Generative AI-enabled Digital Twins for 6G-enhanced Smart Cities

Kubra Duran,Lal Verda Cakir,Mehmet Ozdem,Kerem Gursu,Berk Canberk
2024-11-21
Abstract:6G networks are envisioned to enable a wide range of applications, such as autonomous vehicles and smart cities. However, this rapid expansion of network topologies makes the management of 6G wireless networks more complex and leads to performance degradation. Even though state-of-the-art applications on network services are providing promising results, they also risk disrupting the network's performance. To overcome this, the services have to leverage what-if implementations covering a variety of scenarios. At this point, traditional simulations fall short of encompassing the dynamism and complexity of a physical network. To overcome these challenges, we propose the Generative AI-based Digital Twins. For this, we derive an optimization formula to differentiate different network scenarios by considering the specific key performance indicators (KPIs) for wireless networks. Then, we fed this formula to the generative AI with the historical twins and real-time twins to start generating the desired topologies. To evaluate the performance, we implement network and smart-city-oriented services, namely massive connectivity, tiny instant communication, right-time synchronization, and truck path routes. The simulation results reveal that our approach can achieve 38% more stable network throughput in high device density scenarios. Furthermore, the generated scenario accuracy is able to reach up to 98% level, surpassing the baselines.
Networking and Internet Architecture
What problem does this paper attempt to address?
This paper attempts to solve two main problems faced by 6G wireless networks in smart city applications: 1. **Decrease in network throughput**: As the topology of wireless networks expands to support more devices and larger coverage areas, the network throughput decreases significantly. This is mainly due to the difficulty of network infrastructure in meeting the demands of high - density devices. Therefore, intelligent mechanisms need to be developed to manage a large number of devices to ensure that network performance is not affected. 2. **Lack of effective what - if implementation**: Traditional simulation methods cannot fully capture the dynamics and complexity of physical networks, especially when they are unable to predict future results in the face of parameter changes. Therefore, accurate models that can reflect the potential impact of different scenarios on network performance need to be established for decision - making and service testing. To solve these problems, the author proposes a digital twin framework based on Generative AI. By introducing an optimization formula to distinguish different network scenarios and combining historical and real - time digital twin data, the required network topology is generated. Specifically, this framework solves the above problems in the following ways: - **Optimization formula**: Four key performance indicators (KPIs) are considered, namely device density \( \rho \), packet deadline \( d \), delay \( l \) and buffer size \( \alpha \), and different scenarios are prioritized according to these indicators. The optimization objective function is: \[ O = w_\rho T_\rho+ w_d T_d + w_l T_l+ w_\alpha T_\alpha \] where \( w_i \) is the weight of each KPI and \( T_i \) is the calculated value of each KPI. By maximizing this objective function, the KPI requirements can be efficiently met in network management. - **Generative AI - driven scenario twin creation framework**: Using historical and real - time digital twin data, combined with the optimization formula, future network scenarios are generated. This enables what - if testing of various services before deployment, thereby improving the accuracy of scenarios. The experimental results show that this framework improves the network throughput stability by 38% under high - device - density conditions, and the generated scenario accuracy reaches 98%, significantly outperforming traditional methods. In summary, this paper aims to improve the management and performance of 6G wireless networks in smart cities, especially in dealing with complex and changeable network environments, by introducing Generative AI technology.