This paper seeks to investigate the performance of the Exponentially Weighted Moving Average (EWMA) for mining big data and detection of DDoS attacks in Internet of Things (IoT) infrastructure. The paper will investigate the tradeoff between the algorithm’s detection rate, false alarm and detection delay. The paper seeks to further investigate how the performance of the algorithm is affected by the tuning parameters and how various network attack intensity affect its performance. The performance results are analyzed and discussed and further suggestion is also discussed.
Reference:
Machaka, P., Bagula, A. and Nelwamondo, F.V. 2016. Using exponentially weighted moving average algorithm to defend against DDoS attacks. 2016 International Conference on Pattern Recognition Association of South Africa and Robotics and Mechatronics (PRASA-RobMech), 30 November to 2 December 2016, Stellenbosch, South Africa, Cape Town. 10.1109/RoboMech.2016.7813157
Machaka, P., Bagula, A., & Nelwamondo, F. V. (2016). Using exponentially weighted moving average algorithm to defend against DDoS attacks. IEEE. http://hdl.handle.net/10204/9316
Machaka, P, A Bagula, and Fulufhelo V Nelwamondo. "Using exponentially weighted moving average algorithm to defend against DDoS attacks." (2016): http://hdl.handle.net/10204/9316
Machaka P, Bagula A, Nelwamondo FV, Using exponentially weighted moving average algorithm to defend against DDoS attacks; IEEE; 2016. http://hdl.handle.net/10204/9316 .
2016 International Conference on Pattern Recognition Association of South Africa and Robotics and Mechatronics (PRASA-RobMech), 30 November to 2 December 2016, Stellenbosch, South Africa, Cape Town