Home automation based on IoT is useful for smart living that provides control and comfort. With increase in use of IoT devices, there has been a rise in cyberattacks, with DDoS attacks and data integrity breaches becoming a great risk to network security. The challenge of DDoS protection of IoT enabled home automation has been studied in this paper and edge based security solutions are emphasized. This paper proposes a way to detect and prevent DDoS Attacks in real time using an edge computing methodology based on CNN-LSTM models. In order to successfully detect anomalies, CNNs detect spatial traffic patterns and LSTMs detect temporal patterns. Harmful communications are detected by the system, malicious IP addresses are blocked, attacks are stopped before escalating. Through a real time monitoring interface, the system also supplies the admins with performance information and insights. By using this approach, we can ensure durability against the evolving cyber threats and improve the security and reliability of IoT home automation. Our model achieves 99.99% accuracy on 81,117 records and 17 features whereas such a recent 2024 state of the art study using 10,000 records and 10 features achieved 98% accuracy. The model was successfully run on raspberry pi 4 for testing on resource-constrained IoT devices in .h5 format.