Comsof Fiber to Research Machine Learning Algorithms

Comsof Fiber Press Release

Comsof is proud to announce we are researching the next generation of FTTx planning & design automation, based on Machine Learning. Our goal is to develop smart algorithms which can find the balance between cost optimization and network constructability.

Ten years ago, Comsof was the first to introduce AI-based network planning & design automation in the telecom market. The heuristic algorithms of the Comsof Fiber software are able to create FTTx network designs at breakneck speed, while minimizing CAPex for network deployment.

Operators, equipment vendors and engineering firms around the world soon realized the added values of the tool and today our software is used in projects in over 50 countries. It is recognized as the most comprehensive fiber network planning & design tool.

Practical Optimization

As our company has a culture of constant innovation, we are preparing for the next generation of our software. We have been researching self-learning algorithms, which aim to find the optimal design virtually autonomously. Our focus is on practical optimization.

Optimizing cabinet and closure coverage and sizes. Optimizing cable and duct routes. Optimizing demand point connections by finding the right cost-coverage threshold per connection, are just a few examples.

All will be underpinned by smart algorithms that understand more than just the logic of mathematical formulas. These algorithms will learn from the best network designers and our years of experience on finding the balance between cost optimization and network constructability.

Stay tuned for more updates in 2021