As with any business investment, operators will be seeking to justify femtocell deployment through increased revenues, reduced costs and improved service quality. Individual vendors are conducting trials with a view to satisfying these better than others. Ubiquisys, a leading femtocell OEM, highlights three specific areas on their website where they differentiate:

• Efficient integration into the operator’s core network
  
• Interference and co-existence with the outdoor macrocellular network
  
• Cost-efficiency

It is in these areas that femtocell vendors seek to compete and differentiate from each other. Let’s delve a bit deeper into each of these areas:

Integration of femtocells into the operator’s network

It has been said that there are far too many different system architectures for femtocells. Broadly, they fall into two families – Iu-based and SIP/IMS based. For seamless compatibility and ease of deployment, Iu interfaces will be used first. Femtocell gateways may have several hundred thousand units connected via individual secure IP tunnels, handling thousands of simultaneous voice and data calls. These must be managed to the same standard and in the same way as existing systems.

Existing RAN vendors will argue they have more experience of running a radio network for the operator (they may already be doing so). Independent femtocell vendors may say that these products scale differently and thus require a different implementation and expertise from traditional systems. Network management includes tasks such as updating the configuration of the system, monitoring its health and dealing with faults and outages as they occur, and expanding/optimising the network. Operators will want the same teams of experts to handle the impact of any general femtocells issues. They will want to avoid having to set any parameters within the macrocellular network depending on individual deployments of femtocells.

At this time, it is thought that femtocells are being designed to be completely autonomous and self-configuring with respect to the outdoor network. All femtocell vendors appear to be adopting the TR-069 standard for remote device management. This should be fairly easy to integrate with any other TR-069 management systems the operator already has (for example if they already provide DSL modems). Partnerships or proven interworking with leading TR-069 management systems should help demonstrate the easy integration here.

Interference and co-existence with the outdoor macrocellular network

This is a very complex issue and requires extra “intelligence” within the femtocell to detect and assess the best frequency, code and power level to operate at. Self installing femtocells will “warm up” for about 10 minutes whilst scanning the environment and making these choices.

Additionally, femtocells need to continuously monitor for any changes and adjust their parameters accordingly. Opening patio doors, physically moving the femtocell and external macrocellular outages may all change the situation rapidly and require prompt response.

Handover to/from the external macrocellular network is also important. Initially, handover from the femtocell to outdoors (especially if a 3G to 2G handover) is arguably the most important to implement first. This will require the femtocell to scan and monitor the 2G frequencies so that it can announce the correct “neighbour list” of external sites for the mobile phone to monitor as handover candidates.

An associated issue relates to timing and clock frequency which must be implemented within the tight specifications of 100 parts per billion to ensure that handover works correctly and avoids interference with external networks.

Cost-Efficiency

The market is targeting a unit cost of US $100 or less per unit. This has put intense pressure on vendors to find ways of optimising the cost. The industry has taken up common aspects including:

• Specialised and optimised chips. Whilst many vendors have selected picoChip or other dedicated femtocell chipset suppliers, RadioFrame has developed its own application specific chip. It believes this will give it cost advantage as volumes scale up.
• Reducing the chip count. For example, if the same chip can be used for broadband RF processing and control software, then it also saves board space, power consumption and packaging costs. Integrating the femtocell into another product. Femtocell product packaging can avoid the cost of a separate power supply, box and connectors - piggybacking onto an existing consumer device. This also reduces costs in shipping, managing and configuring the system including reduced potential for equipment mismatches and other end-user issues. Using consumer electronics manufacturing facilities. RAN vendors are used to manufacturing hundreds of thousands of basestations per year, whilst mobile phone vendors make hundreds of millions. Femtocells could be in the millions or tens of millions, and so appropriate volume manufacturing facilities can shave costs down. These facilities often have cost-oriented design engineers who can advise on how to save money through minor changes to system design and alternative component choice.
• Timing and Synchronisation techniques. Smarter algorithms and intelligent choice of components allows substantial savings in this critical area. Differentiation can occur by using a range of techniques to recover and maintain an accurate clock frequency using lower tolerance and thus cheaper components.

• Physical appearance (is it trendy/fashionable). I’d say Motorola has the early lead here with some “wacky” packaging, Airvana has bright colours whilst radioframe’s sleek and shapely space model did attract my attention. If these do become common consumer devices, don’t underestimate the importance of colour and shape in the buying decision.
  
• Existing supplier relationships. Mobile operators typically have only two or three radio equipment providers (as opposed to 10’s of handset vendors). They typically prefer to avoid increasing their supplier base because of the overhead involved with handling a major new player and associated technical management in their network operations, although will do so where competitive or cost advantage makes it worthwhile.
  
• Partnerships: A combination of several products from different vendors can take advantage of the benefits of each, but become worth more than the “sum of the parts”.

TrackBack URI for this entry

Comments (0)

Subscribe to this comment's feed