By Greg Burdett
The concept of a digital twin for a Wireless Network is simple, it is a digital representation of the Wireless Network that can be used to enable better decision making. As described in the article https://www.informationweek.com/it-leadership/boost-your-business-with-digital-twin-technology#close-modal, a Digital Twin of a Wireless Network can help Wireless Operators solve issues faster by detecting them sooner, predict outcomes with a high degree of accuracy, and to design and build better Wireless networks.
However, the article described above concludes with a warning to proceed with caution when building a Digital Twin. This is because the digital twin quality is directly related to the accuracy of the information received, or in other words, garbage in, garbage out. It specifically states, “All digital twin data must be trusted and actionable or else it just becomes a lot of noise”.
This article explores the concept of how to make a Digital Twin of a Wireless Network high quality such that its’ output, which is critical for advanced decision making, is also high quality.
Making a high quality digital twin of a Wireless Network requires the following;
Use input data that is as close to the source as possible
Verify data through analytical methods
Report findings of potentially incorrect data for correction
Let’s explore these points in a bit more detail.
The first item, “Use input data that is as close to the source as possible” is the first step in building a high quality digital twin. Using data that has gone through many levels of conversion, modification, communication, etc. adds risk to the quality of the data. Often data is improperly relayed or converted into another form impacting the quality of the data. If a critical field is incorrectly reset to a default value, this can fundamentally change the Digital Twin representation of the Wireless Network, and more importantly impact the decision making resulting from using the Digital Twin.
The second item, “Verify data through analytical methods”, is critical. Often data can be verified through analytical techniques when multiple sources of data is used. These don’t have to be different sources of the same data, but rather different types of data that should be consistent with each other, and definitely not contradictory, when used by the Digital Twin to build a representation of the Wireless Network.
Finally the third item, “Report findings of potentially incorrect data for correction“ closes the circle to ensure a continuous improvement cycle for the Digital Twin. If the Digital Twin analytics find potentially incorrect data but does nothing to try and correct the source data, not only will the Digital Twin be left using questionable data, but the benefit of correcting the data across the entire organization will be lost. If a key setting of a network element is improperly being communicated to many downstream users this can impact Maintenance and Support, Network Planning, Network Operations, and other critical functions. This is actually one of the key outputs of the Wireless Network Digital Twin in that by detecting and correcting inconsistent data many job functions are improved, in addition to creating a very high quality Digital Twin of the Wireless Network that can be used for advanced decision making.
By implementing the above recommendations the Wireless Network Digital Twin is not just a separate tool used by a decision making or strategy group, but rather an important and critical part of the overall process of building, monitoring, and maintaining a high quality Wireless Network.
JaggedXY is a Wireless Network Digital Twin pioneer and has built a Digital Twin in use by many Mobile Network Operators. The JaggedXY Digital Twin has been built to follow the methodology described to ensure accurate Cell Site and other network data used by the JaggedXY Digital Twin.