Optimized machine learning model for detecting DDoS attacks

Scalable and energy efficient cluster based Intrusion detection against denial of service attacks in WSNs


  •   The paper will discuss how spatial or temporal information can be used to detect and locate multiple adversaries, even if they are using different node identities.
  •   The article will describe a mechanism for detecting DoS attacks using a cluster-based approach that is both scalable and energy-efficient.
  •   The system's performance will be measured using different network parameters, including detection rate, false positive rate, packet delivery ratio, overhead, energy consumption, and average delay of packets.
  •   According to the results, the system has a high hit rate and is more reliable in detecting and locating multiple adversaries than previous systems.

  • Description

Wireless communication technologies have rapidly advanced, leading to numerous real-world applications that will revolutionize robotic exploration, commercial ventures, military operations, battlefield surveillance, border control, and health-related fields. The open nature of the network makes it vulnerable to DoS attacks and can greatly impact the behavior of Wireless Sensor Networks (WSN). The verification of nodes using crypto analysis is difficult due to the limited energy capability of the nodes. Therefore, it is crucial that we find effective solutions to mitigate these vulnerabilities and ensure the secure and reliable operation of wireless networks.