One of the critical issues for mobile operators is the large amount of time and effort required to plan, build, and then optimize the network to ensure the best radio coverage and QoS. It is essential that the accuracy of the radio coverage prediction is as high as possible to reduce the amount of network optimization and re-work required. If the accuracy of the prediction can be improved, significantly less time is required to rework and fine-tune the network to achieve the optimum performance.

MbP (Measurement-based Prediction) software solves these problems by combining real drive test data with coverage prediction results to greatly increase prediction accuracy.

The program's high accuracy coverage predictions are achieved through the unique way in which Andrew applies drive test data to the whole prediction area by using complex interpolation methods.

It increases the reliability of the prediction process and allows coverage maps to reflect details that simply cannot be represented by other models, including those based on drive route pixels only.

Significant savings have been demonstrated by extensive MbP software field trials in Europe, when the number of cell sites could be reduced by up to 18%. Improved roll-out efficiency can lead to further savings of up to 25% in planning costs.

For more information or to order, email networksolutionssales@andrew.com or telephone +44(0)117 9060301.

The main objective of any network design is to achieve the optimum network in terms of radio and services coverage while meeting the cost targets within a given budget. If the radio plan is inaccurate, the network design becomes inefficient and money is wasted on deploying ineffective cell sites and antenna configurations, leading to further re-work to compensate for the poor design. The resulting patchy coverage offers a degraded quality of service to the end customers, which ultimately leads to customer dissatisfaction and churn.

Accuracy of the radio coverage prediction is crucial to the financial success of the mobile network and of the final delivered quality of service to the end customer. Just a 6% improvement in coverage prediction can lead to significant cost savings of up to 18% on deployment costs and an additional 25% on planning costs because of more effective cell site location and configuration.

A radio plan based purely on prediction may look good initially but in reality could be misleading since it is only as accurate as the input data used to configure it. A prediction is just that, only a prediction, and is unlikely to take into account all of the environmental influences found in practice when the network is deployed. By applying MbP software, the optimum radio coverage plan can be achieved by the intelligent combination of real measurement data with the coverage prediction.

The MbP system works by applying the drive test data to each point in the prediction results. Where the measurement agrees with the prediction data, then these coordinates are left alone. However, if there is a mismatch, then the drive test data is used as the reference data, and a new prediction is applied to the whole coverage area.

By applying Andrew's patented complex interpolation algorithm, the prediction results are modified to take into account the new reference data, and a "best effort" prediction is applied to the surrounding area on the plan. This increases the correlation between the predicted data and the actual network over a wider area.

The following examples are taken from a real field trial conducted in the United Kingdom during 2004/05.

The improved path loss prediction shows a greater percentage of hit rates after the application of MbP prediction software because of better correlation between the predicted data and the real network. (Additional high-resolution drive test measurements were taken to verify the improved prediction results across the coverage area.)