The complex web that is the Wi-Fi standard has made significant progress in the first two weeks of the year, after a 2005 dogged by political feuds and delays. Consensus appears to be close on the important 802.11n fast Wi-Fi specification, which looked in danger of going the way of UltraWideBand and being stalled by factionalism. Over at the IETF (Internet Engineering Taskforce) standards body, which governs the higher network layers, Cisco has achieved its aim of getting its LWAPP technology selected as the platform for controlling access points in enterprise networks. And progress is being made on other extensions to 802.11 that will make it far more appropriate to robust enterprise and carrier applications.
Unlike WiMAX, which has many key capabilities such as quality of service and mobile hand-off incorporated into the base standards, Wi-Fi started life as a simple technology with a restricted remit, confined to local area indoor networks in unlicensed spectrum. It has now become a huge and complex system that operators and vendors are seeking to apply to metro ranges far beyond the Lan, and to functions from multimedia in-home systems to semi-mobile voice and data communications to outdoor broadband wireless access.
In many of these, enhanced rate and range will be vital, especially in home networks. This is why 802.11n has been fought over so bitterly. Just as the two groups shortlisted by the IEEE taskgroup ‘ Wwise and TGn Sync ‘ were close to creating a consensus proposal, a breakaway group led by Intel, Broadcom and others, called Enhanced Wireless Consortium, announced its own submission, which it seemed keen to establish as a de facto standard if it were not adopted by the IEEE.
With the eight of most of the largest Wi-Fi chipmakers, and many equipment vendors, behind it, the EWC was almost certain to succeed in the real market even if it did not in the IEEE. In contrast with the similar stand-off in the UWB world – which saw the Intel-backed WiMedia Alliance amassing industry support for its platform and then taking it off to an alternative standards body, ECMA ‘ the Wi-Fi community seems to have accepted the inevitable and the EWC has gained sufficient votes to win the next ballot.
The Joint Proposal (JP) team, which was working to converge the TGn Sync and WWise proposals, has now accepted the EWC technology and will merge it with its own work. The technical challenge will not be great – the EWC proposal had many points of similarity with work in the JP ‘ but consensus will almost certainly lead to the required 75% majority for the proposal when it is voted on in the IEEE summit this week. The main loser will be Airgo, whose smart antenna technology was set to be at the heart of the IEEE standard but which is not part of EWC. The pioneer, which has early customers such as Linksys for its ‘pre-n’ chipsets, will now presumably have to rework its products for the new standard and will lose its market lead.
It is good news for the development of Wi-Fi ‘ which is lagging that of pre-standard platforms that offer 150Mbps-plus speeds ‘ and for the IEEE itself, whose relevance is put into doubt by the antics of the breakaway groups.
Also furthering the cause of useful real world solutions, this time mainly for the enterprise, is agreement on the IETF standard for how Wi-Fi access points are controlled. Cisco’s LWAPP (lightweight access point protocol), which it acquired along with switchmaker Airespace last year, has been selected. Ironically, the goal of the original proponents of this standard was to limit Cisco’s power by allowing switches to work with multivendor access points, reducing the chance for large suppliers to lock companies into their architectures.
The LWAPP protocol was backed by Airespace but had been overshadowed by newer proposals such as SLAPP (Secure Light AP Protocol) from switchmakers Trapeze and Aruba. This offered a fairly simple approach that would have had the benefit of being quick to bring to market. But once Cisco acquired Airespace, it revived LWAPP in order to regain the initiative in this IETF project, and now it seems the influence of the giant has won through. Other proposals had been submitted by Panasonic and Siemens/Chantry, the latter a prime mover in this area through the CTP tunnelling technology created by the wireless router start-up, which was acquired last year by Siemens. LWAPP will only be part of a broader standard, CAPWAP, which will address control, interoperability and provisioning of APs.
The 802.11 platform itself will gain other functions over the coming 2-3 years. An important one, which will see draft proposals submitted this quarter, is 802.11v, which addresses control of Wi-Fi end points such as laptops and VoIP phones, rather than just access points and other infrastructure. It would define procedures by which a wireless Lan can control key parameters on wireless client adapters, such as identifying which network or access point they should connect to.
Most 802.11v support is likely to be implemented in software, making rapid adoption, including upgrade of older kit, feasible. Among the capabilities envisaged for the new standard are client control, network selection, network optimization, and statistics retrieval and monitoring.
Wireless client control involves several functions, most importantly load balancing, which would automatically point clients to the best available access point in terms of its loading. Currently, clients can be prevented from connecting to overloaded APs but this can disrupt client sessions.
The proposed specification would also reduce the time to connect a client to a WLan securely, enabling configuration ‘ with SSID and security credentials ‘ to be automated from the central point. Currently this can only be done with relatively expensive specialized equipment.
Schemes that will save battery life on low power devices such as handsets are also being considered for 802.11v, specifically by reducing management traffic by simplifying client configurations and controlling Management Information Bases over the air.