12 APRIL 2016

The benefits of automated geo-location driven analytics

The rise in network complexity and increased focus of many CSPs on improving organizational efficiency and processes, has created some new challenges for radio network engineers, impacting how they cope with their day-to-day tasks. Today’s engineers are expected to handle, manage, optimize, monitor and troubleshoot multi-technology 2G/3G/4G and multi-vendor networks. To do their jobs effectively, they are required to analyze multiple dimensions of data sources, looking at cell, subscriber and device level KPIs, monitoring for faults in network equipment and more. They then need to combine and analyze all this data in order to create standardized root cause analysis workflows, but in most cases this can be difficult. Engineers must first have a transparent view of both network performance and subscriber experience at any given time, and have the ability to perform tasks quickly in order to maintain high quality of service, while minimizing the effects of network faults.

There are solutions on the market today that are designed to help engineers reduce the time required for troubleshooting and analyzing the network. For instance, some of today’s geo-location based solutions can deliver the network transparency engineers need. They often include various sets of traffic and performance maps that show subscriber usage, then apply additional logic to more quickly identify network faults. Additional analysis methodologies can be applied, where the engineer defines a problem using a combined set of criterion with certain thresholds. For any set of thresholds that are violated, an alarm is generated to alert the engineer to a potential problem. In some cases, additional pre-defined root cause analysis is applied to identify the actual cause of the problem, and not just the problem itself. The next step is to propose and implement corrective actions.

TEOCO as a market leader in RAN optimization and geo-location solutions, providing unique and innovative approaches that embed geo-analytics capabilities into its offerings. Built on TEOCO’s rich expertise in RAN Optimization, this suite of solutions provide engineers with the ability to automatically identify the root cause of many network problems by using a decision support system (DSS) approach. These solutions enable radio engineers to use simplified workflows and smarter network management tools that utilize best-practices, based on TEOCO’s many years of experience. An automatic, comprehensive analysis is performed over multiple dimensions, such as network topology, performance counters and geo-located call traces. This is done through a sophisticated fuzzy logic algorithm, where abnormal network or subscriber usage behavior is automatically identified and searched. Each of the identified problems is then reported and categorized by level of severity, based on a weighted score mechanism, which includes the dimensioning of multiple KPIs.

Engineers can use either pre-existing or customized user-configured problem criteria, but there is no need to waste time defining highly specific static thresholds, since criteria can sometimes vary between different clusters or different time periods, can be subjected to change and require ongoing maintenance. TEOCO’s unique algorithm can use static thresholds, or it can automatically geo-locate underperforming KPIs, examining all the areas in question. The results are then mapped out across geographical polygons, identifying problems by order of severity and suspected cause.

This analysis can be followed up with traditional drill down investigations, using intuitive work flows. The engineer can arrive to work the next morning and view all the network faults that happened the previous day, highlighting the geographical areas that were affected and their suspected cause, all ranked by level of severity. High severity cases will need to be addressed as soon as possible, since they can impact service, while lower severity cases can be addressed as required.

Automatic problem detection can be a real time-saver in the day-to-day life of a radio engineer. And even if everything is okay, the engineer still wants to have visibility and the assurance that things are under control. These solutions help engineers sleep at night. Working as around-the-clock control monitoring systems, they indicate if network performance and subscriber QoS is at their required levels, while also providing an accurate view of network conditions for radio engineers and their managers.

Author: Yony Dvory, Mentor Product Manager