This paper proposes a hybrid Autoencoder–Multi-Layer Perceptron (AE–MLP) model for detecting Distributed Denial of Service (DDoS) attacks in Mobile Edge Computing (MEC) environments. The MEC framework faces significant security challenges, particularly from DDoS attacks, which threaten the availability and performance of MEC services. Traditional Intrusion Detection Systems (IDSs) rely on shallow Machine Learning (ML) models, which are limited in their ability to detect and mitigate DDoS attacks. The proposed hybrid AE–MLP model leverages autoencoders' feature extraction capabilities to capture intricate patterns and anomalies in network traffic data. This extracted knowledge is then fed into a Multi-Layer Perceptron (MLP) network, enabling deep learning techniques to further analyze and classify potential threats. The hybrid model achieves higher accuracy and robustness in identifying DDoS attacks while minimizing false positives. The proposed model was evaluated using the NF-UQ-NIDS-V2 dataset, which contains a wide range of DDoS attacks. The results show that the hybrid AE–MLP model achieves a high accuracy of 99.98%, outperforming several existing techniques. The model's performance was tested with different data splits (80/20, 70/30, 60/40), and the highest accuracy was achieved with the 80/20 split. The model also demonstrated efficient training and prediction times, with training time of 4.96 seconds and prediction time of 0.74 seconds. The hybrid AE–MLP model offers several advantages, including effective anomaly detection, adaptability to evolving attack scenarios, and the ability to learn patterns in large amounts of traffic data. However, the model is computationally intensive, and its training time is longer than other models. To mitigate this, cloud-edge collaboration is suggested, where the model is trained in the cloud and deployed to the edge for real-time DDoS attack detection. The study concludes that the hybrid AE–MLP model is a promising solution for DDoS attack detection in MEC environments, providing high accuracy and robustness while minimizing false positives. The model's integration of autoencoder's data compression and feature extraction capabilities with MLP's classification capabilities enables effective processing and extraction of significant features from data at the edge, optimizing latency and bandwidth utilization. This makes the hybrid AE–MLP model an ideal solution for MEC applications that require real-time decision making.This paper proposes a hybrid Autoencoder–Multi-Layer Perceptron (AE–MLP) model for detecting Distributed Denial of Service (DDoS) attacks in Mobile Edge Computing (MEC) environments. The MEC framework faces significant security challenges, particularly from DDoS attacks, which threaten the availability and performance of MEC services. Traditional Intrusion Detection Systems (IDSs) rely on shallow Machine Learning (ML) models, which are limited in their ability to detect and mitigate DDoS attacks. The proposed hybrid AE–MLP model leverages autoencoders' feature extraction capabilities to capture intricate patterns and anomalies in network traffic data. This extracted knowledge is then fed into a Multi-Layer Perceptron (MLP) network, enabling deep learning techniques to further analyze and classify potential threats. The hybrid model achieves higher accuracy and robustness in identifying DDoS attacks while minimizing false positives. The proposed model was evaluated using the NF-UQ-NIDS-V2 dataset, which contains a wide range of DDoS attacks. The results show that the hybrid AE–MLP model achieves a high accuracy of 99.98%, outperforming several existing techniques. The model's performance was tested with different data splits (80/20, 70/30, 60/40), and the highest accuracy was achieved with the 80/20 split. The model also demonstrated efficient training and prediction times, with training time of 4.96 seconds and prediction time of 0.74 seconds. The hybrid AE–MLP model offers several advantages, including effective anomaly detection, adaptability to evolving attack scenarios, and the ability to learn patterns in large amounts of traffic data. However, the model is computationally intensive, and its training time is longer than other models. To mitigate this, cloud-edge collaboration is suggested, where the model is trained in the cloud and deployed to the edge for real-time DDoS attack detection. The study concludes that the hybrid AE–MLP model is a promising solution for DDoS attack detection in MEC environments, providing high accuracy and robustness while minimizing false positives. The model's integration of autoencoder's data compression and feature extraction capabilities with MLP's classification capabilities enables effective processing and extraction of significant features from data at the edge, optimizing latency and bandwidth utilization. This makes the hybrid AE–MLP model an ideal solution for MEC applications that require real-time decision making.