E-mail for more information Company's other products E-mail to a colleague See similar products Printer friendly format

By Gur Lavie

The majority of cellular service providers plan, maintain and troubleshoot their Radio Access Network (RAN) performance by performing extensive drive tests and using prediction-based models.

Since radio access networks comprise most of the providers’ CapEx investment (50%-70%), and have a major impact on the revenue potential through user experience of the network, it is no surprise that major efforts are being made to maximize RAN throughput, capacity and quality. As service providers deploy more demanding 3G

click to enlarge Figure 1. All day traffic analysis.

services and as adoption rate increases, quality of service may be compromised and customer satisfaction put at risk. So, the main question is, how can service providers sustain network performance while adding new 3G services?

Traditional network maintenance and optimization methods that rely on drive-test information and simulated signal-strength propagations (predictions), suffer from three major drawbacks: inaccuracy, cost and time-consumption.

Inaccuracy
Drive-test data provides an easy solution to the wrong problem. In 3G networks, approximately 70% of traffic originates from stationary, in-building environments — malls, office buildings, residential housing etc. Drive tests provide excellent information about network quality for moving terminals in open areas (only), but fail to provide information on the areas where the bulk of traffic actually occurs. The service provider loses a great deal of traffic potential when basing network optimization on drive-test data alone; this translates to lost Minutes of Use (MOUs) and reduced Average Revenue Per User (ARPU).

Simulated signal-strength propagations are only an approximation of the actual RF (radio frequency) signal quality across the cellular network coverage area. Service providers often find that 20% to 40% of the network-quality simulation is erroneous; this leads to higher churn rate, wasted CapEx investment and incorrect operational priorities.

Cost
Collecting, preparing and analyzing the data necessary for network planning and optimization, requires expensive geographical maps and high quality drive-test equipment. Drive tests must be executed repeatedly and extensively throughout the network. For the process to be efficient and to ensure that the results are accurate, these costly expenses must be budgeted on a periodic schedule and cannot be reduced to a one-time expense.

Time Consumption
The combined process of conducting drive tests, calibration and ray tracing consumes considerable time and effort. Drive tests, by nature, consume time and resources. Signal-strength calibration, whereby the simulated signal strength is modified to better correlate and reflect drive-test results, are particularly complex. Creating signal-strength propagation for ray-tracing-based algorithms can take hours and even days. All of these methods require special skills, training and experience to be used effectively.

As networks become more complex, new services are being launched and traffic volumes are increasing constantly, operations move into a service-centric approach.

Traditional methods of maintenance and optimization are not in sync with the service providers’ ever-changing time and quality constraints. Therefore, they seek ways to monitor their network performance in real-time and keep their network constantly tuned to the changing traffic patterns.

Finding a Better Solution
Service providers will be satisfied with predictions and drive tests for only so long. They need a solution that drives planning, maintenance and troubleshooting activities from an accurate, cost-effective and real user experience point of view.

A new approach that has already proved to provide an excellent solution to these challenges is based on the collection of advanced service quality reports originating from

click to enlarge Figure 2. Busy hour traffic analysis

the end-user’s handsets. As subscribers move around the network and use their mobile handsets, their mobile phones are constantly engaged in complex interactions with the network. This interaction yields a specialized quality-of-service report called mobile measurements that reflects the customers’ experience with the network quality (signal strength, noise level, candidate handover lists, etc.).

These measurements are collected from 3G terminals and provide a valuable information source that products like Mentor, offered by Schema, use to create accurate, real-time traffic and service coverage maps, positioning of drop calls hot spots and terminal type quality analysis.

For the first time ever, service providers can get a valid and accurate view of their network quality inside buildings, which amounts to 70% of the traffic. They now have a tool that will reveal traffic spread across the network and revenue-yielding hot spots. Service-degradation locations can be proactively identified and eliminated.

By making use of these reports, together with advanced analysis algorithms, an operator can achieve great advantages over the current methods. These types of solutions can help with network maintenance and troubleshooting by automatically identifying network quality degradations and reduced user experience in key areas as well as proactively provide solutions.

As mobile subscribers, most of us take part in the adoption of the emerging 3G technology and experience its “teething pains”. However, when we go for the upscale devices, fancy services and expensive air time, we want to be impressed. We want high bandwidth data, speedy game and music downloads and good video quality. Service providers that manage to excel in identifying their hot spots usage for each service, exact problem areas and the right priority in network investment, will attain higher minutes of use, increased ARPU and the best customer satisfaction.

Gur Lavie is director, product marketing, at Schema. He is responsible for global product strategy, as well as working closely with customers designing optimization solutions for UMTS, HSDPA, CDMA and EVDO networks. Schema is located at 218 Rt. 17N, Rochelle Park, NJ; (201) 556-3100.


Schema Inc.