Wireless Local Area Networks have brought about significant benefits for innovative companies in specific industry segments over the last couple of years. The ability to roam untethered around a working location has brought about changes in work practices with resulting efficiencies. However, due to low data rates and relatively complex deployment, wireless applications have to date been relegated to vertical niches such as warehousing, retail, and manufacturing. Using radio is no novelty to industry, but we are witnessing the emergence of newer technologies (chip-sets, adapters, access points) and standards, faster data rates coupled with voice/data integration together with license-free operation in the 2.4GHz radio frequency band making it easier for more customers to deploy and enjoy the productivity benefits of wireless operation.
Factors Influencing the Growth of Wireless LAN Deployment
Ease-of-use (consistent) user-friendly interfaces
Wireless device battery life
Memory and storage
Application and data synchronization
Applications and solutions (end-user utility to drive acceptance)
The growth potential for this wireless technology is particularly interesting in the home and small business markets. In the home, and for the small businesses sector, removing the need to invest in a fixed wiring infrastructure provides cost efficiencies and flexibility, or to quote Mack Sullivan. Managing Director of the Wireless LAN Association: "Because there's no wiring, they take their system with them; they can set up the system themselves and save money on installation, too".
CI estimate that in small companies with less than 5 employees, less than 6% are currently networked. For companies with less than 20 employees, only some 25% are networked and less than half the companies with 20-100 employees are networked. IDC predict an order of magnitude increase in the number of SOHO local networks over the next 3 years to 8 million. In the same timeframe (by 2002), Access Media International expect that the 20 million households that currently have 2 or more PCs will double. Frost & Sullivan forecast that for the wireless LAN market, propelled by the above drivers, over the next five years revenues will triple from the expected $400million this year, with the volume of shipments going up five-fold.
Genesis of the Technology
The first commercially available wireless LAN technologies appeared in the mid to late 1980s. The major requirement at first was mobility-to give employees the capability to roam with a computing device and still have connectivity to the resources available on the corporate LAN. Early implementations were in specialist industries such as warehousing, manufacturing, retail store management, and public utility environments where computing traditionally could not go and yet where significant productivity benefits were to be gained. As a result, early adopters of the technology were able to justify the cost and inconvenience of isolated implementations of proprietary technologies in discrete locations.
Wireless LANs work by superimposition of data on the radio carriers. By utilizing different frequencies, multiple users can coexist in the same radio space. In a typical deployment a transmitter/receiver (referred to as an access point) is connected to the fixed LAN or another fixed location. Wireless users then communicate with this access point (AP) using wireless adapter cards on their wireless information device (WID). This communication is transparent to the LAN operating system. Access points have a range of approximately 150 meters indoors and 300 meters outdoors. The aim is to position access points so that clients might roam freely without losing network contact. Multiple access points can exist to hand the client off from one to another transparent to the client. Similarly, designers may employ an extension point (EP) to extend the reach of a network. An EP differs from an AP in that it is not connected to the fixed network, but in other respects functions like an AP.
The Drive for Enterprise Wireless LAN Standards
As with any new technology, there emerged a few competing and independent proprietary wireless LAN (WLAN) approaches forcing the early adopter community to choose a single vendor technology, and commit to it throughout their corporate deployment. There was also a drive to establish a standard to allow interoperability between various manufacturers' WLAN implementations. The early end-users of WLANs were seeking to influence the vendor community to improve the speed and reliability of wireless LANs to approaching that which is experienced on fixed connection LANs. As a result, the Institute of Electrical and Electronics Engineers (IEEE) was given a mandate to develop a standard for spread spectrum wireless LANs operating in the unlicensed 2.4GHz Industrial Scientific & Medical band. This work was undertaken by the LAN Interoperability Group IEEE 802.
In 1997 the original 802.11 standard for 2.4GHz wireless networks was delivered, with feedback from some sectors of the industry that it had done little to promote interoperability among products from competing vendors as it allows for both frequency-hopping and direct-sequence-spread spectrum-based radios; two incompatible technologies. The standards also failed to deliver the framework for the required speed improvements as it only allows for low data rates (2 Mbits or less) which proved too noticeable a gap between this and the performance to which a typical Ethernet LAN user had become accustomed. Nevertheless, 802.11 products based on Ethernet LAN protocols are now freely available, and are being implemented in corporate networking infrastructure.
In 1991 ETSI (the European Telecommunications Standards Institute) began work on the definition of HIPERLAN/1, a high performance LAN which operates at a higher frequency than 802.11 in the 5.15GHz to 5.25GHz band. The standard was delivered in 1996 and like 802.11 it too is an implementation of Ethernet LAN protocols, but provides greatly superior maximum air interface speed of 23.5 Mbits delivering usable data rates of up to 18Mbits at a range of 50 meters. The issues of quality of service are also addressed in this standard.
Factors End-Users Consider When Deploying Wireless LANs
- Security and reliability
- Desired throughput
- Compatibility and interoperability of wireless devices and the fixed (LAN) infrastructure
- Environmental interference
- Licensing issues (for certain global deployments)
- Desired range and coverage
The IEEE responded with the 802.11 HR (High Rate) DSSS offering a standard which allows industry compliance for 11 Mbps in the 2.4GHz radio frequency band. This allows wireless LANs to be extensions of 10 and 100 Mbps wired Ethernet networks. IEEE 802.11 has emerged as a key element of a new service model taking a step further towards the convergence between the Internet, voice, and data, allowing services and communication constructs to be built where the user is an active and constant participant in the communication space.
Meanwhile the work of ETSI has progressed and HIPERLAN/2 is intended to provide access to IP, ATM, and 3rd Generation mobile networks. As opposed to IEEE 802.11 and HiperLAN1 which were based on wireless Ethernet technology, HIPERLAN/2 is connection-oriented with wireless connections being time-division multiplexed, and as such, aims to provide common connectivity for mobile communications in corporate, public environments, and home to ensure the production of interoperable products.
Close co-operation has been developed by ETSI with the IEEE 802.11 committee and the Multimedia Mobile Access Communications Promotion Council (MMAC) in Japan, which led to agreement on a common standard and approach for worldwide spectrum allocation.
New Approaches for New Markets
Computer and telecommunications technologies that are proven in commercial settings are rapidly absorbed into other markets. Wireless LAN technology is no different with residential and personal area network offerings being introduced. Sometimes a new approach is justified in order to simplify implementation and administration overhead, but often a simplified or reduced functionality implementation would serve a similar purpose.
Residential and SOHO Market
In the late 1990s, a consortium known as the HomeRF working group was formed to focus on creating a shared wireless access protocol (SWAP) for interconnecting data and voice equipment in the home environment. This area was already being addressed by the European Digital Enhanced Cordless Telecommunications (DECT) standard, a Time Division Multiple Access (TDMA) structure which although not specifically aimed at residential use, allows for full-duplex data communications at 552kbps, and was designed to permit short range access to voice, fax, data services via wireless LAN, and wireless PBX services.
The HomeRF definition which offers support for 127 devices and six full-duplex conversations per network, uses 2.4GHz FHSS technology, is compliant with ETS300 328, and compatible with IEEE 802.11. As a result it may offer a useful low-cost alternative for shared, high speed Internet access solutions. HomeRF has also ensured a level of compatibility with DECT, part of the IMT2000 specification which has so far achieved widespread implementation in digital cordless telephones.
Meanwhile the ETSI backed DECT has recently released standards for approval which have important implications for low-cost technology for high-speed data transmission in the 50-300 meters range. One of these, DECT Packet Radio Service (DPRS), is intended to be the basis for all interoperable packet radio services that can be offered over the DECT air interface. Embracing the deployment of home, small office, enterprise, and public services it is proposed as the base standard for simultaneous operation of voice and high speed data transfer in the same mobile terminal. With over 200 products on the market, a total of nearly 45 million terminals delivered by the end of 1999, adoption in 110 countries, recent acceptance as a WAP carrier, and adoption in the U.S. on the 2.4GHz band under the name of Personal Wireless Telecommunications, it is interesting to see how DECT’s use in the domestic WLAN market will develop.
We are beginning to see the proliferation of other low-cost residential WLAN networks based on proprietary 2.4GHz protocols, and a lack of product interoperability seems inevitable, giving the residential customer integration challenges which they are not expecting, and almost certainly not equipped to resolve.
Personal Area Network (PAN)
A few years ago, the telecommunications and computing industries recognized that a truly low-cost, low-power radio-based cable replacement, or wireless link, was desirable to provide the basis for small portable devices to communicate together in an ad-hoc fashion.
A study was performed, and the technology code named "Bluetooth" began to be defined. The goal was to enable an easy-to-use service for mobile and business users by means of a small, short range radio-based technology for integration into production line models of a range of different devices, or in other words a small form factor, low-cost, short range radio link between mobile PCs, mobile phones, and other portable devices to provide a cost-effective solution for the replacement of the many proprietary cables that connect one device to another.
Bluetooth is an open specification for wireless communication of data and voice. It is based on a 9 x 9mm microchip, facilitating protected ad hoc connections for stationary and mobile communication environments. In addition to a hardware description, it also offers an application framework and interface support with interoperability requirements.
For instance, Bluetooth radio technology built into both the cellular telephone and the laptop would replace the cable used today to connect a laptop to a cellular telephone. Printers, PDAs, desktops, fax machines, keyboards, joysticks, and virtually any other digital device can be part of the Bluetooth system. But beyond removing the tethers from devices by replacing the cables, Bluetooth radio technology employs FHSS and provides a universal bridge to existing data networks, a peripheral interface, and a mechanism to form small private ad hoc groupings of connected devices away from fixed network infrastructures.
The Bluetooth world refers to a piconet as a collection of devices connected via Bluetooth technology in an ad hoc fashion. A piconet starts with two connected devices, such as a portable PC and cellular phone, and may grow to seven connected devices. All Bluetooth devices are peer units and have identical implementations. However, when establishing a piconet, one unit will act as a master and the other(s) as slave(s) for the duration of the piconet connection.
In addition to the significant weight of the founder members-Erricson, IBM, Intel, Nokia, Toshiba-Bluetooth has snowballed in popularity and has over 1000 adopters in the industry stewarded by a special interest group (SIG).
The working group IEEE 802.15 resulted from a study group within 802.11 that produced a draft project authorization request for wireless personal area networks to be discussed. The study group has solicited industry input on market requirements and technical solutions for a wireless personal area network (WPAN) with 0 to 10 meter range, data rates of less than 1Mbit/s, low power consumption, small size less than 0.5 cubic inches, and low cost relative to target device. Sectors of our industry are wrestling with the potential overlaps and conflicts between the implementation of Bluetooth and the IEEE work.
Other Specifications Relevant to this Space
The IEEE Wearable Computers activities are peripheral to the wireless LAN space. Worn on the body, they provide constant access to computing and communications resources. Wearable computers are defined as unobtrusive computing devices, networking devices, software, and peripherals, which are worn or carried by individuals and enhance their ability to perform productive work as well as provide entertainment. Input devices for wearables range from innovative text input devices, video capture devices, microphones, GPS locators, affective sensors (blood pressure, GSR, heart beat, EMG), infrared positional beacons and probes, digital cellular modems, PDAs, and so on. Output devices include head-mounted displays, speakers/headsets, PDAs, tactile feedback, speech, and so on.
IAPP (Inter-Access Point Protocol) is a specification that defines how access points from different vendors in an IEEE environment communicate with each other to support roaming. The IAPP specification defines how access points will communicate to hand over mobile stations. This specification provides interoperability when used with IEEE 802.11. It is sponsored by Lucent, Aironet, and Digital Ocean who expect that it will spur the development of products compatible with IAPP to achieve wireless LAN interoperability. The IAPP specification builds on the capabilities of IEEE 802.11 that address the physical and media access control layers of the OSI reference model-the IAPP specification on the other hand tackles higher-level OSI layers such as logical link control to achieve inter-access point communications. In addition to the three sponsors, IBM have also indicated their support for IAPP.
Firewire is another IEEE standard relevant to this world-IEEE 1394. It offers multimedia connection enabling simple, low-cost, high-bandwidth realtime (isochronous) data interfacing between computers, peripherals, and consumer electronics products such as camcorders, VCRs, printers, PCs, TVs, and digital cameras. The attributes of multimedia orientation, self-configurability, peer-to-peer connectivity, and performance of 1394 have encouraged applications that include non-linear (digital) video presentation and editing, desktop and commercial publishing, document imaging, home multimedia, and personal computing. The low overhead, high data rates of 1394, the ability to mix realtime and asynchronous data on a single connection, and the ability to mix low-speed and high-speed devices on the same network provides a range of benefits to the world of wearable computing.
Wireless Ethernet Compatibility Alliance (WECA)
The Wireless Ethernet Compatibility Alliance was founded on August 23rd 1999 by six major players in the wireless market-3Com, Aironet, Intersil (formerly Harris Semiconductor), Lucent Technologies, Nokia, and Symbol Technologies-to certify and promote multi-vendor interoperability of wireless LAN systems.
WECA’s goal is to certify interoperability of IEEE 802.11 high-rate products and promote that standard for SMEs (small to medium enterprises) and the home. Over the past several months, WECA has worked closely with Silicon Valley Network Labs (SVNL) to develop an interoperability test bed. This task was completed in February 2000 and interoperability testing is now underway with SVNL operating as an independent test facility. However, only WECA members can submit products to the lab for interoperability testing. When a product meets the interoperability requirements as described in the test matrix, SVNL notifies WECA, which then grants a certification of interoperability, which allows the vendor to use the Wi-Fi logo on advertizing and packaging for the certified product. The Wi-Fi seal of approval assures the end customer of interoperability with other network cards and access points which also bear the Wi-Fi logo.
Wireless LAN Association (WLANA)
The Wireless LAN Alliance, also known as WLANA, is a non-profit trade association with a declared educational focus, made up of leading local area wireless technology vendors. WLANA provides information about wireless local area networking applications, issues, future directions, and trends, offering a reference resource to customers, press, and analysts. Information available includes industry studies, white papers, application stories, and links to related topics and member web sites. As an example, at the end of 1998 WLANA commissioned a study to research the ROI gained from the implementation of wireless LANs in five industries, with an executive summary of the results available on the WLANA web site. The study indicated a high level of satisfaction with the investment customers, with average payback quoted as less than 9 months.
In addition, the Alliance provides an industry voice to government agencies and third-party vendors. WLANA is committed to establishing wireless LANs as a key component of future local area network technology.
Early in 1999 WLANA announced that it had broadened the scope of its educational efforts to include all forms of local wireless applications and technologies, including personal area networks (PANs) and home radio-frequency networks (HomeRF and Bluetooth) as a response to perceived market needs. Jeff Abramowitz, WLANA President, was quoted as seeing the $450 million local wireless segment of the industry as about to cross the chasm into three broad segments: general-purpose enterprise applications, home networking applications, and ubiquitous personal area connectivity.
In the last few months WLANA has been active in the debate around the NPRM in which the FCC proposes to amend the Part 15 rules regarding the rules that govern FHSS systems operating in the 2.4GHz band. In addition, the FCC is attempting to refine measuring the processing gain of DSSS systems operating in the 2.4GHz ISM band.
The WLANA is opposing the changes to the FHSS rules to curb multiple standard confusion in the wireless market. The IEEE 802.11 committee has established a standard for high data rate (HDR) operation to allow DSSS systems to achieve data rates up to 11Mb/s. The WLANA is trying to dissuade the FCC from expanding the FHSS in an attempt to promote the acceptance of a single standard for high-speed operation. The FCC expects to determine its action on these rules by early 2000.
Wireless LANs Research Laboratory (WLRL)
In the academic world, the Wireless LANs Research Laboratory was established within the Center for Wireless Information Network Studies (CWINS) at the Worcester Polytechnic Institute, in Massachusetts, U.S. The key objectives of the WLRL are to serve as a focal point for wireless LAN technology including issues involving benchmarking and performance, to specify and develop key benchmark criteria and test software, including compatibility, interoperability, and standards verification software and systems. This includes designing and developing installation planning, simulation, and debugging tools for both hardware and software. They publish their output as articles, presentations, conferences, and on the Internet.
Wireless Information Networks Forum (WINForum)
The Wireless Information Networks Forum is an association for manufacturers of unlicensed communication systems, including PCS products. WINForum's membership includes Lucent, Nokia, Hewlett-Packard, Apple Computer, Ericsson, IBM, Motorola, NorTel, and Rockwell. WINForum was active in its FCC efforts with the allocation of 300MHz of spectrum in 5GHz band called the Unlicensed National Information Infrastructure (U-NII) bands by the FCC within the ranges 5150 to 5350MHz and 5725 to 5825MHz. (ISM users also occupy the latter band.) WINForum's work with the FCC is ongoing with the finalization of U-NII bands exclusively for high-speed multi-media services and to lock out these bands for other purposes such as low data rate applications.
Wireless Interoperability Forum (WLI Forum)
The Wireless Interoperability Forum (WLI Forum) member companies span a number of mobile computing product and service suppliers focused on the interoperability of a wide range of wireless LAN products and services to foster the expansion of this market. They have published an open interface specification enabling independent parties to develop compatible products, and have established a certification process for wireless LAN product interoperability, including opening existing interfaces, defining and executing interoperability tests, and increasing market awareness.
Their charter is not intended to conflict with the IEEE 802.11 objective of developing new interface specifications. They intend to develop a transition path to new standards as they emerge, including 802.11 in the 2.4GHz band and other open standards in different bands from the current RangeLAN2-based interface. The RangeLAN2 RF wireless LAN technology was introduced in 1994. It is a 2.4GHz FHSS architecture that operates at a data rate of 1.6Mbps per channel, with 15 independent channels, or hopping pattern, available that allows up to 15*1.6Mbps independent wireless LANs to operate in the same physical space, providing up to 24Mbps of aggregate network bandwidth. The additional bandwidth is critical in maintaining high throughput for each user even when there are a large number of users.
HIPERLAN/2 Global Forum
HIPERLAN/2 Global Forum is an open industry forum which was formed in the third quarter of 1999 with the objective of providing common connectivity for mobile communications in corporate and public environments and in the home. Founding members, Bosch, Dell, Ericsson, Nokia, Telia, and Texas Instruments formed this consortium to promote the HiperLAN2 standard worldwide, and to ensure interoperable products are produced against the standard.
The industry is seeing the major players setting out their stalls for the current and emerging standards in wireless LAN technology. As we have seen from the taxonomy above, the predominant use in wireless LANs is spread-spectrum technology. Even though there are many standards bodies and affinity groups in operation, our canvassing showed that there remain in the market concerns over divergence of wireless LAN standards and development activities which will threaten the interoperability of, and investment in, wireless LAN products.
We document below the issues voiced by those we canvassed which are seen by end-users as barriers to purchase and deployment, or by vendors as inhibitors to the growth of the WLAN industry, and the widespread adoption of the technology.
Spread-spectrum technology is a wideband radio frequency technique that transmits a message on different frequencies that are reassembled by the receiver, technology already in use, as an example, by GPS systems and CDMA telephones. Spread-spectrum improves reliability and security by reducing interference among users sharing a common spectrum. The two most typically in use are FHSS (Frequency-Hopping Spread Spectrum) and DSSS (Direct-Sequence Spread Spectrum). FHSS appears as a single logical channel by using a narrow-band carrier that changes frequency in a pattern known by both transmitter and receiver. DSSS generates a bit pattern called a chipping code in which data is combined with a pseudo-random digital sequence (PRN) which is demodulated by a similar technique employed in the receiver.
Here is the rub-FHSS and DSSS do not interoperate. So what are their relative strengths? FHSS claims scalability as a strength-it is easier for multiple uncoordinated FHSS networks to be co-located (think of Bluetooth). DSSS claims greater range with QPSK/BPSK modulation being more efficient than FSK with a higher data rate (today 11Mbps to FHSS 2Mbps). The FCC is considering a rule change as a result of NPRM 99-231 (notice of proposed rule making) with a final result expected early in 2000. The NPRM addresses DSSS certification procedures and FHSS channel widths. IEEE 802.11 HR will comply with the proposed rule changes. For FHSS, the wider bandwidth would allow increased data rates with an adverse impact on scalability and interference due to the requirement for overlapping channels with a higher hop rate. But this party is not over yet-there is still time for some intense debate before a conclusion is reached.
Lack of Conformance Testing and Certification for WLAN Standards
Until very recently (February 2000), even within what is purportedly a well-defined standard, end-user organizations have experienced issues resulting from a lack of conformance and certification testing program for 802.11 HR equipment. End-users have been frustrated by the lack of guarantees of interoperability even when they take pains to ensure that equipment purchased from competing vendors is based on a common chip-set. This is seen by end-user organizations as a major failing in the deployment of standards. The recent work announced by WECA should go some way to solving the issue for 802.11HR equipment developed by its members, but an industry-neutral conformance testing and certification infrastructure must be implemented for all wireless networking technologies including Bluetooth, HIPERLAN (1&2), and HomeRF.
There is a multiplicity of WLAN technologies being allocated spectrum within the 2.4GHz band. These include IEEE802.11, Bluetooth, DECT, and HomeRF.
HIPERLAN has the potential to interfere with satellite systems operating in the same 5GHz spectrum. To avoid any short-term interference, the HIPERLAN/2 standard states that the transmitter power will be reduced from its current 1 watt to 200mW in parts of the band where conflicts are prone to occur. A longer-term solution is to make more of the 5GHz spectrum available and there are currently public consultation processes in place in most countries.
General consensus among the user community to which we spoke was that they wanted to see some co-ordination between divergent WLAN standards and one worldwide 5Ghtz approach.
Many of the end-user organizations we interviewed were of the opinion that HomeRF is overlooking a basic human behavior dynamic which will threaten its success. Many of the early adopters of WLAN technology in the home will be the same people who have become accustomed to using it in the workplace. Workers that bring their laptop home with them will fully expect it to interoperate with any WLAN technology they implement in their home environment. Early experiences show that observance of IEEE 802.11 is not sufficient to guarantee this and, since the corporate standards are likely to prevail, residential users will be prepared to pay a premium for peace of mind.
Lack of User and Device Profiling Capability
Bluetooth continues to be developed successfully, but feedback from the industry raises concern that the group is not paying enough attention to the important topic of user and device profiling, and the need for a directory which is capable of storing multiple profiles for each device depending on their current mode (work/home/travel/recreation). The Bluetooth SIG should be encourages to work with an organization already well versed in the definition and deployment of directories.
What is required to foster uninhibited growth of the high-speed wireless connectivity market? At minimum the smooth coexistence, and at best the highest possible level of interoperability between Bluetooth and IEEE 802.11 and .15, which requires the oil and water mix of FHSS and DSSS.
Industry effort is required to achieve alignment of activities in WECA, WLANA, HomeRF, and so on, to ensure that differences do not retard market growth, and conformance and certification testing programs are implemented to ensure end-user confidence. The Open Group can leverage both its conformance and certifications program skills and the power of its corporate and government buyers to promote a common industry consensus.