KX enables CSPs to proactively plan their network investment, develop build schedules, and introduce new technologies and services such as 5G and IoT in the most optimal fashion.

By applying machine learning processes to past deployments it can automatically determine how to improve new deployments and anticipate required configuration changes.

What-if analytics and machine learning-based models enable operators to prioritize deployment by assessing the impact of configuration changes or increased volume across the network.

KX provides processes and analytics for defining network deployments based on performance and quality targets with accompanying functionality for real-time detection, diagnosis, and remediation of network underperformance.

The underlying platform can ingest, store and analyze hundreds of network metrics such as service traffic over time, session data, geolocation, cell configuration, interference, noise, signal strength, and many others, for each cell of the network with maximum time granularity.

This functionality enables mobile operators to supervise network performance in real-time and to make statistical inferences to more efficiently manage their network. User quality data can also be included in the analysis to enable detailed mathematical modeling of network quality of services as it affects the customer experience.

As experienced telco solution providers, we have extensive experience in applying machine learning to techniques to not only improve customer quality measures but to automate them as well. KX offers specific tools and processes that enable CSPs to build mathematical models of customer quality for their network.

By consolidating and correlating data from a multitude of interfaces, you can evaluate customer satisfaction, reduce customer churn, improve service quality, and provide more attractive service offerings to increase customer loyalty and retention. Moreover, it can be done in real time.

KX can be integrated into your optimization process to focus on both quality and cost and to achieve the appropriate balance of both.

The solution implements models that infer customer quality in each cell and automates root cause analysis of deviation from defined service quality levels. Automation is achieved via real-time monitoring of network performance that measures deviation from the target quality model. Diagnostics identify root causes and propose the appropriate remedial action. The results then feedback into the same process to enable a continuous analysis of both quality and cost-effectiveness.