Group 5 - Spectrum Requirements

The Powerpoint Slides from the Workshop are available
(Need a free Powerpoint viewer?)

Jim Nixon Facilitator - RA
Ian Tebby Scribe - RA
Don Pearce - Tele2
A. Rhodes - Probot Comms
David Winder Balloon-A-Gram
C. Walden - Sony
W. Dobbie - BT
Chris Hibbut - On-digital
Peter Simpson - Adaptive Broadband
Paul Francis - Francis Walker & Co.

Syndicate group 5 was tasked with discussing the following questions:

• What are the required bit rates ?

• How many devices will be sold ?

• How much coverage is required, over what distances ?

• What are the re-use distances ?

• Is equipment going to be service specific or will one box serve all?

The syndicate meeting was opened by Jim Nixon (Facilitator) who introduced the subject and outlined the aims of the group.

The group quickly formed into proponents of two types of service. These were Fixed Wireless Access (FWA) services and mobile or nomadic home connection systems. These two types of service could then potentially be subdivided into public or private systems.

It was agreed that the applications to be used (the "Group 1 question") drove the estimation of the spectrum required. The proponents of home connection systems remarked that private home video applications were likely to be a high demand application. These digital video systems are expected to require data rates of approximately 30 Mbit/s and are intended for the leisure market to link, for example, a domestic video camera to a video recorder and monitor / TV / PC as required.

The severe case of the high demand home video applications was considered to be in urban areas, particularly in apartment residences constructed with wooden floors and 'non-solid' walls. Supporters of home video systems stated that unless the whole band were devoted to this type of system unacceptable interference was likely to occur. This view was disputed by many delegates who believe that the band should be divided between applications dependent on demonstrable need for spectrum.

It was questioned whether the home video applications needed immediate consideration, as equipment is not available and may not be on the market until 2002. It was agreed that present-day applications needed allocation soon, but needed to take account of future applications.

One delegate explained that some initial studies into spectrum requirements for home video applications have indicated that if each home used only one channel, then twenty seven 30 Mbit/s channels would be required for a typical reuse plan. The study also suggested that it is likely that each home using the system would require a minimum of two channels, making a total requirement of 54 channels.

Supporters of FWA systems noted that telephony, internet access and e-commerce applications over ATM and Ethernet required access to spectrum now, so a realistic evaluation of future needs of all systems is required. They suggested that market forces should be allowed to define the division of spectrum, as in the USA.

The possibility of using 2.4 GHz 'Bluetooth' equipment for home interconnect / video was discussed and it was noted that although Bluetooth may provide some degree of domestic interconnectivity, it is unlikely to provide sufficient bandwidth or range for home video. This argument was countered with suggestions that video transfer from camera to recorder does not need to be real time, therefore the bandwidth requirement should be significantly less as packet-based systems could be used.

After some considerable discussion, it was agreed that a common channel plan should be adopted in the interests of spectrum efficiency.

This was followed by disagreement whether there would be a requirement for high definition video transfer systems within the home. This question was not resolved but it was noted that any bandwidth-hungry technology, such as high definition video, must make maximum use of technology to give the best possible spectrum efficiency. A number of delegates expressed the view that permitting digital video systems in this band would be effectively providing yet another band for the broadcast of digital TV.

A wide ranging discussion on the efficiency of home video questioned why home video use should require exclusive use of the whole band. This requirement was based on the premise that to be successful, home video systems needed sufficient spectrum to be interference free, as this would assure market success. It was also stated that even if the entire band were to be allocated for home video there would still be a potential of interference in some areas of high density use. Counter remarks were that if all of the available spectrum were insufficient to avoid interference then inappropriate technology was being used and FWA technology could be used for this application in far less spectrum.

It was also noted that the prime areas for FWA system deployment were likely to in regions were ADSL and similar systems are not available. It was decided that FWA systems were likely to be dominant in rural and suburban areas and home video systems would probably be mainly in densely populated areas. Therefore, there was a possibility of regional spectrum allocations, based on need for bandwidth in specific areas, by use of flexible band partitioning dependent on demand. This could give protection for all operators in the designated areas.

There was much support for the argument that internet access would be the major application for FWA systems and this use would be more attractive in the marketplace than home video by offering data rates of between 2 Mbit/s and 5 Mbit/s to the user. The supporters of FWA systems repeated that the FWA market would not require exclusive use of the whole band. It was suggested that 200 MHz of spectrum would provide sufficient spectrum to plan a system giving attractive data rates to users. Co-existence with other applications was seen as a priority and and it was restated that a common channel plan was essential.

The subject of type approval was raised, noting that HIPERLAN Type 2 specified requirements for the MAC layer to facilitate interoperation of equipment. It was remarked that HIPERLAN Type 0 concept offered more flexibility for proprietary systems as it contained only requirements for coexistence of systems. It was stressed that a HIPERLAN Type 0 standard would need to include the requirements of the ERC Decision on HIPERLANs (CEPT/ERC/DEC(99)24).

It was also noted that Europe would see the UK as a testing ground and are likely to follow whatever system proved successful in the marketplace.

Jim Nixon summarised the discussion for the plenary presentation as follows:

1. Spectrum requirements are inextricably driven by the applications to be deployed. Two applications had been considered FWA and home video.

2. Home video system proponents claimed that exclusive use of the entire band was required to minimise the interference potential. These systems may not be on the market until 2002.

3. FWA system proponents claimed that a total of 200MHz was required for national use by a number of operators / service providers. Spectrum for FWA is required now for deployment of systems.

4. Band partitioning could be regional, based on population density and availability of other services, such as ADSL.

5. It was agreed that a common channel plan is required.

Any system must meet the requirements of the R&TTE Directive and the ERC Decision on HIPERLANs.