TY - GEN
T1 - Cybersecurity Fusion
T2 - 2023 IEEE Global Communications Conference, GLOBECOM 2023
AU - Javadpour, Amir
AU - Ja'fari, Forough
AU - Taleb, Tarik
AU - Ahmadi, Hamid Reza
AU - Benzaïd, Chafika
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Mafia, also known as Werewolf, is a game of uncertainty between two teams, which aims to eliminate the other team's players from the game. The similarities between detecting the Mafia members in this game and botnet detection in a computer network motivate us to solve the botnet detection problem using this game's winning strategies. None of the state-of-the-art researches have used the Mafia game strategies to detect the network's malicious nodes. In this paper, we first propose the Mafia detection strategies, which are applied using linear relation and reinforcement learning techniques. We then use the suggested strategies in a network infected by the Mirai botnet, using Mininet, to evaluate the performance of botnet detection. The average results show that the suggested strategies are 11% more accurate than the existing ones for the Mafia game. Additionally, the true positive and true negative detection rates of a network modeled by the proposed Mafia game are 71% and 91%, respectively.
AB - Mafia, also known as Werewolf, is a game of uncertainty between two teams, which aims to eliminate the other team's players from the game. The similarities between detecting the Mafia members in this game and botnet detection in a computer network motivate us to solve the botnet detection problem using this game's winning strategies. None of the state-of-the-art researches have used the Mafia game strategies to detect the network's malicious nodes. In this paper, we first propose the Mafia detection strategies, which are applied using linear relation and reinforcement learning techniques. We then use the suggested strategies in a network infected by the Mirai botnet, using Mininet, to evaluate the performance of botnet detection. The average results show that the suggested strategies are 11% more accurate than the existing ones for the Mafia game. Additionally, the true positive and true negative detection rates of a network modeled by the proposed Mafia game are 71% and 91%, respectively.
KW - Botnet detection
KW - Distributed denial of service (DDoS) attacks
KW - Mafia game
KW - Network security
KW - Reinforcement learning
KW - cybersecurity
UR - https://www.scopus.com/pages/publications/85187393449
U2 - 10.1109/GLOBECOM54140.2023.10437968
DO - 10.1109/GLOBECOM54140.2023.10437968
M3 - 会议稿件
AN - SCOPUS:85187393449
T3 - Proceedings - IEEE Global Communications Conference, GLOBECOM
SP - 6005
EP - 6011
BT - GLOBECOM 2023 - 2023 IEEE Global Communications Conference
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 4 December 2023 through 8 December 2023
ER -