aggregation (PPDA) methods presented in Wireless Sensor Network (WSN)/IoT. Finally, this paper proposes a privacy-preserving data aggregation scheme more suitable for medically sensitive data of WBSN. The proposed scheme takes account of important differences between the WSN and WBSN, such as data redundancy and the role of individual sensors.

Analysis and Detection of Brain Tumor Using U-Net-Based Deep Learning

Abstract: Brain tumor could be a life threatening disease and the survival rate of such disease is low It is generally the abnormal growth of cells inside the brain Early and accurate detection of the brain tumor is very difficult The manual segmentation of the brain tumor extent from 3D MRI (Magnetic Resonance Imaging) volumes is a time consuming process and depends a lot on the operator’s experience The automatic tumor segmentation has the potential to decrease lag time between diagnosis tests and the treatment for the same Hence, there is a high demand of time and memory efficient, and reliable computer algorithms to do this accurately and quickly In this paper, we first highlight limitations of the image processing based solutions and subsequently present a novel deep learning based technique The proposed technique relies on U-Net based Deep Convolutional Networks for the automatic detection and analysis of brain tumors

Improving the Effectiveness of Moving Target Defenses by Amplifying Randomization

Abstract: Cyber security is a major concern nowadays. In the arms race between the attackers and the defenders, attackers claim control by employing deception techniques against defense technologies. To deal with this, there is a cyber defense strategy Moving Target Defense (MTD) which constantly changes the target surface by using counter-deception techniques. It forces the attacker to operate in unpredictable environment. This paper proposes an efficient approach that integrates diversity-based MTD at the OS level with shuffle-based MTD. This integration increases attacker efforts and makes the attack costly. Further, an algorithm for amplifying randomization in the system is proposed with the aim of improving the effectiveness.

Proactive DDoS Attack Mitigation in Cloud-Fog Environment using Moving Target Defense.

Abstract: Distributed Denial of Service (DDoS) attacks are serious cyber attacks and mitigating DDoS attacks in cloud is a topic of ongoing research interest which remains a major security challenge. Fog computing is an extension of cloud computing which has been used to secure cloud. Moving Target Defense (MTD) is a newly recognized, proactive security defense that can be used to mitigate DDoS attacks on cloud. MTD intends to make a system dynamic in nature and uncertain by changing attack surface continuously to confuse attackers. In this paper, a novel DDoS mitigation framework is presented to support Cloud-Fog Platform using MTD technique (CFPM). CFPM applies migration MTD technique at fog layer to mitigate DDoS attacks in cloud. It detects attacker among all the legitimate clients proactively at the fog layer and isolate it from innocent clients. CFPM uses an effective request handling procedure for load balancing and attacker isolation procedure which aims to minimize disruption to cloud server as well as serving fog servers. In addition, effectiveness of CFPM is evaluated by analyzing the behavior of the system before and after attack, considering different possible scenarios. This approach is effective as it uses the advantage of both MTD technique and Fog computing paradigm supporting cloud environment.