Electronic Design

The Wireless Business Sliced And Diced

Wireless is where it's at. We have single-chip radios for virtually any application. It's cheap, easy, and fun to make virtually any product wireless. Just choose your technology and protocol, and away you go.

From the very beginning, Wi-Fi has been a true winner. In only a few years, this sophisticated technology has evolved into an all-purpose wireless tool. Single-chip, multiband, multimode versions are now available to meet the 802.11a/b/g standards. Prices have dropped to where it's a de facto inclusion in most laptops and some PCs.

The most recent development is continued standard work for improved security, quality of service, and higher speeds. Eyes are on the forthcoming 802.11n standard, which seeks 100-Mbit/s-plus rates up to 100 meters. The new system uses multiple-input, multiple-output (MIMO) radios and diversity antennas to improve speeds. A final spec and standard is expected later this year if agreement is reached. This faster version will open new applications, particularly in consumer electronics where video is key.

We've been hearing about UWB for years now, yet no real products or applications have surfaced. Progress is under way, but a lack of consensus in the IEEE 802.15.3a WPAN group bogs it down. Two camps are vying for the standard, and the current deadlock doesn't seem to be coming to resolution. While the Multiband OFDM Alliance (MBOA) led by Texas Instruments and Intel and over 100 other companies has the critical mass to finalize a standard, the pulse/cdma-type UWB faction led by Freescale and Artimi is still alive and well. It appears the groups will take separate paths and let the market sort out the winners and losers.

But we're still a year or so away from any real quantity of UWB products. It's a promising technology, because no other offers rates to 500 Mbits/s, even if it's over a range of less than 10 meters. Look for UWB to scale to 1 Gbit/s or perhaps more. Despite forthcoming competition from faster versions of Wi-Fi, it should find a niche in consumer electronics.

Most vendors are also seeing a fully wireless USB port with a data rate to the 480-Mbit/s USB max as the killer app. Could a wireless IEEE 1394 version be far behind? UWB is also expected to play a role in mesh networking as an option for short-range communications of very high speed data.

Business and government are finally recognizing that radio-frequency identification (RFID) is the way to monitor inventory, keep track of shipments, identify and locate items, and provide controlled access. It has been around a while in automatic toll road access and gas-pump charge systems. These RF-powered tags use the 125/134-kHz, 13.56-MHz, and 915-MHz UHF and 2.4-GHz bands to permit interrogating and reading distances up to many feet.

Thanks to dropping prices—falling from a couple of bucks each to 40 to 50 cents—many companies can afford to use single-chip transceiver tags. High-ticket items and skids of products can afford them now. But when prices hit 5 to 10 cents, they could replace the bar code.

Walmart is pushing its suppliers to use RFID to identify shipment pallets and eventually individual products. The Department of Defense, which made RFID a mandate for the military by this year, has been adopting it to track, ship, and inventory almost everything. As with many wireless systems that deliver massive amounts of information, back-end software is the key to making it work.

A mesh network is a topology that provides multiple paths between network nodes. Wired networks have used the mesh topology to get redundancy and reliability, but it's messy and expensive. Mesh makes the most sense with wireless, because wireless nodes can be set up to form ad hoc networks that connect any nodes.

If interference or excessive distance between nodes causes a dropped link, the mesh system will find an alternate path through the mesh automatically. The nodes themselves may generate messages to be sent elsewhere or be available to receive data or both. They also act as repeaters to move data from point to point when they're not transmitting or receiving their own data. What results is a very robust network at low cost.

The greatest potential seems to be with wireless sensor networks. Each node contains a sensor to monitor some physical condition, like temperature, pressure, light, and position. This data is then transmitted via the mesh to a central control point where the data is stored, interpreted, and used for followup control purposes. Examples include building monitoring and control, security, factory automation, and public-safety services like fire and police. Even the home is a target for mesh, where almost anything can be enabled for a mesh network.

However, standards and software still require development. The IEEE's wireless personal-area-networking (WPAN) standard 802.15.4 implements mesh networks on a small scale with a technology known as ZigBee. Designed for very low-cost, low-speed applications, ZigBee is being used to pioneer the mesh field.

New standards are no doubt in the future for the larger networks. Cisco and Intel are helping the IEEE kick off a standards study this year. In the meantime, proprietary standards will be developed and deployed until a standard arrives.

As for the software, expect lots of mesh action when the price of the nodes (sometimes called motes), including the transceiver, an embedded controller, and the sensor, drop to less than $10. But how will designers deal with all of the data to be collected? Software must be developed for the application so the data is saved, interpreted, and otherwise processed for whatever purpose. It's going to happen soon.

Two wireless technologies with the greatest promise are smart antennas and cognitive or agile radios. When designing a wireless system, the engineer's greatest challenge is power management and the antenna. Antennas are essential to any wireless system. They're the link to the electromagnetic ether, so the more efficient they are, the better the communications link.

Smart antennas can adapt to system changes and provide variable directivity and gain as needed. By varying beam width and azimuth as well as manipulating multiple beams, they can minimize interference, locate specific radios, and track signals. Thanks to new forms of diversity and DSP algorithms, adaptive antennas are making an impact on all wireless systems.

Cognitive (or agile) radio may bring relief to the spectrum crisis. There's never enough spectrum to assign to everyone. Yet at any given time, 60% to 80% of the spectrum is unused. Cognitive radio identifies available spectrum and nearby services and resources. It then shifts the radio to that unused segment for transmission. These radios would have multiband, multimodulation, multi-access, and multiprotocol capability and could switch from one to another as required. they also will "think" and provide a reliable connection under almost any circumstances.

TAGS: Freescale
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.