802.16: A Look Under the Hood
By enabling quick and relatively inexpensive deployment of broadband services infrastructure, the IEEE 802.16 standards for wireless broadband access have the potential to finally address the long-standing “last mile” problem that has plagued the data and telecom carrier industries. In Part 1 of our in-depth look at 802.16, we discussed how the new technology could be utilized and what was happening in the industry. Now let's delve into the nitty-gritty details of how the standards work and what data networking services they enable.
Wireless Support for Data Networking Services
In addition to the physical layer discussed in Part 1, 802.16 defines a Media Access Control (MAC) layer. The capabilities of this layer allow 802.16 to support a wide array of data networking services, including many services that are already familiar to corporate and residential users using copper or fiber networks.
Because they provide the basis for these services, support for both ATM and packet operations was a requirement in the 802.16 design. ATM is important because of its role in telecom carrier infrastructure. For example, ATM is often used to support DSL services. ATM is also widely used to support voice transmissions. When it comes to packet operation, 802.16 supports all of the “usual suspects,” including IPv4, IPv6, Ethernet, and VLAN services.
802.16 accomplishes all of this by dividing its MAC layer into separate sublayers that handle different services, provide common core functions, and implement wireless privacy. Overall, this design gives 802.16 both flexibility and efficiency at the same time.
The convergence sublayers map the different services into the core MAC common part sublayer. In addition to relating service data units to MAC connections, the convergence sublayers are responsible for decisions about bandwidth allocation and QoS. They also embody functions to get the most efficient use (maximum effective bits transmitted and received) out of the radio frequencies themselves.
The common part sublayer is connection-oriented. All services, even connectionless services such as Ethernet and IP, are mapped into a MAC connection. The common part sublayer includes mechanisms for requesting bandwidth, including bandwidth on demand — a very attractive option for many carriers.
Security and More Security
Authentication and registration are part of the 802.16 MAC common part sublayer. Authentication is based on the use of PKI technology-based X.509 digital certificates. Just as every Ethernet interface comes with its own unique Ethernet MAC address, every 802.16 customer transceiver will include one built-in certificate for itself and another for its manufacturer.
These certificates allow the customer transceiver to uniquely authenticate itself back to the base station. The base station can then check to see if the customer transceiver is authorized to receive service. If the database lookup succeeds, the base station sends the customer transceiver an encrypted authorization key, using the customer transceiver’s public key. This authorization key is used to encrypt and protect any transmissions that follow.
Link privacy is implemented as part of another MAC sublayer, called the Privacy sublayer. It operates below the common part sublayer. It is based on the Privacy Key Management protocol that is part of the DOCSIS BPI+ specification. The changes to the DOCSIS design are aimed at integration with the 802.16 MAC. They also enable 802.16 to take advantage of recent advances in cryptographic techniques.
Features and Goodies
802.16 supports a wide variety of QoS (Quality of Service) options, based on mechanisms used in DOCSIS. Bandwidth can be allocated to a customer transceiver and managed on that basis, or it can be allocated to individual connections between the base station and the customer transceiver. Some customer transceivers will manage their own allocations, even to the extent of stealing bandwidth from one connection to help another. Customer transceivers are permitted to negotiate with the base station for changes in allocations.
These design choices enable services as diverse as connection-oriented, constant-bandwidth ATM and connectionless, bursty IP traffic to co-exist in the same box. 802.16 is flexible enough to permit a single customer transceiver to simultaneously employ one set of 802.16 MAC connections for individual ATM connections and another set for sharing among numerous IP end users.
802.16 uses scheduling services to implement bandwidth allocation and QoS. Unsolicited grant services provide a fixed, regular allocation. This mechanism is well suited for ATM or T1/E1 over ATM. There is relatively low overhead because there is no need to support requests for changes to the allocation. At the same time, delivery delay and jitter are minimized.
For flexibility, 802.16 also specifies a wide variety of mechanisms to request bandwidth allocation changes, including MAC protocol requests and various types of polling. The same mechanisms also can be applied to deliver best effort service, which makes no guarantees for throughput or delay.
In addition to extending 802.16 operations to the 2-11 GHz range, 802.16a also extends the reach of 802.16 beyond the limits of communication between a base station and a customer transceiver. It does this by enhancing the base standard to support mesh deployment. In mesh deployments, a customer transceiver can act as an intermediary between another customer transceiver and the base station. In other words, the customer transceiver is acting as a switch between locations.
802.16 vs. 802.11
It is natural to ask whether 802.16 will replace or compete with 802.11. This question will become even more pertinent once the 802.16 working group completes its work on mobility. Assuming the ratification of a standard for 802.16 mobility and good non-line-of-sight operation inside buildings, which standard should you use, or both?
802.11 is rapidly becoming established. It is cheap and easy to install, and its well-publicized problems with security are being addressed. 802.11 is normally deployed using a hotspot approach. Hotspots are chosen to provide the desired campus coverage. The access points are then attached to the corporate LAN backbone.
In comparison, there are a variety of enterprise network architectures that can be implemented using 802.16. The technology could simply connect campuses to each other or could also work directly with end-user laptops and desktop systems, perhaps replacing all or part of the wired campus backbone.
While there will certainly be some overlap, the two standards have some important differences. 802.11 has wide 20 MHz channels and a MAC that is designed to support tens of users over a relatively small radius of 100-300 meters. (MACs that use more power to attain the 300m limit may be non-standard.) On the other hand, 802.16a allows the operator to control channel bandwidth, and its MAC is designed to support thousands of simultaneous users over a 50 km radius. (This reach has not been demonstrated yet; working products may have a somewhat smaller range.).
The maximum data rate for 802.16 is higher than that of 802.11, partially because it gets nearly twice the number of bits per second from a single Hertz of frequency. In addition, 802.16 offers a variety of QoS choices, while 802.11 supports only best-effort service (with the possible addition of priorities, as in 802.11e).
Because of these options, 802.16a requires more configuration in order to manage the users and the services they receive. The fact that 802.16a supports mesh network topology while 802.11 does not may be more significant to carriers than to enterprise IT managers, given the wide radius of coverage offered by a single 802.16 base station.
Even more important than any of these technical differences are the issues of when the 802.16 standards will be completed and when 802.16 products will become available. Millions of 802.11 NIC cards are being installed today, while 802.16 products will not be available for another twelve to eighteen months. By that time, there will be a very large and significant installed base of 802.11 interfaces in offices and homes. This will provide considerable inertia against any change from 802.11 to 802.16. For the pendulum to swing in the 802.16 direction, there must be significant and compelling benefits for enterprises and individual users to make the switch.
What Does the Future Hold?
As you can see, much care and work have gone into the design of 802.16. Becoming an expert will mean learning many details. However, it is worth understanding at least the rudiments of this technology because it has the potential to revolutionize how companies and carriers design and evolve their networks.
At the end of the day, everyone would like to be able to do more while spending less money, and obviating the need for wires can result in a considerable reduction in infrastructure costs, which means wireless data networking — in its many forms — is clearly here to stay.
Beth Cohen is president of Luth Computer Specialists, Inc., a consulting practice specializing in IT infrastructure for smaller companies. She has been in the trenches supporting company IT infrastructure for over 20 years in a number of different fields, including architecture, construction, engineering, software, telecommunications, and research. She is currently consulting, teaching college IT courses, and writing a book about IT for the small enterprise.
Debbie Deutsch is a principal of Beech Tree Associates, a data networking and information assurance consultancy. She is a data networking industry veteran with 25 years experience as a technologist, product manager, and consultant, including contributing to the development of the X.500 series of standards and managing certificate-signing and certificate management system products. Her expertise spans wired and wireless technologies for Enterprise, Carrier, and DoD markets.