RF optimization work is a process through which different physical parameters (like antenna adjustments) in a cell and logical parameters like RF parameters of a cell/BSC (collectively called BSS) are modified to provide good coverage, quality and capacious network. Drive testing a route with a software tool in a laptop, and iterative actions to improve the service, is the conventional way of performing the optimization work. GSM RF team that I am working with are doing this sort of hit and trial work to address the problem so far. Drive testing an area identifies a very common problem for the cell, that is degrading the quality of service for the users in that area but the extensive analysis of the DT log file requires a post processing tool that determines the cause of severe problems and helps solve it immediately. Absence of post processing tool in RF optimization has paid a lot in being efficient/fast in solving a RF related problem. But when it comes to solving a problem in whole of the network, DT and post processing even falls short for the limitation of analysis only in the DT route (as most of the DT are done in streets and roads) and the time it takes in analysis and solving the problem. Also more than half of the network usage happens indoor, drive testing and post processing in every corner requirement makes it out of date. Virtual Drive test (VDT), is one of the new techniques that enables every mobile users to be a drive testing tool, and acquires the signal measurement of serving and neighbor cells by the user along with its position. Those measurement data from probe can be analyzed with tools to identify problems in whole of the network and act instantly.
Integration of prediction data and Drive test or Measurement data (from Probe) is also possible in the latest planning/optimization tools in giving better work plan for RF optimization and is pretty significant in yielding good KPI (Key performance Indicators). This sort of tool (Capesso with Aircom Asset) is the latest to reach me which has not only such integration but has the ability to efficiently optimize the number of the sites for the coverage/capacity/quality requirement. Measurement report accumulation requires additional probe in BSC/RNC/Core which can even monitor and analyze customer problem happened in the near past. Other tools (that may come to my hand in near future), even has the ability to automatically modify/plan network configuration, being based on the live data taken from the network. Such tools remove the need for periodic and cuber some manual data processing, and is one step further towards the fully automated Self Organizing networks (SON).
Self Organizing Networks (SON) is a fully automated technology that has the functionality to self configure, self optimize and self heal in a mobile Radio access network (RAN), such that rapid roll out of better quality network is possible with least manual interactions. Self configuration enables a site to be just plug and play, with the automatic cell configuration, yielding faster roll-out. Self optimization include automated load balancing to handover optimization to coverage optimization for speeding the best optimization for good quality network. Self healing automatically adjusts cell parameters and recover failures in case of detection of a failure. SON being a feature of LTE (Long Term Evolution),can also be applied to GSM and UMTS radio network but is not yet launched full fledgedly.
Trends in GSM RF Optimization, from drive test based to integration of prediction, measurement and live data, is now moving towards full automation, which is the future of next generation mobile networks. All in all, such tools help a telco run a good quality network, satisfying customers in every nook and corner.