Intel- Wimax wireless chip

Intel reveals Wimax wireless chip
BBC News | Last Updated: Monday, 18 April, 2005, 13:57 GMT 14:57 UK

Tech giant Intel has unveiled a chip that gives access to wide-reaching wireless networks.

The new chip uses Wimax, which enables broadband connections over several miles, unlike the short range of wi-fi used in specific locations.

The company has announced a series of trials with partners, including BT, around the world.

Wimax is seen as a method of extending broadband reach to whole cities, rural areas and countries with low net use.

It is also seen by some analysts as a potential competitor to mobile phone networks, although such a clash is still some way off.

Intel and other Wimax supporters will be hoping that the release of a chip will do for the technology what Centrino chips did for the spread of wi-fi.

The implementation of wi-fi-ready chips inside laptops by Intel helped propel the rapid growth of wi-fi networks around the world.

[Girl using a wi-fi laptop
Will laptops soon be making use of Wimax technology?]

Sean Maloney, head of Intel's mobility unit, said many trials were under way, but the actual deployment of mobile Wimax would not start this year.

"You'll probably see things at the back end of 2006, you'll probably see some... trials earlier than that," he told the Reuters news agency.

Intel's new chip, codename Rosedale, is expected to power devices to receive Wimax signals.

Unlike wi-fi, which can be set up by individuals to form home networks, or in individual offices, cafes and airports, Wimax is engineered to cover an entire city or physical area.

It offers the potential of always-on, always-accessible broadband networks across a wide area and could have a huge impact on the use of the internet on the move, and in areas such as net telephony.

Scott Richardson, general manager of Intel's broadband wireless business, said Wimax equipment was probably still too expensive for wide adoption, but that Intel and networking equipment makers were working to push equipment costs below $200 (£110) from the $300 to $500 level.

"It's our vision and our strategy to really drive that price point down," Mr Richardson said.

Intel is working with a number of partners to develop standards for Wimax, and early versions of the technology are already in limited use.

A pre-Wimax technology powers broadband connections on the London to Brighton train service in the UK.

"Most of the world does not have (broadband) infrastructure," said Mr Maloney.

"For most of the world this will be how they will get broadband," he said.

Meanwhile, Intel announced on Monday that the first PCs to use its new dual-core chips have gone on sale.

Dual-core chips boost the processing power of chips and the new machines, made by Alienware, Dell and Velocity Micro, are aimed at computer hobbyists and gamers.

Source here

++ Related

Intel's Rosedale chip brings WiMax closer
First network trials based on Rosedale technology will start next year
By Sumner Lemon, IDG News Service
September 07, 2004

here

BT sees DIY WiMax in 18 months
Dan Ilett | ZDNet UK | April 19, 2005, 12:25 BST

With Intel's Rosedale chip now out, the UK's incumbent telco predicts that it will soon be offering self-installation WiMax kit

BT could be offering self-installation kits for WiMax to businesses and home users with 18 months, the telco claimed on Tuesday.

Speaking at the UK launch of Intel's Rosedale WiMax chip, BT Retail's Chet Patel said that the company was looking to mimic the services already available on Wi-Fi and duplicate them for WiMax.

"For fixed broadband, they [self-installation kits] took 2 to 3 years", said Patel, a general manager for the telco. "Today is the start of it for WiMax. It will be a year to two year until the equipment and networks are available. We're hoping it will be 18 months until plug and play is available".

The launch of self-installation ADSL kits in January 2002 helped to drive the take-up of broadband in the UK, as users no longer had to pay for an engineer to visit their premises to install the service.

DIY WiMax could soon be a reality in Tokyo, where communications operator Yozan plans to roll out 600 WiMax base stations, which will be compatible with routers that users can apparently install themselves.

Intel's Rosedale chip, otherwise known as the Intel Pro/Wireless 5116 broadband interface device, is based on the IEEE 802.16-2004 standard. WiMax provides data links at distances of up to 30 miles at a maximum speed of 70Mbps.

If WiMax lives up to its promise, it could solve the dilemma of delivering high-speed Internet connections in areas where the cost of running suitable cables to homes and offices is prohibitively expensive.

Some pre-certified WiMax equipment is already available, but Patel believes the launch of Intel's chip should help to lower the cost of WiMax kit. "The announcement of Rosedale means we have more suppliers and vendors. There will also be lower costs coming through," Patel predicted.

Scott Richardson, general manager of Intel's Broadband Wireless Group, said he thought WiMax modems would fall in price from $500 (£262) to $200 or lower over the next few years. "Our view is we can get this below $200," he said.

At $200, WiMax modems would still be a much more expensive than ADSL, or even 3G datacards. However, they could make sense for those who need a very fast Internet connection, or who can't get other forms of broadband.

Richardson added that the WiMax Forum had begun testing WiMax equipment at its laboratory in Malaga, Spain.

"The WiMax Forum has opened labs for certification, and we're on the clock," said Richardson.

Intel and BT were speaking at a WiMax even at London Olympia.

ZDNet UK's Graeme Wearden contributed to this report

Source here

deviceforge.com | Intel ships "Rosedale" WiMAX chip | Apr. 19, 2005

Intel is shipping its highly integrated WiMAX chip, formerly codenamed "Rosedale." The Pro/Wireless 5116 chip has two ARM9 cores and an OFDM (orthogonal frequency division modulation) modem, and targets low cost, low chip-count access points and gateways supporting WiMAX, an IEEE standard for long distance wireless broadband.

WiMAX (worldwide interoperability for microwave access, IEEE 802.16-2004) was designed to support broadband-like (384 Kbps and up) speeds over long-distance wireless links, and could greatly increase the number of people with fast Internet access, Intel said when it began sampling the Rosedale chip last September.
(...)

Source here

intel.com | WiMAX: Wireless Broadband for the World - An interview with Jim Johnson


"Our role is to make sure you can take your notebook everywhere you go, and wherever there is a network deployed, you can access it wirelessly. Our strategy is to make sure the fundamental technology and products and building blocks are cost-effective enough so that people will deploy [broadband wireless] networks around the world."

—James A. Johnson, Vice President, Intel Communications Group
General Manager, Wireless Networking Group


James A. Johnson, Vice President, Intel Communications Group and General Manager, Wireless Networking Group, recently explained how quickly WiMAX technology has grabbed the attention of the industry and is gaining traction among developers. Mr. Johnson also discussed the role of Intel in developing WiMAX standards, ensuring interoperability among manufacturers, and delivering leading-edge silicon for the next generation of wireless computing devices.

WiMAX: Wireless Broadband for the World
Q: What is WiMAX?
A: WiMAX technology involves microwaves for the transfer of data wirelessly. It can be used for high-speed, wireless networking at distances up to a few miles. The term WiMAX comes from 'Wireless (Wi) Microwave Access (MA).' WiMAX is very similar to Wi-Fi in that it uses the same core technology of wireless modulation developed way back in the '60's and '70's. It's called OFDM (Orthogonal Frequency Division Multiplexing), for those that care about the technical terms.

The real benefit of WiMAX technology is that you can run signals very, very close to each other on wireless channels. You can have super narrow lanes, so you can put a lot of traffic over them and they don't disrupt each other.

Q: How is WiMAX different from Wi-Fi?
A: Although the fundamental technology is the same, over time we can add levels of sophistication to WiMAX. Wi-Fi channels occupy a fixed width of the spectrum. But with WiMAX, we're going to enable the traffic lanes – or channels – to get smaller and narrower. This helps service providers seeking to offer wireless last-mile DSL or cable-type service because they can provide a narrower channel that uses less bandwidth and serve more users. You can take what used to be a fixed Wi-Fi lane and make a bunch more lanes and serve more people.

The other big difference between Wi-Fi and WiMAX – starting right away – is that we're going to use licensed spectrum to deliver WiMAX. To date, all Wi-Fi technology has been delivered in unlicensed spectrum. WiMAX will use one of the unlicensed frequencies, but we're also supporting two other frequencies that are licensed. What that means is that you can turn up the output power and broadcast longer distances. So where Wi-Fi is something that is measured in hundreds of feet, usually WiMAX will have a very good value proposition and bandwidth up to several miles.

Also WiMAX is designed to be a carrier-grade technology, which requires a higher level of reliability and quality of service than are now available in typical Wi-Fi implementations.

Those fundamental differences make WiMAX more of a metropolitan area access technology versus hotspot.

Q: What is the WiMAX standard and when will it be published?
A: We are now in testing and trial of the 802.16-2004 standard, which is the fixed, point-to-multi point broadband wireless access version of WiMAX. The first generation of the technology will allow service providers to deploy fixed broadband services (similar to DSL or cable). We expect to see commercial trials of that standard in the second half of this year. Based on our development schedule, we believe Intel will have first silicon position for 802.16-2004 WiMAX premise equipment – the WiMAX units that get installed at the consumer's home or office.

Later this year we will publish 802.16e, which some people call the 'portable' or 'nomadic' phase of WiMAX. It will offer broadband connectivity similar to Wi-Fi. When you use a notebook in a hotspot or in a building, you'll be able to move around at pedestrian speeds and maintain your broadband connection. Because the 802.16e standard is already largely defined, we plan to deliver our silicon solution at roughly the same time the standard gets published, which will allow deployment either in a trial basis or some commercialization in late 2006. So 'nomadic phase' will roll out about a year after the fixed phase.

Solutions built on the first generation of 802.16e in 2006 won't allow for the fast hand-off like you're used to with a cell phone. The true mobility – which most people think of as high speed roaming and fast hand off – will be ready roughly a year later, in late 2007.

Q: What will users be able to do via WiMAX that they can't do with existing technologies?
A: There are areas of the world - especially in emerging markets and rural areas - where deploying wired broadband infrastructure is not cost effective. WiMAX is very cost effective technology to quickly deploy in the regions which otherwise would not have broadband access. So WiMAX helps spread broadband to more users more quickly than existing technologies.

Another benefit of WiMAX is the ability to get higher connection speeds farther away from the transmitter. Right now you can get a really high speed connection in Wi-Fi close to the transmitter. The other option is that you can get a pretty slow Internet connection using a cellular technology, which spans a greater distance. WiMAX fits between those two offerings. You'll get speeds similar to close-up Wi-Fi connections out to several miles away from the transmitter.

WiMAX will also be much easier to install, which makes it more cost-effective for service providers and hopefully some of those savings will accrue to users. For example, with 802.16-2004, service providers will be able to offer users last mile access with an external antenna mount. Shortly thereafter we expect to offer that same service so people can put that antenna inside on their desk near their window. Either way, it's easier to install than Wi-Fi, which requires precise alignment between access points. Our goal here is to enable self installation.

When 802.16e comes out in 2006, the improvements become more obvious. This is where we're adding scalability and multi-access capabilities to the fixed OFDM technology. We'll make the channels scalable and the lanes different sizes to extend broadband wireless access across a larger geography. As I mentioned earlier, fixed wireless access is known as OFDM, and the industry term for this scalable technology is SOFDMA – or Scalable OFDM Multi-Access. From the technical side, it's a pretty big change. So it's very similar in concept to today's more evolved, multi-access cellular technologies.

Q: How does WiMAX compare to 3G?
A: 3G has been built on the foundation of a voice network. And the 3G community is adding data capability to it. Our objective for WiMAX 802.16e is to be a high speed data service that can be used to extend and complement 3G service. We know of several 3G service providers that are worried that as data use grows they will not have enough spectrum for both their voice customers and their data customers. So they're interested in WiMAX as a complementary data service technology that they can deploy with their 3G voice service. So essentially, WiMAX is very targeted for wireless data, not wireless voice.

There's another big difference between WiMAX and cellular technologies in general. Cellular grew up – more or less – where the whole system was architected from the network and base station all the way out to the client – typically the phone – and different vendors' equipment didn't necessarily work together. Eventually there were a lot of separate networks. Roaming agreements have been put in place, so that more so and more so you can work on other people's networks, although you'll be charged extra for it. With WiMAX we're working to get that type of interoperability and more - from Day One. That means if you buy a client made out of silicon by Intel, you could use it on several different networks even if the base stations are provided by different companies.

Q: Intel is said to be working toward a future of "anywhere, anytime connectivity". Where does WiMAX fit into this vision?
A: Today, Wi-Fi kind of lives by what we call the "five minute rule." If you live in a city, most likely you can walk five minutes and find a hotspot. Or if you're in your car in the suburbs or a village, you can usually drive within five minutes and find one of those. With WiMAX we're trying to offer that same type of service without having to drive or walk five minutes. Eventually, you can just open your notebook and get a connection, wherever you may be.

When WiMAX is fully developed, you'll no longer be limited to 300 feet within the Wi-Fi hotspot. And you won't have to drive around looking for a connection. Even though it's only five minutes, it's still five minutes, and that's just not as natural as getting a connection anywhere.

When people have a broadband connection they tend to use their computer more, they leave it on and they integrate it more into their lifestyle. WiMAX technology extends the range of broadband wireless access to more users in more geographies. This happens first with last mile connections where anyone wants them, and eventually in notebook mode for mobility. We believe WiMAX can be deployed in any metro area, using the licensed spectrum that we are working with today, and it moves around the world.

Q: What will it take to put WiMAX out around the world?
A: We really only need to make sure we have access to spectrum – the same frequencies – all the way around the world. We're specifically working on three frequencies for WiMAX deployment worldwide. 5.8 GHz unlicensed is the same spectrum as Wi-Fi, and so all the benefits of that Wi-Fi spectrum can be made available with WiMAX technology. However, because it's unlicensed, we'll be limited on the distance we'll be able to transmit. In the licensed frequencies we've targeted for WiMAX – 2.5 GHz and 3.5 GHz. The WiMAX community and Intel are working with various governments around the world to allow the use of WiMAX in this spectrum.

Q: What is Intel doing to support the development of the WiMAX communications standard?
A: Intel is one of the primary leaders in the development of WiMAX standards. At the first level, the IEEE standard is known as 802.16-2004 and eventually 802.16e. We're very active in leading that standard development effort, which has grown quite a bit. There's another standards body called the WiMAX Forum, which will drive interoperability and some of the standards in terms of network capability above the IEEE standard. One of Intel's managers is the president of the WiMAX Forum, which has grown to over 200 companies, with some of the biggest names in communications you'd expect. We are seeding that body with a WiMAX network using the first draft ratified by several Forum members. Included in that network are the pre-published standards for network security, roaming, and those types of capabilities that actually make it a workable solution.

The broadband wireless industry's been around for about 10 years doing proprietary implementations, so it's really not that different to bring it out in standards. By introducing standards to broadband wireless technology – via the 802.16 standards – we will bring the cost of equipment down and make it more universally available. What Intel is bringing to the party are open standards, which means interoperability and traditionally lower cost equipment. And we're also delivering silicon integration, which also delivers lower cost equipment.

Q: What products and technologies does Intel have planned to support WiMAX-based computing?
A: In September of 2004 we announced our first WiMAX silicon code named Rosedale, and we began sampling to OEMs at that time. Our 802-16-2004-based silicon solutions are focused on the premise equipment. Most of these equipment manufacturers have been making products using proprietary wireless broadband access technology for quite some time, and they already had their base station and transmitter solutions underway. So our role there is to develop lower cost premises equipment and make sure it interoperates with the base station as part of the standard. So that silicon is in lab trials at all of our OEMs today. And we expect that to turn into commercial solutions the middle of next year.

When 802.16e is ready, Intel will develop silicon again for premise equipment as well as notebooks as part of Intel Centrino mobile technology. We also plan to develop silicon for the neighborhood transmitters. As you deploy out a metro area, we believe there will be many, many smaller transmitters that might be situated on lampposts or on the top of neighborhood buildings. We'll offer the silicon and software solution for those devices. And we'll work for larger base station companies as they develop their larger transmitters that they usually use in cellular technology.

Q: What are some of the challenges for WiMAX moving forward?
A: In terms of challenges, we've talked about some already. Number one is spectrum. We're pretty optimistic. There still is work to do, but we're committed to do it.

Another challenge is meeting people's expectations. There is a lot of excitement for WiMAX now – ahead of the standard, ahead of products, and ahead of the deployments. There's as much excitement now for WiMAX as there is for Wi-Fi, even though Wi-Fi has been around for about eight or nine years. I think our biggest challenge at Intel is to manage this excitement so that the industry doesn't get discouraged. We'll continue reinforcing the timeline for WiMAX, clearly explaining what is going to be available and when it will be available. We believe WiMAX is on a very fast track, and it will continue to add more and more capability every generation. But it will take time to get these capabilities to where service providers and users can really benefit. And that's why we continue to talk about our three-phase rollout and implementation.

Lastly, our mission is a global standard, so that we can deliver a global SKU as part of Intel Centrino mobile technology, so that similar to Wi-Fi, you can get a connection regardless of who made the network and the equipment. We want the same thing with WiMAX. You won't need to carry four network cards as you travel around the world. That's the goal, but there's a lot of work to make that the reality.

About James A. Johnson

James A. Johnson is Vice President, Intel Communications Group, and General Manager, Wireless Networking Group. In this capacity Johnson leads the team that is responsible for the architecture, design, development and marketing of Intel's Wireless LAN and WiMAX solutions.

Prior to his current position, Johnson was General Manager of Intel's Platform Networking group and held various engineering and management assignments in the Technology and Manufacturing Group and the Networking and Communications Group. Johnson was on the team that launched Intel into the networking business. He was responsible for the initial production of networking products and manufacturing ramp at sites in the United States and Puerto Rico. Other positions held include Plant Manager for Intel's PC and server manufacturing plants worldwide and General Manager for three different networking and communications businesses.

Johnson joined Intel in 1984 after receiving his bachelor's degree with honors in industrial engineering from Oregon State University in 1984. Johnson sits on the advisory board for Oregon State University's College of Engineering.

Source here




intel.com | Intel's New WiMAX Silicon Solution Expands Broadband Wireless Offerings

Overview: Technology Can Reach Billions More

Intel has released its first WiMAX product, providing equipment manufacturers and carriers the ability to deliver next-generation wireless broadband networks around the world. Previously codenamed "Rosedale," the Intel® PRO/Wireless 5116 broadband interface device (herein referred to as "Intel PRO/Wireless 5116") is based on the IEEE 802.16-2004 standard, giving carriers and end-users the confidence that equipment from different vendors will work together.
(...)

here

intel.com | Intel® PRO/Wireless 5116 broadband interface

Overview

Innovation for WiMAX Fixed Wireless Broadband Platforms

The Intel® PRO/Wireless 5116 is a highly integrated, IEEE 802.16-2004 compliant system on chip (SoC) for both licensed and license-exempt radio frequencies. The unmatched level of integration streamlines the design process and delivers a solid foundation for the development of cost-effective customer premise equipment (CPE).

When combined with third-party RFICs and power amplifiers, manufacturers can create a broad range of outdoor and indoor self-installable WiMAX modems and residential gateways capable of delivering high-rate IP-based data, voice, and real-time video. To further reduce cost and speed hardware product development, Intel PRO/Wireless 5116 hardware implementations will be available through third-party ODMs.

here

intel.com | Intel and WiMAX, Accelerating Wireless Broadband
here

zdnetuk | Inside Intel's Pentium M/Centrino | Rupert Goodwins | March 12, 2003

How does Intel's Pentium M processor and Centrino kit of notebook parts deliver new levels of performance and battery life? Our Tech Guide gives you the details.

With the launch of Intel's Pentium M/Centrino technology, the company is putting its silicon where its mouth is. Intel has long said both that the notebook market will be one of the major growth areas in the future, and that performance need not be sacrificed for portability and battery life. Unusually, the chip giant has been quite open prior to launch about the technical details of the project codenamed Banias (the major redesign of the Pentium for the portable market), and our initial tests suggest that Intel's promises about performance and battery life appear to have been kept.

Banias is now called the Pentium M, while the chipsets formerly codenamed Odem and Montara-GM are now the 855PM and 855GM respectively. The wireless side is handled by Intel's PRO/Wireless 2100 Network Connection Mini-PCI card (11Mbps 802.11b now, with faster 54Mbps variants to follow). The whole family is called Centrino, and while manufacturers are free to take individual parts, they can only get the approved logo (the one that looks like a set of dart flights) if they use a complete set.


Intel's Centrino family comprises the Pentium M processor, the 855 chipset and a PRO/Wireless 2100 Network Connection Mini-PCI card. A notebook vendor can only use the logo (above) if a system uses all of these Intel components.

Pentium M architecture

The Pentium M will be available initially from 1.3 to 1.6GHz speeds, with low and ultra-low voltage variants at 1.1 and 0.9GHz emerging later. Those are maximum speeds: like previous Intel mobile processors, they're designed to run at slower rates to save battery life. One of the Pentium M's innovations is that different components within the processor chip can effectively shut down altogether when not in use, even during a single clock cycle. Intel says that most of the processor is, in fact, turned off most of the time.

The same is true of the Pentium M's very large Level 2 cache, where most of its 1 megabyte is kept in a state as close to unpowered as possible. It has dedicated circuits that keep track of recent accesses, working out how much of the circuitry can be stood down. When applications or the operating system use data from a small section of memory, the entire cache can be disabled on the fly and the data produced from a small, dedicated part of the chip -- effectively a sub-cache.

Lots of power is normally wasted doing pre-fetch and branch prediction -- two high-performance mechanisms where a processor gets information in from memory ahead of time whether it's going to be needed or not. Intel has analysed billions of lines of existing code to make the best guess as to when these mechanisms are needed, so reducing wasted memory accesses. Normal processors can waste up to 30 percent of their time recovering from mistaken predictions: the Pentium M has what Intel calls the 'best-in-class' branch prediction, with many independent circuits checking their areas of expertise and combining to produce a best guess that's around 20 percent more efficient than previous efforts.

Another standard part of processor architecture is the stack -- an area of memory that contains temporary data and memory address information. It's used intensively by all software, but the instructions that manage the stack have to go through the same processing as any others. The Pentium M spots these instructions and very early on hands them over to a special hardware unit dedicated to stack management. Intel says that this removes around 5 percent of the load on the rest of the processor.

Likewise, the Pentium M reduces processor load by combining different aspects of the program flow into single internal operations, a process called 'micro-op fusion'. Thus, multiple instructions look like single operations for most of their life within the processor, only being split up into their independent bits just before actually being executed. This creates a further 10 percent reduction in the amount of time the chip spends processing operations.

All of the above not only reduce power consumption but helps efficient processing, says Intel, and our tests bear this out -- a 1.6GHz Pentium M outperforms a 2.4 GHz Pentium 4M while delivering around 50 percent longer battery life. At full tilt, the 1.6GHz part takes 24 watts; in its most severe power saving mode it runs at 600MHz and consumes just 6 watts. The ultra-lower voltage variants are expected to manage on around half a watt at the full-blown economy setting. Manufacturers can take this advantage either as extra battery life with the same size and weight of machine as before, or keep the same battery life and make their notebooks smaller and lighter.


This graph summarises the performance and battery life advances made by the new Pentium M/Centrino notebooks. The previous generation of Mobile Pentium 4-M notebooks (green dots) were good performers (30-50 on Business Winstone 2001), but typically delivered between 2 and 3 hours' battery life. Older Mobile Pentium III-M systems (blue triangles) could exceed 3 hours' battery life, but performance was lower (20-40 on Business Winstone 2001). The three new Pentium M notebooks (pink diamonds) we have tested so far occupy the hitherto-elusive top-right quadrant of the graph, where performance is excellent (>50 on Business Winstone 2001) and battery life long (3-5 hours).

855 chipsets

The 855 chipsets also have numerous power saving tweaks, driving LCDs much more effectively and monitoring the power requirements of different system components, reducing power where possible. They are very similar, with the 855PM supporting off-chip graphics via a power-managed AGP 4X bus and the 855GM having on-chip support for LVDS output, CRT and two DVO ports. Otherwise, both chips have the same features, including a memory controller for up to 2GB of 200MHz or 266MHz DDR DRAM, in four banks, and -- of course -- extensive power management. Both chips talk to a further Intel chip, the ICH4-M, which handles input/output.

The ICH-4M has a built-in 10/100Mbps Ethernet controller, a 33MHz PCI interface, AC 97 v 2.3 for audio and modem, USB 2.0, and the Low Pincount Bus (LPB) for serial support and security token interfacing. Intel's reference design for Centrino systems expects the PCI bus to be used for its 1000MT Gigabit Ethernet product -- and the company would prefer designers to do that rather than use the ICH-4M's built-in slower Ethernet. Intel also expects the PCI bus to support CardBus and Mini-PCI, with CardBus being used for GPRS or GSM interfaces and Mini-PCI connecting to the wireless LAN solution, which is codenamed Calexico. The first wave of Centrino systems only offer 11Mbps 802.11b via a Philips Wi-Fi chip, but Intel's forthcoming wireless silicon will support 54Mbps 802.11a as well. Calexico also includes diversity reception, where it chooses the best antenna out of two for a connection depending on signal strength, and reduces interference from other devices on 2.4GHz and 5GHz via integrated narrow-band filters. An 802.11g version and a tri-standard (a/b/g) variant are expected later this year.

Intel is pushing its Wireless Coexistence Solution for Centrino notebooks, which reduces or removes interference between 802.11b and Bluetooth in the same system. This involves being careful about placing antennas and understanding their mathematical relationship, and includes some hardware and software additions in both the 802.11b and Bluetooth components.


Source here