The IEEE has assigned the task of developing MAC and Physical Layer standards for wireless LANs to the 802.11 working group. The working group formed in late 1990 but produced very little in its initial three years of life. Comprising almost 200 members, it could not agree on foundational issues. The major disagreement centered on the issue of the type of MAC to be standardized.
Wireless MAC schemes can be coordinated in either a distributed or centralized fashion. The former is called distributed coordination and the latter is called point coordination. Distributed coordination requires the individual devices to share responsibility for arbitrating access to the LAN. This is the model used in most wired LANs. In point coordination systems, a centralized device controls access to the medium.
In November of 1993, the issue was finally resolved when the 802.11 committee voted on a MAC standard based on the proposal jointly submitted by NCR, Symbol Technologies Inc., and Xircom, Inc. The conflict between distributed coordination and point coordination was resolved by including both models in the standard. The IEEE approved the initial 802.11 standard in June 1997.
The fundamental MAC scheme is a distributed coordination function, which is essentially a contention-based MAC scheme implementing a version of CSMA/CA. In this scheme, collision avoidance includes a Request to Send (RTS) and Clear to Send (CTS) exchange. After determining if the medium is free and calculating an appropriate wait time, an RTS is sent to indicate the intent to transmit and the length of the transmission. If a CTS is returned, frame transmission begins. All other stations are aware that the transmission is taking place and know the expected length of the transmission, thereby avoiding collisions. A point coordination function (PCF) scheme is also supported, which operates more on the centralized polling model (i.e., a central point controls access to the medium).
The LAN is deployed over a wireless logical bus topology. Options available for the Physical Layer include infrared, direct sequence spread spectrum (DSSS), frequency hopping spread spectrum (FHSS), and orthogonal frequency division multiplexing (OFDM). The original standard only supported infrared and spread spectrum. The spread spectrum operated in the 2.4 GHz band and supported 1 Mbps with an option of 2 Mbps in “clean” environments.
Developments in the standard have moved more rapidly since its early days. Transmission rates up to 54 Mbps are now supported, with work underway for transmission rates up to 200 Mbps.
The Wi-Fi Alliance promotes the use of wireless LANs and provides certification of vendor interoperability.
As with any of the IEEE working groups, the 802.11 working group for wireless LANs has several task groups that focus on specific areas. Each task group contributes to the standard or standards that emerge from the working group. The accompanying chart identifies the more significant task groups in the 802.11 wireless LAN efforts.
The 802.11 standard defines the requirements of the Physical (PHY) and Medium Access Control (MAC) Layers of the OSI network model. Separate task groups worked to define these requirements. The original 802.11 standard, released in 1997, was developed from the efforts of the two groups defining these layers' requirements. See http://www.ieee.org for more details.
- The PHY task group defined three different physical implementations as part of the original 802.11 standard. The three choices were infrared as an optical option, and DSSS and FHSS as radio frequency choices in the 2.4GHz band. Data rates were defined with peak rates of either 1 Mbps or 2 Mbps.
- The MAC task group defined the frame structure and the access control scheme to be used. The frame structure is similar to Ethernet, but more robust in that it provides controls for acknowledgments, fragmentation, security, and power management. The access control scheme utilizes CSMA/CA.
- The 802.11a and 802.11b task groups defined enhancements to the Physical Layer to provide for higher speeds. The a task group defined a new Physical Layer implementation utilizing OFDM. It operates in the 5.8 GHz band and allows for speeds up to 54 Mbps. The b task group uses the DSSS modulation technique from the original 802.11 effort, although with a different keying mechanism. It provides for up to 11 Mbps. Both the a and b enhancements were included as addendums to the 1999 edition of the 802.11 standard.
- The e task group provided QoS mechanisms in 802.11 LANs. It defined enhancements to the MAC Layer, defining service classes and augmenting the PCF to be friendlier to QoS applications.
- The hybrid control function (HCF) defined in IEEE 802.11e enhances the PCF defined in the original specification. As with the original PCF, the access point (AP) issues a periodic beacon that defines time windows that include a contention period (CP) and a contention free period (CFP). During the CP the devices in the basic service area (BSA) use the EDCF to arbitrate access.
- During the CFP, they use the enhanced PCF defined by IEEE 802.11e. The basic function is enhanced over the original PCF in four ways. First, traffic classes are defined to give preferential access to higher priority systems. The AP is also no longer required to poll in a round-robin fashion but can use more sophisticated strategies to select the next device to be granted access. This is enabled by the third enhancement: the devices in the BSA can provide the access point with queue depth information that reflects their need to gain access to the medium. Finally, devices in the BSA can be granted access to transmit more than one frame at a time.
- The 802.11g task group defined enhancements to the Physical Layer, this time using OFDM in the 2.4 GHz band. It provides 20 Mbps in the 2.4 GHz band.
- The 802.11i task group provided improvements in security by enhancing the MAC Layer capabilities for encryption and authentication.
- The 802.11n task group is an ongoing effort to provide higher speeds in the Physical Layer (54 Mbps). The draft standard was accepted in January 2006.
- The r task group is working on a handoff process for mobile Wi-Fi.
- The s task group handles Wi-Fi mesh networks.
The Physical Layer defined by the IEEE 802.11 PHY task group provides for the actual transmission of data between wireless nodes. The original 802.11 standard released in 1997 provided for three Physical Layer options. One is an optical-based transmission using infrared light. The other two are radio frequency (RF) options: frequency hopping spread spectrum (FHSS) and direct sequence spread spectrum (DSSS).
The infrared option provides for a 1 Mbps peak data rate, with an optional 2 Mbps rate. It uses pulse position modulation (PPM) to achieve these rates. Infrared products have all but disappeared from the market, leaving only the RF options as critical. Furthermore, 1 and 2 Mbps transmission rates have also faded except as a fallback for higher speed WLAN options.
The 802.11 Physical Layer is divided into two sublayers, regardless of the actual transmission option being used. The top layer is the Physical Layer Convergence Protocol (PLCP). The PLCP maps the payload (the actual MAC Layer protocol data unit (MPDU)) into a physical protocol data unit (PPDU) suitable for the transmission option being used. It adds a PLCP preamble and a PLCP header to the MPDU to generate the PPDU for transmission. The preamble helps with synchronization between devices, allowing a device to sense when transmission is possible. It also acts as a start of frame delimiter. The PLCP header contains fields that identify the speed being used and the length of the payload. The actual format of the preamble and the header will vary according to the transmission option being used.
The bottom sublayer is the physical medium dependent (PMD) sublayer responsible for forming and transmitting the actual bit stream of the PPDU created by the PLCP. The PMD interfaces directly with the air medium and performs the necessary modulation and demodulation.
The 802.11 MAC Layer defines the frame structure used by all 802.11 devices and is responsible for maintaining order in the shared wireless environment. The 802.11 MAC layer defines two modes for maintaining order. The first is a distributed coordination function (DCF) in which nodes act as peers. The DCF defined for 802.11 uses a CSMA/CA scheme. It is the mandatory mode for all 802.11 devices. The other mode is a point coordination function (PCF) in which one node acts as a master node, or base station. PCF follows a basic polling mechanism, which is optional for 802.11 devices.
The frame structure is shown in the visual. It includes a Frame Control field that has a number of flags. There is a flag to indicate whether the frame is to/from another 802.11 node or to/from a node on the distribution system (DS). Another flag indicates the type of frame: Frames can be control frames used by the CSMA/CA scheme, management frames used by stations and access points as probes or beacons so they can locate each other, or data frames. Other flags indicate whether the frame has been fragmented, whether the frame is a retransmission, whether sequencing is being used, or whether the data is protected via WEP encryption.
Four address fields are defined since frames might be ultimately destined for a non-802.11 node via a distribution system or destined for another 802.11 node via an access point. Addresses are 48 bits long so that 802.11 devices can interoperate with other IEEE 802 devices such as Ethernet devices.
802.11 Security Issues
Perhaps the most interesting and also perplexing parts of wireless LANs are the issues surrounding security in an unbounded medium. Surely at some time or another, we have all see the thriller in which the bad guys' radios are tapped by the good guys and their best laid plans are “foiled again.” If they used some form of encryption and authentication, they could plan their acts without fear of intervention. A similar problem exists in the WLAN environment.
In the beginning, the standards developers searched for a security process that would be equivalent to that found in the wired LAN world. In fact, their solution was called wired equivalent privacy (WEP), which is part of the original IEEE 802.11 standard and can be used with any of the IEEE 802.11 Physical Layer options. It was at best a band-aid on the problem and at worst, a totally ineffective security process.
At the core of the process was the idea of service set identifiers (SSID) that were broadcast by the access points (AP). Only those legitimate users would respond and become part of the wireless LANs. The basic problems with this approach were twofold. First, this is a broadcast process so everybody could monitor the process. Second, all APs came with a default AP SSID. For a Linksys device from Cisco, for example, the default SSID is “linksys.” Anyone who has looked at the SSID list that their wireless LAN adapter has detected has seen these common names. Many network administrators and private AP owners have never changed these default names.
To compound the problem, WEP is turned off by default and has no user authentication. As a result, it is constantly under attack and there are numerous reports of weaknesses with WEP.
802.11 Wireless LANs
802.11 Wireless LAN Products
An 802.11 wireless LAN (WLAN) typically consists of one or more access points (AP) and numerous wireless end-user stations. The AP is the focal point of the wireless basic service area. Multiple APs can be connected via a distribution network, usually a high-speed Ethernet, which can also connect the WLAN to the main wired LAN infrastructure. Individual stations can be PCs equipped with 802.11 compliant network interfaces via PC cards, card bus cards, or PCI cards. Both 802.11b and 802.11g cards are available. IEEE 802.11a is also available but less popular. IEEE 802.11n cards are also on the market, but they are pre-standard. The draft of 802.11n was accepted in January 2006.
A big issue in choosing a WLAN product set is deciding which standard to follow. 802.11b is most commonly being shipped. 802.11a offers higher speeds, but it operates in the 5 GHz band and is not compatible with 802.11b. Organizations that have already invested heavily in 802.11b might be reluctant to move to 802.11a. 802.11g, however, is both available and backwards compatible with 802.11b. 802.11n will be compatible with 802.11a/b/g.
Most AP vendors support 802.11b and 802.11g. Some also support 802.11a by providing devices with dual radio slots so they can operate 2.4 GHz and 5 GHz radios simultaneously. However, the lack of popularity for 802.11a coupled with the work underway towards 802.11n makes the future of these products dubious.
Many vendor products compete in this marketplace—a quick search of the Wi-Fi certified companies supplying internal adapters resulted in more than sixty companies.
Wireless bridges allow building-to-building interconnection, which can be much less expensive than wired solutions, such as leased lines.
The Wi-Fi Alliance was formed to certify compliance to the 802.11 standards. It performs independent testing of vendors’ 802.11 devices to ensure that they meet the 802.11 specifications and that they will interoperate with other certified devices. Devices that pass the certification test are referred to as wireless fidelity (Wi-Fi). There are three categories to be certified in—a, b, or g—depending on which 802.11 standard or standards the device is using. The Wi-Fi Alliance is also involved in defining and testing security procedures. Wireless Protected Access version 2 (WPA) is a Wi-Fi Alliance certification that devices are 802.11i compliant. 802.11n will be added when the standard is ratified.
Enterprise Wireless LAN Applications
Although proponents of wireless networking look forward to a totally wireless world, today wired LANs are still the norm. What, then, are some of the enterprise uses for 802.11 wireless LANs (WLAN)?
- Employees could gather for meetings in corporate conference rooms, bring their laptops, and create an ad hoc network for an impromptu workgroup. If the conference room had an access point nearby, the employees could bring their laptops and reconnect to the corporate intranet to have all of the corporate data available to them during the meeting. There is no need to have the conference room wired with a dozen network ports and no access cables to trip over.
- On the manufacturing floor, wireless machine tools can have their numerical control programs downloaded without the need for cables that might restrict the movement of the machines or even prove to be safety hazard.
- At a trade show, setting up a booth with a network of a dozen nodes is much easier when there is no need to wire up a network infrastructure.
Extending the wired LAN:
- WLANs can extend the reach of the wired LAN in difficult to wire locations (e.g., an old building).
- A wireless access point makes it easy for mobile workers (e.g., sales personnel) to connect to the corporate intranet when they are in the office.
Bridging between buildings:
- Using directional antennas, distances up to 25 miles can be spanned. Companies can find this an easier, more cost-effective solution to installing/leasing buried cable or fiber.
Infrastructure solution for small businesses:
- WLANs offer a quick, cost-effective way to build and grow the corporate network, especially for smaller start-ups. When the time comes for the company to move into larger quarters, it is much simpler to take the network right along.
The bottom line is WLAN products can replace or extend a wired LAN. These products provide many advantages over a wired LAN environment, primarily mobility and lower cost of ownership. The increased use of laptop computers in the corporate environment provide a market driver for mobility, while the cost of wiring or rewiring existing facilities often makes WLANs an affordable option.
Examples of Wireless LAN Deployment
Although many enterprises hesitate to jump on the wireless bandwagon because of distance and speed limitations, not to mention security concerns, wireless LANs (WLAN) have already proven themselves in several environments.
Federal Express is one company that has fully embraced wireless technologies. FedEx first started using WLANs in the late 1990s for packaging, sorting, and aircraft maintenance. The convenience of having mobile devices in these areas was the obvious drawing point, and the result was a significant increase in the efficiency of workers. FedEx initially deployed proprietary 802.11 WLANs and switched over to 802.11b in 2000 when devices implementing that standard emerged. The success the company had using WLANs in the first two areas encouraged it to use them as a way to extend the wired LAN into conference rooms and for their sales force. FedEx currently has about 600 wireless locations with about 10,000 access points throughout the company.
A hospital can reap the benefits of WLAN technology as well. Doctors and nurses can use laptops to enter and retrieve patient information, moving from room to room while still connected to the network via the 802.11 NIC in the laptop. A barcode on the patient’s wristband identifies the patient, and the laptop displays the relevant patient information. Such technology results in much faster access to patient data and less time spent entering new information.
The retail business world also benefits from the 802.11 WLAN. Scanning devices used for inventory control can download their data via the wireless connection. Mobile point of sale terminals with 802.11 connections can be moved to where customer service representatives need them most. Large hotels can use mobile terminals (even at curbside) to speed up check-in during peak hours.
The casual household user should not be forgotten. The rapid proliferation of home computers has given the WLAN a place to live. Few home owners want to rewire their homes for data. A WLAN in the home makes it easy to host multiple computers, and easy to support laptops that can move from the kitchen table to the home office to the den. But if you live in a duplex, apartment, condo, or tight neighborhood, beware of the neighbors! If you do not have the right security measures in place, the neighbors could be hitching a ride on your WLAN and accessing the Internet or even your home computer.
|<mp3>http://podcast.hill-vt.com/podsnacks/2007q2/wi-fi.mp3%7Cdownload</mp3> | Wi-Fi|