CDMA networks are inherently more complex to design and optimise than TDMA networks. The system-level performance interdependencies of CDMA systems, limits the corrective actions that can be taken to rectify design inefficiencies after launch. This is a well established fact of life in IS-95 networks and is now being slowly recognised throughout Europe where many networks were designed using a 2G mindset, methodologies and tools.

An effective mechanism that can provide an objective view of the design quality in early stages of the work is the Preliminary Design Review (PDR). The PDR is usually carried out at network level as a mechanism to identify major problems and prevent these problems being reflected in the network design. PDR is most effective when carried out by a separate group to the design team as it provides a fresh look into the design.

The input to the PDR is a 3G preliminary plan and a 2G impact analysis report. The 3G plan will primarily include a report as well as the design data in an agreed format. The 2G impact analysis is used when modifications to an existing 2G site is proposed. The impact analysis report includes any coverage variation as a result of the deployment of multi-band antennas, azimuth changes or any other changes that could impact the performance of existing 2G sites.

2G Impact and Interference Studies

Co-locating UMTS and GSM sites, results in serious degradation in GSM performance if certain criteria are not met during the design and implementation process. Some of the serious implications include the spurious emissions from UMTS transmitters and the effects on GSM channel quality. This effect can be detected by measuring the C/I level on the downlink of GSM channels (BCCH channels for example).

Furthermore, a number of GSM parameters could be modified when UMTS transmitters are co-located on existing GSM sites. However, when antenna sharing is applied, most of the GSM parameters are modified. Therefore, GSM coverage and consequently the network performance could be affected.

A multi-band Scanner from Comarco can provide a typical solution for non-intrusive UMTS-GSM impact assessment. The Scanner can be used for drive testing of both the 900 MHz and 1800 MHz bands of the GSM network. The following parameters can be measured or determined for each of the BCCH Channels:

Best Server Received Signal Level: This will provide good knowledge of the coverage of each cell.

C/I at every BCCH channel: This will provide us with the signal quality at the BCCH channel. Therefore, we can detect the impact of changes in other cells configurations on the performance of the investigated cell. Another advantage of the C/I parameter, is that we can detect the spurious UMTS interference if it occurs during the drive testing process.

BSIC parameter can also be identified at every measurement point. Therefore, identifying the base station to which the measured BCCH belongs, enabling pinpoint accuracy in spotting the problem and facilitating its correction.

A survey of this nature would typically include the following:

• Baseline measurements on the performance of a GSM900/1800 network prior to the deployment of UMTS2100 network,
• Auditing the existing GSM900 and GSM1800 networks for RxLev and C/I,
• Conducting measurements over the operator’s 2100 MHz band to:
  • Identify the areas where in-system interference exists in clusters where the UMTS2100 network is already deployed,
  • Identify the areas where there are external interferers in clusters where UMTS2100 is not yet deployed,
• Taking measurements to capture areas/sites where deployment of UMTS2100 infrastructure is adversely affecting the operation and performance of the existing GSM900/1800 networks.