Machine Learning Methods Assessment for Denial of Service Detection in Wireless Sensor Networks

Quincozes, Silvio E.; Kazienko, Juliano F.

doi:10.1109/wf-iot48130.2020.9221146

Cited by 21 publications

(14 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This algorithm is a decision tree classi er technique that consists of an open-source java implementation constructing the trees from the training dataset [14]. It predicts the classes of unknown samples using the entropy concept.…”

Section: J48 Classi Ermentioning

confidence: 99%

See 1 more Smart Citation

Detection and Analysis of Flooding Attacks in Wireless Sensor Networks

Gebremariam

Panda

Indu

2022

Preprint

View full text Add to dashboard Cite

Wireless sensor networks are distributed networks and randomly deployed in harsh environments. The scattered nature of the sensor nodes in an unattended environment makes them vulnerable to various types of attacks. Multiple attacks like flooding attacks degrade the network performance and lifetime by sending various requests to malicious nodes. This paper proposes the detection and analysis of flooding attacks in mesh wireless sensor networks using the Blue tooth communication package MATLAB 2020a toolbox. The flooding attack disabled the relay nodes during the routing path between sources and destination nodes for each simulation scenario. The simulations are conducted by the number of sensors variation for the random generator. After simulating the meshed nodes, the statistical results are the message delivery chance, average hope count, and critical relays. The opportunity of message delivery having a route between sources and destinations is 88.64%. The effectiveness of the proposed system for detection and analysis of flooding attacks is evaluated by random forest machine earing techniques with an average detection rate and accuracy of 99.83%, which is greater than by 3.8%.

show abstract

Section: J48 Classi Ermentioning

confidence: 99%

“…S. E. Quincozes and J. F. Kazienko [14] Assessed the detection and classi cation accuracy of ooding, black hole, and gray hole DoS WSNs attacks using machine learning approaches. The detection of ooding attacks is 98.07% which is less than the new approach, 99.83%.…”

Section: Ac = (5)mentioning

confidence: 99%

Detection and Analysis of Flooding Attacks in Wireless Sensor Networks

Gebremariam

Panda

Indu

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Quincozes et al (14) described that the major goal was to evaluate the efficacy of the ML (Machine Learning) techniques for detecting various DoS (Denial of Service) attacks WSNs. WSN-DS dataset was applied to stimulate the presented techniques concerning accuracy and speed.…”

Section: Literature Reviewmentioning

confidence: 99%

DoS Attack Detection Mechanism in Wireless Sensor Networks

Sharma

Shajahan

Elangovan

et al. 2022

Salud, Ciencia y Tecnología

View full text Add to dashboard Cite

Wireless sensor networks (WSNs) are not like traditional networks in terms of their characteristics. Unlike WSN, in classic networks, networking mechanisms have decision-making power with reference to the management of an incoming packet only based on its internet protocol (IP) destination address. Assailant nodes can activate a denial of service (DoS) attack after entering the network. This research work focuses on tracing these collusive nodes by applying a trust-based scheme. The trust-based scheme includes measuring the degree of trust among all nodes. Nodes with minimum trust are designated as malicious nodes. Trust is measured based on the overall packets transferred during the time slot allotted. The node forwarding maximal packets and exploiting minimal assists will be tagged as malicious. The nascent architecture was deployed in NS2, and the outcomes were analysed based on particular performance metrics. To calculate trust, the overall packets forwarded by nodes in the allotted slot were considered. The node that transmits the largest number of packets and uses negligible assets was declared a vindictive node. This task implementation presents the approach in the NS2 software and analyses the results based on certain metrics.

show abstract

“…Em 2016, a IoT está em alta. Contudo, governos e empresas notam a intensificação de códigos maliciosos e de ataques de negação de serviço, do inglês Distributed Denial of Service (DDoS), percebendo a necessidade de combate-los através da segurança da informação [Quincozes and Kazienko 2020]. Em 2018, a versão 5.0 do protocolo MQTT é disponibilizada.…”

Section: Evolução Históricaunclassified

Desvendando a Camada de Aplicação na Internet das Coisas: Teoria, Prática e Tendências

Quincozes

Quincozes²,

Kazienko

2021

Minicursos Da ERSI-RJ 2021 - VII Escola Regional De Sistemas De Informação Do Rio De Janeiro

View full text Add to dashboard Cite

The Internet of Things (IoT) is a reality that is already part of our daily lives. In this context, it is essential to understand the fundamentals of prominent protocols, especially at the application layer level, as well as aspects for implementing solutions for IoT. In this document, we introduce the application layer and its protocolos, namely MQTT, CoAP, MQTT-SN, XMPP and DDS. In addition, MQTT and CoAP protocols are presented in practice, as both have been gaining prominence in the literature and the industry, representing two alternatives for the IoT application layer. Finally, potential applications and trends in areas, namely: security, machine learning, computational paradigms, cyberphysical systems and user interfaces are addressed in order to provide insights to readers and encourage them to the development of IoT applications. ResumoA Internet das Coisas, do inglês, Internet of Things (IoT) é uma realidade que já faz parte do nosso cotidiano. Nesse contexto, é fundamental a compreensão dos fundamentos de protocolos proeminentes, especialmente a nível de camada de aplicação, bem como aspectos para a implementação de soluções na IoT. Neste documento, é apresentada a camada de aplicação e seus protocolos MQTT, CoAP, MQTT-SN, XMPP e DDS. Além disso, os protocolos MQTT e CoAP serão apresentados de maneira prática, visto que ambos vem ganhando destaque tanto na literatura quanto na indústria, representando duas alternativas para a camada de aplicação da IoT. Por fim, as potenciais aplicações e tendências nas áreas de segurança, aprendizado máquina, paradigmas computacionais para IoT, sistemas ciber-físicos e interfaces do usuário serão abordadas.

show abstract

Machine Learning Methods Assessment for Denial of Service Detection in Wireless Sensor Networks

Cited by 21 publications

References 16 publications

Detection and Analysis of Flooding Attacks in Wireless Sensor Networks

Detection and Analysis of Flooding Attacks in Wireless Sensor Networks

DoS Attack Detection Mechanism in Wireless Sensor Networks

Desvendando a Camada de Aplicação na Internet das Coisas: Teoria, Prática e Tendências

Contact Info

Product

Resources

About