This paper presents a novel hybrid model combining entropy-based features and support vector machine (SVM) techniques for effective detection of DDoS attacks in software-defined networks (SDNs). The proposed model significantly enhances detection accuracy and is validated through simulations and practical implementations, addressing limitations of existing methods that generally operate in simulated environments. By leveraging a dynamic threshold for entropy calculation, the model offers real-time anomaly detection and aims to enhance network security against evolving cyber threats.