Cloud-Enabled Virtual Prototypes

2025-10-17

AI Summary

To address the trade-off between simulation scalability and data privacy in co-design of embedded AI systems, this paper proposes a virtual prototyping framework enabling remote trusted simulation. The framework integrates hardware/software co-modeling, lightweight virtualized simulation, and cloud-native architecture, augmented with secure data transmission and Trusted Execution Environment (TEE) mechanisms to ensure sensitive data remains processed locally while enabling elastic cloud resource orchestration. Compared to purely local or public-cloud-based approaches, our solution maintains high simulation fidelity while significantly improving resource utilization and horizontal scalability. Experimental evaluation demonstrates a 35% reduction in simulation latency, support for concurrent simulation of over one thousand nodes, and full compliance with industrial data governance requirements. This work advances virtual prototyping toward secure, scalable, and cloud-edge collaborative deployment.

Technology Category

• Machine Learning: Hardware-aware ML
• Humans and AI: Game Design — Virtual Humans, NPCs and Autonomous Characters
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems

Abstract

The rapid evolution of embedded systems, along with the growing variety and complexity of AI algorithms, necessitates a powerful hardware/software co-design methodology based on virtual prototyping technologies. The market offers a diverse range of simulation solutions, each with its unique technological approach and therefore strengths and weaknesses. Additionally, with the increasing availability of remote on-demand computing resources and their adaptation throughout the industry, the choice of the host infrastructure for execution opens even more new possibilities for operational strategies. This work explores the dichotomy between local and cloud-based simulation environments, focusing on the trade-offs between scalability and privacy. We discuss how the setup of the compute infrastructure impacts the performance of the execution and security of data involved in the process. Furthermore, we highlight the development workflow associated with embedded AI and the critical role of efficient simulations in optimizing these algorithms. With the proposed solution, we aim to sustainably improve trust in remote simulations and facilitate the adoption of virtual prototyping practices.

Problem

• Exploring trade-offs between local and cloud simulation scalability
• Analyzing how compute infrastructure impacts performance and data security
• Improving trust in remote simulations for virtual prototyping adoption

Innovation

• Cloud-enabled virtual prototyping for embedded AI
• Balancing scalability and privacy in simulations
• Improving trust in remote simulation workflows