Research Article
[Retracted] Lightweight Statistical Approach towards TCP SYN Flood DDoS Attack Detection and Mitigation in SDN Environment
Table 3
Comparative study of various DDoS detection strategies.
| Ref. no. | Approach | Method | Benefits | Limitations |
| [40] | Anomaly-based detection | Entropy, source IP index, and packet rate | Low computational time, low false-positive/negative rate, and high detection throughput | Low accuracy and low adaptability | [41] | Machine learning | DNN | High accuracy | High computational time | [42] | Statistical | Used FGPA | Nine types of DDoS attacks can be detected based on incoming traffic | High detection time, computational complexity, and low flexibility | [43] | Rate limiting | FlowSec | Low computational time and high detection throughput | Low accuracy and high false-positive rate | [44] | Statistical | Switch statistics | Low computational time, High accuracy, and flexibility | High false- positive/negative rate and complexity | [45] | Machine learning | LSTM + CNN | High detection accuracy and minimal false-positive rate | Memory consumption and computational complexity are high | [46] | Hybrid | Long short-term memory + fuzzy logic | High accuracy, flexibility, and minimum damage | Computational complexity |
|
|