WiMAX & family: threats and opportunities


Senior Member
Sep 22, 2003
<i>From an article published by Goldman Sachs, 7 September 2004. (excuse the copy and paste of selected parts of the report)</i>

<u><b>Europe Telecom Services</b></u>
Lots of noise, but only modest impact. Near term, wireless broadband market share is unlikely to be significant, but could grow sharply in the medium term if mobile operators with strong brands use it to attack DSL. Unlike its predecessor technologies, it works and is arousing operator interest.

<u>Wireless Broadband provides an efficient, carrier-grade service</u>
The wireless broadband technologies covered, UMTS TDD (IPWireless),
802.16 (WiMAX), Flarion/802.20 and CDMA450, provide high-speed data connections of 1-3 Mbps in cell sites with a radius of 1-3km without the need for line-of-sight which is expensive to accommodate. They are specifically designed for carrier-grade service.

<u>Not tested for mass market yet, but a threat to DSL and 3G medium term</u>
Current commercial services are modest and the key mass market test is still outstanding, but we believe the technology is ready to compete with DSL for the local loop. We expect the technology to evolve to include voice over IP and in due course to grasp a share of the mobile data pie, affecting 3G operators. Mobility is unlikely to be included in commercial systems before end-2005 and spectrum availability remains a major issue and limiting factor.

<u>Operator interest shows potential significance of the technology</u>
Vodafone, Nextel and other major operators have shown interest in wireless broadband with both trials and live services. No common strategy has yet emerged, but we believe that cellular operators see a potential commercial opportunity (and threat) here and are quite likely to deploy WiMAX-type wireless broadband systems themselves, with WiFi at arms length.

<u>With spectrum being a major constraint, we expect competition for it</u>
Although the 3.5 GHz spectrum has generally been reserved for fixed wireless services in Europe, licenses in this band prohibit the offering of mobility (through cell hand-off). The more attractive spectrum bands are those used by UMTS (WCDMA) and planned for UMTS
expansion, making actual available licenses scarce.
<hr noshade size="1">
Fixed wireless failed miserably a few years ago but the technology has improved and is arousing operator interest. Such new broadband wireless technologies help to blur the distinction between fixed and mobile and actually have several advantages over DSL and UMTS. Indeed, we believe existing operators like Vodafone may themselves explore and ultimately embrace such new technologies, in addition to UMTS, to enhance their breath and range of services to customers.

Nonetheless we expect these new wireless technologies to evolve to include VoIP and in due course grasp a share of the mobile voice and data pie, impacting 3G operators. But the impact will only be felt over the next three to five years.

We believe that emerging wireless broadband technologies such as UMTS TDD (IPWireless), 802.20 (backed by Flarion), 802.16 (WiMAX) and CDMA 450 mainly pose a threat to fixed-line operators as they provide an alternative to the local loop, particularly as a replacement of DSL.

However, given the relatively small scale of the operators using the service so far, the actual impact may only be modest in the near term. For the mobile operators, the technology poses a threat more in the medium to long term unless embraced by the operators themselves, for example using their allocation of UMTS TDD spectrum and their strong brand to attack DSL.

In the short term however, wireless broadband has several advantages over DSL as well as UMTS, partly due to technical flaws in both technologies relating to “crosstalk” on DSL and data capacity constraints on current UMTS equipment.

Out of the four standards we focus on in this report, UMTS TDD looks the most promising in the short term given it is incorporated in the 3GPP specification and almost all UMTS license holders own some TDD spectrum in the 1.9-2.0 GHz band.

WiMAX, often wrongly used as a synonym for all wireless broadband technologies,is still in the specification process at the IEEE, and thus no live operations exist and equipment is not expected to become available until 2005. Flarion developed a proprietary OFDM system which is expected to form the core of the 802.20 standard as and when completed – probably not before 2005/6. CDMA 450 potentially offers
an attractive alternative for providing high-speed data services in rural areas and it is gaining some momentum mainly in the Nordic countries and Eastern Europe.

Having learnt some lessons from the WLL debacle, equipment manufacturers started to focus on standardizing their technology within either the IEEE or the 3GPP. The largest group of companies is organized in the WiMAX Forum working on the IEEE 802.16 “WiMAX” standard, of which Intel is the most publicly outspoken supporter. Alvarion, a supplier to Firstmark, is part of this group and the company claims that its technology will ultimately be compatible with WiMAX.

WiMAX is still in the standardization process and thus compatible equipment is only expected to become available commercially over the next 18 months. The original version of the standard, 802.16 was amended in January 2003 to become 802.16a. It is effectively a fixed-wireless local loop technology requiring outdoor aerials mounted by
engineers, thus making it suitable for business customers and for providing backhaul links. Additionally, BT has said that it will use the proprietary Alvarion (fixed-wireless) technology to provide broadband coverage in rural areas of Scotland and Northern

In June 2004, a revised version, 802.16 Rev d, was specified, which is much more advanced given it supports portable user devices and works with much smaller aerials. The next revision of WiMAX, 802.16e, which will support fully mobile applications, is currently being specified by the WiMAX Forum and it is expected to be finalized by the
second half of 2005, with compatible equipment not being available until 2006 the earliest.

Given that the technology has not been fully specified and due to the fact that there is neither a sufficient variety of equipment available commercially or any live network in operation, we believe that among the technologies covered in this report, it is the furthest
away from having a commercial impact, although currently at least seems to capture most of the hype surrounding wireless broadband.

Another technology which is already commercially available is Flarion’s FLASHOFDM. It is a proprietary technology, though Flarion is now trying to standardize it via the IEEE 802.20 working group for Mobile Broadband Wireless Access (MBWA). However, progress towards a standard has been difficult due to internal conflicts of interest within the working group, which particularly in 2003 delayed the group’s work significantly, causing it to fall further behind WiMAX/802.16, its rival for the first IEEE wireless broadband standard. The group’s list of backers is significantly shorter than that of WiMAX/802.16, and although it is supported by Cisco, it is currently not capturing as much public attention and hype as Intel-backed WiMAX.

The Flarion technology works in different frequency bands (400 MHz-3.5 GHz), with the UMTS spectrum (1.9-2.1 GHz) being the ideal licensed band from the company’s point of view. However, as pointed out above, the UMTS spectrum itself, both the FDD and the TDD part, cannot be used for any non-WCDMA technology, thus making actual
deployments of FLASH-OFDM in Europe difficult.

Although Flarion’s technology can be licensed by independent equipment vendors, interest from third parties proves to be less than enthusiastic according to the company as it is still seen as proprietary given that the IEEE standardization process is still ongoing. The company is currently providing a Wireless Network Card in PCMCIA format for laptop connectivity and it has internally developed a prototype Mobile Broadband Handset. The network card achieves download speeds of 1-1.5Mbps and uplink speeds of

Although the network technology was originally optimized for packet data transmissions, with a few enhancements it can also be used for Voice-over-IP services, which would mean Airdata users could give up their fixed-line completely, making the product more valuable and attractive. However, the company says that it does not consider VoIP on its network mass market quite yet given the negative perception many consumers still have about VoIP due to bad experiences with the first generation of Internet-based VoIP applications in the late 1990s.

It has established the first partnerships to provide a VoIP product, starting with SIPGate, the VoIP service of Indigo Networks in Germany. To be able to use the service, customers need to purchase a VoIP adaptor from SIPGate for EUR99, connecting their standard analogue phone to Airdata’s aerial. For EUR8 per month, customers get free onnet calls and 100 min of calls to domestic fixed and mobile networks, though other packages without monthly charge are also available.

An important factor determining the voice quality on IP networks is the network’s latency, i.e. the average roundtrip delay of a packet. This is currently less than 100ms on Airdata’s live network in Stuttgart. As a reference, the ITU says that carrier-grade voice
services should achieve a latency of better than 100ms for acceptable voice quality. The current version of the IPWireless technology (v5.01) can prioritize voice packets, though Airdata says that only when dedicated bandwidth can be made available for voice and a
more efficient voice codec is available (G729, also used Skype), will the quality improve sufficiently to push VoIP as a commercial product.

The new version 5.1 is expected to be implemented by the end of 2004 and only software updates are needed both in the network and the CPE. The company expects the latency to improve to 50-80ms across its whole network and with the new version it expects to be able to offer a guaranteed Quality of Service for users on premium packages.


Expert Member
Aug 6, 2003
Looks good.

When is WIMAX gonna be available, and is it going to be affordable? E.g setting up a neighbourhood WLAN?

Chow, Nick

Nick Smit
broadband@nicksmit dot za dot net


Senior Member
Sep 7, 2004
"equipment is not expected to become available until 2005"

Sheesh people, to much scanning and not enough reading.


Expert Member
Aug 6, 2003
Yeah, soz... But: http://www.wimax.com/
And, it is relatively near 2005.

Chow, Nick

Nick Smit
broadband@nicksmit dot za dot net


Well-Known Member
Oct 4, 2003
Client devices will be available end 2005 for $100.
Intel owns key patents and licensed it to as many developers
as possible. Intel would like to see client devices for
free if they could. Intel uses Wimax as a means to drive
their laptop sales. AT present client devices cost $1000.

Wimax is the one thing that could save us from the crime wave
in our suburbs. God help us when the ANC finally confiscates all
guns and residential CCTV/WIMAX are not installed... Just how the
farming community will survive nobody knows.