When US/Israel-based Schema decided in 1998 to tackle the issue of optimization in wireless networks, the company was open about its lack of experience in the telecoms industry. In fact, the company sees this as a strength, and its customers and investors seem to agree. Schema has just completed its third round of venture capital funding and raised $25 million.
The company`s founder, Yuval Davidor, is passionate about looking at the wider picture to solve problems, and Schema was founded on this premise in 1994. "With complex problems people often make their minds up too early," he said.
Before Schema turned its attention to the telecoms market, it focused on defence, medical applications and logistics. "We were never masters of the domain that we tried to address, and it`s the same with telecoms," said Davidor. "We did not know RF so were forced to bypass an ego issue. It allowed us to see problems through the eyes of end users."
Davidor recognises that mobile network operators face the "small blanket problem", where different departments pull the financial and radio resources in different directions. "So much is at stake that nobody wants to share information even between the marketing, finance and engineering departments. But all of these departments affect the spectrum efficiency," he said.
Marketing departments want to offer subscribers incentives like lower prices and free minutes but these conflict with the capacity limits of the network. To increase capacity, engineers must build new infrastructure but this reduces the capital efficiency of the company. A solution to the capital efficiency problem is to increase the number of subscribers but this conflicts with the capacity limits.
If these factors are not balanced properly they can result in churn - subscribers switching to another operator. Lower costs elsewhere and quality issues account for almost two-thirds of churn. Schema`s telecom resource management (TRM) solution - known as Falcom - promises to address both of these issues.
The optimization tools start by integrating network information such as results from propagation, and drive or walk tests of network quality. The tools also examine switch data: the logging of traffic throughout the network; the configuration of the network; physical data such as the position of base stations; and the demands of the operator and customers.
The many gigabytes of data gathered in the initial phase are used to construct an impact matrix. This predicts the effect that any change in the network would have on every base station, and allows operators to test several "what if?" scenarios before making any physical changes to their network.
After the impact matrix has predicted different scenarios the next stage is optimization, when the requirements for the new network are fed into the system. A scenario that the operator selects can then be implemented throughout the network.
The results from the TRM investigation may differ significantly from the plan that would be suggested by a single department in the company. For example, changing the quality distribution of the radio channels could decrease the proportion of excellent channels but also result in fewer very poor channels. "Generally, an RF engineer would not take a very good channel and make it worse," said Davidor. "However, the operator is not going to get a special bonus from the subscriber for providing an excellent channel rather than a good one. On the other hand, if the subscriber does not get a good connection they are likely to churn."
Davidor says that optimization solutions help operators migrate to new technologies. European operators will face a challenge as they overlay GPRS onto their GSM networks. "If we take one cell with three voice channels and overlay GPRS across the spectrum we either have to take one of the existing channels for data - leaving just two for voice - or create a new channel by spectrum clearance," he explained. The second solution is better because it does not penalize the voice customers, but it is hard to do in an established network where spectrum cannot readily be reassigned. There are interference problems and the system has to be upgraded without disrupting the service to existing subscribers.
The role of optimization tools in migration has already been proven in the US. "The US operator Bell Atlantic bought the Falcom TRM tool primarily to assist with its migration from analogue to CDMA," said Davidor. "The technologies are different but the migration issues are the same."
.gif)
