High-tech

Uweeba (uwb) - from bandwidth to bucks? Part two

By The Chilli analysts

Editor's note: owing to the large number of requests for part one of the Uweeba market analysis, The Chilli has decided to publish the second part.

How does Uweeba fit in with the other standards?

Who is in on it?

A large shadow cast by Intel

The Chilli perspective

How does Uweeba fit in with the other standards?

Table three: positioning between uweeba and other wireless standards

Uweeba will have to go through a full deployment cycle, encompassing initial deployment, standardisation, maturity (in interoperability and stability), cost reduction, etc. Wi-Fi and Bluetooth are already in the midst of this, with volume deployment and downward price pressure.

Uweeba can provide greater bandwidth than Bluetooth, although for many of Bluetooth's target applications and end products, uweeba's high capacity won't be required. Rather than supplant Bluetooth across the board, uweeba will expand the market for applications that can be targeted by short-range wireless, and will ultimately challenge wired standards such as usb 2.0 and 1394 Firewire, and may complement them by becoming their wireless version.

The situation relative to Wi-Fi for home a/v streaming is more interesting. Although uweeba has a shorter range, it is adequate for home networking, and its higher bandwidth enables several streams of standard definition, mpeg-2, digital video - something that Wi-Fi is currently hard pressed to do. The case for Wi-Fi in the public arena has cooled off due to economics, but it will continue to proliferate in the enterprise as a cable replacement for Ethernet.

Who is in on it?

Table four below summarises some of the leading startups attacking this space. Established companies, including Broadcom, Focus Enhancements, Fujitsu, Mitsubishi, NEC, Panasonic, Samsung, STMicroelectronics, Texas Instruments and IP vendor Parthus are tracking and participating in the uweeba market space. Larger vendors, both fabless and idm (integrated device manufacturer), tend to enter markets when standards are more stable and mature. They also suffer from slower decision-making, resource allocation and a focus on existing commitments. Philips and Motorola have already partnered with startups to get early access to uweeba systems and silicon. In return, the startup gets a strategic partner, wider channels to market and process technology know-how. This could be a potential barrier for startups that do not have a potential partner on their radar screen.

Table four: players in uweeba silicon space

The primary market for uweeba silicon vendors is consumer electronics, with radar and tracking as secondary markets. For this reason, new startups can either wait for the 802.15.3a standard to be finalised before producing silicon or allow enough flexibility in their design to make last minute changes. Vendors such as Time Domain and General Atomics may offer proprietary solutions and gain traction in other niche markets. As uweeba is a phy standard, with the mac determined by the application, some startups will focus on the complete solution with the balance focusing on the phy and partnering with others for the mac expertise. Most vendors claim experience in wireless systems development, with Staccato claiming significant experience in developing all-cmos radio solutions. A key ingredient for successful uweeba startups will be a development team that has previously developed a complete wireless chipset, baseband, rf, software and reference designs.

A large shadow cast by Intel

Currently an influencer in the wpan space for the computing segment, Intel is potentially a very strong player and competitor in the longer term. Intel has been researching a re-programmable, re-configurable radio architecture implemented in cmos that could be targeted at uweeba. The company has also been pursuing development of a combined cmos, analogue and sige 0.09 nm process and can use this to integrate wlan, and later uweeba into its core chipset. This will be the latest instalment of an integration strategy that has encompassed uart, Ethernet and irda functionality.

The Chilli perspective

There are three key questions to ask at this stage:

1. Is there a market for uweeba solutions?

2. Will it create a herd mentality leading to yet another mini bubble?

3. What are the risks and threats?

The answer to question one is a qualified 'yes'. For example, Allied Business Intelligence forecasts unit sales of 45.1m units with revenues of $1.39bn by the end of 2007. Taking a lesson from history, Bluetooth was promoted heavily from 1999 onwards, but it was not until three years later in 2002 that it shipped 35m units - uweeba projections may follow a similar pattern. Assuming that interoperable, standards-compliant chipsets reach the market early 2005, with end products hitting the streets late 2005/early 2006, it is hard to see a ramp-up to 45.1m units in just 24 months. Another factor in any estimate is the fact that uweeba is not going to be a cable replacement in all its target markets - Bluetooth already exists in laptops and mobile phones and is quite sufficient and more cost-effective. Uweeba will therefore have to create its own niche and co-exist with existing wireless communications technologies for some time.

Consumer a/v streaming in the home looks a good bet if the sub $10 bom (bill of materials) and growth inhibitors are addressed as promised, although cross-platform interoperability between content formats and security will also be required - the objective of the newly formed Digital Home Working Group (dhwg). The computing segment will be attractive, with uweeba supplanting wired usb 2.0, perhaps using usb 2.0 for the mac layer. Intel will inevitably integrate uweeba into its core logic at a future date, at which point uweeba vendors will need to diversify into other market segments. Both markets will require startups to develop global networks to work with influencers in Europe and North America, as well as commercial and engineering support activity for the Far East contract manufacturing houses. This requires adequate planning for Chilli R2 and R3 funding, which most startups underestimate at the beginning. Investment in suitable third party representation and alliances should also be planned from day one.

The industrial and automotive segments have a different mix of business factors compared to consumer electronics products, but can provide more stable, albeit lower volume business. The industrial market consists of many small companies and projects with medium-sized volumes - the kind of accounts that are handled by distributors. To attack such market segments, uweeba vendors should consider partnering with a niche distributor from day one to get a feel for the quirkiness of this business segment. The automotive segment has particular challenges and is not best suited to startups. In terms of tracking applications, uweeba will co-exist with rfid technology and presents some interesting opportunities for defence, physical security and biometric applications.

To answer the second question, it is very likely that other vendors will enter the market for uweeba silicon. Investors and entrepreneurs must consider this very carefully, as there are several requirements:

• Domain expertise and competence in rf design
• Reference design expertise, focusing on interoperability, standardisation and co-existence issues
• Expertise in delivering price-sensitive, high volume chip designs
• Investment in Asian representation, channel partners

As reported previously in The Chilli, it takes a total of approximately $25m of total funding for a fabless chip startup to reach primetime. Lets assume a headcount of 60 staff, a fairly typical number for an R2 fabless startup, giving a burn rate of around $6m. Assuming that the chipset average selling price (asp) goes from sub $30 down to sub $10 over the first three years of availability, a lot of chips have to be sold to breakeven. The more vendors participate, the lower the asp, and the more chips you must sell - against more competitors. Sound familiar? At its peak during 2000-2002, the Wi-Fi market had over thirty fabless chipmakers vying for a share of the market. The mass participation of vendors forced a price crash, with some companies exiting the market, and others becoming the subject of fire sales. Clearly, those vendors that can demonstrate the relevant domain expertise and business agility can thrive in this environment. Longer term, the functionality of uweeba chipsets will be mostly integrated into application-specific standard products (assp) for embedded applications. This will provide opportunities for ip aggregators to expand their portfolio and uweeba startups seeking an exit through trade sale to an idm.

The Chilli believes there is an opportunity to avoid an investment bloodbath if vendors partner across the food chain - lowering the bom to increase proliferation, whilst maintaining a reasonable margin.

There is an opportunity for partnering between:

• Startups and idms: the idm gets early access to silicon to kit with its complementary devices. In return the startup gets access to channels, markets, geographic representation and potentially foundry capacity and technology know-how, e.g. rf manufacturing expertise
• Module, antenna, chip vendors and ip aggregators to provide different form factors - pc, embedded - requiring ip, single-chip, module or subsystem solutions, depending on the chosen end market

In terms of risks, the further development of Wi-Fi could potentially hurt uweeba, but the need to improve qos, bandwidth and support mesh networking will take time, and it is not yet clear if the consumer segment has warmed to the existing Wi-Fi solutions in sufficient volume. Its useage model for end users, who are not as tech-savvy as traditional expert computer users, is a weakness. This is where uweeba has the biggest advantage if the body of players can agree a reasonable standard, cost and quality of service. On balance, the coexistence of multiple systems, low cost, transmit power, security (little chance of interception and deception) and ease of use and deployment makes uweeba a better consumer proposition than Wi-Fi.

Comments on this story? Send an e-mail to Editor@theChilli.com

© Chilli Publishing Ltd 2003

04AUG2003