Will 802.11e solve problems with excessive packet loss and jitter with VoWLAN?

Will 802.11e solve problems with excessive packet loss and jitter with VoWLAN?

Will 802.11e solve problems with excessive packet loss and jitter with VoWLAN?

    Requires Free Membership to View

    Login

    SearchMobileComputing.com members gain immediate and unlimited access to expert guides for mobile deployment, management and security, industry trends, and more-- all at no cost. Join me on SearchMobileComputing.com today!

    Kate Gerwig, Editorial Director

    By submitting your registration information to SearchMobileComputing.com you agree to receive email communications from TechTarget and TechTarget partners. We encourage you to read our Privacy Policy which contains important disclosures about how we collect and use your registration and other information. If you reside outside of the United States, by submitting this registration information you consent to having your personal data transferred to and processed in the United States. Your use of SearchMobileComputing.com is governed by our Terms of Use. You may contact us at webmaster@TechTarget.com.

802.11e is being developed to address over-the-air (layer 2) QoS in wireless networks. As it is currently being specified, 802.11e will be useful for home and small business networks, but insufficient for enterprise scale applications. WMM Quality of Service (or IEEE 802.11e EDCF) provides prioritization based on traffic classification from the wireless LAN access point to the client as well as from the client to the access point. This provides prioritized channel access on the upstream direction however does not scale up with large numbers of WMM voice clients. This is due to the fact that in a high voice density system with lower backoffs used for the high priority the chance of a collision is actually statistically worse than with standard non-WMM DCF. Therefore, with high numbers of active WMM voice flows, this will create contention that can impact voice quality. Additionally on the downstream direction, since the access point would fall into the same WMM priority queue with lower backoffs for the voice flows it will be contending equally with all of the voice clients that are sending upstream traffic. Contention is the enemy of predictable communication in wireless Ethernet.

In addition, it is important to note that this standard does not address some other areas that may be important to enterprises. 802.11e does not provide Quality of Service on a per application basis. This means that if a laptop is simultaneously running a soft phone for a wireless VoIP call as well as checking email, the device receives the high priority assigned to it, not just the voice application. This situation worsens contention.

Also, 802.11e does not solve the problem of collisions being caused by neighboring cells, or co-channel interference. A call could have priority in the cell that it is communicating in but with only three channels available in 802.11b/g (which is what phones run on) there will inevitably be overlapping traffic from other cells which will further increase contention for that client. So the contention problem is not how many clients attached to that single AP but how many clients within "hearing range" of each other on the same channel.

Lastly is the inter-AP handoff problem. Today most WLAN systems (though not all) do not manage client handoffs during roaming. The roaming handshake is initiated and controlled by the client. The process of handing off as you roam can take anywhere from 200 - 300ms. This can destroy a VoIP call that can't tolerate more than 20 to 30ms handoffs.

So in a small environment where inter-AP roaming and co-channel interference aren't an issue 802.11e will help a great deal. In larger environments if you're relying on 802.11e as your sole QoS mechanism you'll be pretty disappointed as more users get on the Wi-Fi network.

This was first published in July 2005