AI-Driven DDoS Detection in 5G Edge Networks: A Performance Comparison Study
Bibliographic record
Abstract
Distributed Denial-of-Service (DDoS) attacks are one of the harmful attacks that is caused due to the increase in number of wireless devices, which in turn impact the service availability and network reliability. DDoS attacks occur in various applications such as websites, servers and DDoS attacks flood a website or service with traffic to make it unavailable, which would have a negative impact on businesses, online security, sales, and reputation. Real-time detection of such attacks at the edge may require lightweight, accurate, and responsive machine learning solutions. In this paper, a comparative study of four AI techniques—CNN, LSTM, Autoencoders, and XGBoost is performed—using a synthetic, imbalanced dataset simulating 5G network traffic. This study is done to illustrate the effectiveness, feasibility, and adaptability of AI-driven methods for DDoS detection in 5G edge networks, addressing critical limitations of traditional systems. Other reasons include early detection of DDoS attacks using AI models and real-time response, detection of sophisticated and evolving DDoS attacks, and comparison of various model performance metrics such as accuracy, latency, resource efficiency, and scalability of AI models at the edge. CNNs and autoencoders can be used to show how deep learning can automatically extract useful features from raw traffic data. The dataset consists of 500 traffic flows, each represented as a time-series with three steps, reflecting both normal and DDoS patterns. The evaluation done considers classification performance metrics and visualizes the results. XGBoost model demonstrates the highest balanced performance, while Autoencoders shows high precision in detecting anomalies but with limited recall. The research work provides practical perceptions about the suitability of each model for edge-based DDoS detection, highlighting the trade-offs between accuracy and interpretability under real-world constraints.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".