A new peripheral caching method using a “digital twin” significantly improves wireless network performance by predicting user data needs and optimizing data storage, thereby improving network reliability and speed. The researchers aim to test the method in real applications.

Computer scientists have developed a new technique that improves the speed and reliability of wireless networks by predicting the data needs of wireless computer users before they occur. This innovative method enables proactive data management using “digital twin” technology that replicates the network it supports.

This is something called edge caching. Caching refers to storing data on a server that a system or network believes users will use (or reuse) in the near future. This allows the system to meet user needs faster than if the system were to retrieve data from the original source. Edge caching is when a system caches data on the server closest to the end user, such as computers connected to network routers or co-located.

“The two big challenges are determining what data should be cached and how much data the edge server should store at any given time,” said Yuchen Liu, lead author of the paper and an associate professor of computer science at North Carolina State University. “Systems can’t fit everything into the edge caches, and storing excess data on an edge server can slow down the server if the data uses too much computing power. As a result, systems are constantly deciding which data packets to keep and which to discard.

“The more accurately the system predicts what data users will actually need and how much data the edge servers should store, the better the system’s performance,” Liu says. “Our work here is focused on improving these predictions.”

D-REC: A New Edge Caching Optimization Method

The new edge caching optimization method, called D-REC, uses a computational simulation technique called a digital doppelganger. A digital doppelganger is a virtual model of a real object. In the case of D-REC, the digital doppelganger is a virtual model of a particular wireless network (a cellular or Wi-Fi network).

“This method can be applied to any wireless network, depending on the needs of the system administrator or network operator,” Liu says. “D-REC can be customized according to the user’s needs.”

In D-REC, the digital twin receives real-time data from the wireless network and uses it for modeling to predict what data will be most requested by users. These predictions are then sent back to the network to inform the network’s edge caching decisions. Since the simulation is done by a computer outside the network, it does not slow down the network.

Performance and prediction capabilities of D-REC

To determine whether a wireless network performs more efficiently with D-REC, the researchers used open-source datasets. They conducted large-scale experiments that aimed to account for many variables, such as the scale of the network and the number of users on the network.

“D-REC outperformed traditional approaches,” Liu says. “Our technology improved the network’s ability to accurately predict what data should be cached. D-REC also helped systems better balance data storage across their networks.”

Additionally, since D-REC’s digital twin focuses on predicting network behavior, it can detect potential problems in advance.

“For example, if a digital twin thinks that a particular base station or server is likely to be overloaded, the network can be notified and allowed to redistribute data across the network to maintain network performance and reliability,” Liu says.

“We are currently open to working with network operators to explore how D-REC can improve network performance and reliability in real-world conditions.”