DesignTwin-SDN: A digital twin-driven AI framework for smart architecture and interior planning

Abstract

In today's world, the need for smart buildings along with smart interior design is quickly growing, requiring advanced AI-based methods for better area usage, power conservation as well as safety. Usual methods battle in continuing real-time adaptability, with several security concerns, and in sufficient data processing, making them inadequate for multidimensional architectural and interior design systems. For countering these challenges, this research introduces DesignTwin-SDN, a Digital Twin-based AI system in combination with CRL-Net, a Deep Learning (DL)-based clever model fusing CNN, ResNet-50, and LSTM for the creation of more precise spatial-temporal analysis and anomaly detection. Also, for possessing strong security, the system includes Adaptive Flow-Encrypted AES (AFEA), a dual-mode encryption strategy of combining Advanced Encryption Standard (AES) with Flow-Based Dynamic Encryption (FBDE). Furthermore, Starfish-Orangutan Optimization Algorithm (SOOA), which is a combination of the Starfish Optimization Algorithm (SFO) with Orangutan Optimization Algorithm (OOA), is proposed for the selection of the optimal secret key for purposes of maximizing that encryption efficiency. The Zero-Knowledge Proof (ZKP) system further secures authentication by precluding forbidden use in smart spaces. Experimental results show the suggested framework's effectiveness. These results show a 99.05% accuracy and a 98.51% recall, a large increase over the current method. DesignTwin-SDN merges DL and digital twins with SDN-IoT to change smart architecture and interior layout planning into more secure, efficient, and flexible systems with dynamic environments.