With this final section of our five-part series on 802.11n deployments you’ll learn about delivering video over your 802.11n wireless LAN. You’ll understand the 802.11n features that affect video streaming while identifying benefits and limitations related to techniques like multicast and airtime fairness that must be considered in order to deliver high-quality video over wireless without starving other applications. After reading this section, you will understand the common pitfalls to avoid when adding video to your wireless LAN.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
Don't miss any of the articles in this tutorial on deploying 802.11n for improved wireless LAN performance:
- 802.11n for improved WLAN performance
- 802.11n WLAN design best practices
- Supporting legacy devices
- Optimizing 802.11n for voice over WLAN
- Delivering video over your 802.11n WLAN
Businesses often have years of experience with VoIP over Ethernet but far less with video. This sums up the state of the WLAN industry as well. Old APs and clients were not good at delivering high-quality video over wireless, but 802.11n is a game-changer.
According to In-Stat, 802.11n dissolves technical issues that constrained Wi-Fi adoption in video-centric devices. 11n is expected to foster an explosion of consumer electronic video devices, including set-top boxes, game consoles and media servers, topping 200 million units by 2014. Business use will follow as Wi-Fi video products mature.
Video directly benefits from 11n's higher throughput and better range. For those planning to deploy video over wireless LAN, key 11n features include 3x3 or 4x4 MIMO, channel bonding and SGI (to boost throughput), and A-MPDUs and block ACKs (to minimize streaming overhead). Although methods vary by vendor, transmit beamforming can also improve rate over range, boosting effective video throughput. In addition to these standard features, recommended practices for video include the following:
- Antennas and spatial streams: Understand how many antennas your devices have and how they can be used. For example, a 3x3 AP communicating with a 1x1 client is limited to MCS values of the client – that is, 65 Mbps or 150 Mbps. Fortunately, those "extra" AP antennas can still be used for STBC or TxBF to make the best of those supported data rates.
- Rates: Video format and encoding determines throughput needs and therefore minimum acceptable data rate. For example, a 480i60 QuickTime video encoded in MPEG-4 might be fine at 5 Mbps, but 1080p30 HDTV in MPEG-2 could require 20 Mbps. Establish video traffic requirements and verify that your WLAN can consistently meet them.
- WPA2-AES: To encourage stronger security in newer WLANs, 11n prohibits HT data rates (>54 Mbps) on associations secured with WEP or TKIP. To avoid this artificial constraint, verify that all video clients support AES and that no video SSIDs are configured for WPA2-Mixed-Mode (i.e., WPA2 with TKIP or AES).
- Channels: HD video benefits from bonded channels in the 5 GHz band. Even with buffering, video can be latency sensitive, however, so deploy video on non-DFS channels. Some devices do not support bonded channels; others simply don't support them well. If this becomes a problem, load-balance sessions across two 20 MHz channels instead of putting all on a bonded channel.
- Priority: Like voice, WMM access class should be used for video QoS. However, contention rules for video are very different from those for voice, giving video better than best-effort throughput, but without letting video "hog" the channel. Video access class should be mapped to 802.1p or DSCP markings at the AP.
- Airtime fairness: WMM can give video traffic more airtime but does not help video clients share channels with one another. Slower or distant clients take longer to transmit the same bits, dragging fast 11n clients down to their level. To fix this, enable proprietary Airtime Fairness features, preferably based on real-time RF measurement. Products differ, but Airtime Fairness can be important to let 11n video sessions (or any other high-throughput application) reach their full potential.
- Multicast: Some video is distributed using IP multicast streams, but many WLANs do not handle multicast over 802.11 well, dropping data rates to ensure delivery to the weakest clients. In fact, multicast can consume more airtime than sending N unicast frames would have. To fix this, look for wireless multicast optimizations – for example, WLAN products that convert wired-side multicast frames into many discretely addressed unicast frames.
Here again, WLAN performance should be verified against video design goals. Individual metrics like throughput, media loss rate (MLR) and delay factor (DF) contribute to overall video quality, typically measured as Media Delivery Index (MDI).
Bring your 802.11n deployment all together
In this series, we explored benefits and key features associated with 802.11n, mechanisms important for coexistence, and practices to optimize performance of voice and video over 802.11n WLANs.
Bear in mind that best practices are common recommendations for typical WLANs. Just as every building and workforce is different, so every WLAN is unique. Don't rigidly follow best practices – consider the underlying rationale to decide how they apply to your WLAN. For example, a WLAN designed to support dedicated mobile VoIP handsets may look very different from one designed for laptops and smartphones using email, softphone voice and video conferencing over the same Wi-Fi connection.
Ultimately, you will probably need tools to verify that your multimedia WLAN is in fact meeting business needs. Conventional TCP/UDP measurement tools like iPerf are helpful for data but don't tell the whole story for voice and video. Use tools that can measure MOS and MDI and are designed to facilitate testing over wireless, throughout a coverage area. Finally, establish a baseline so that you can concretely assess the impact of proposed upgrades and unplanned changes in the future.
About the author:
Lisa A. Phifer is president of Core Competence Inc. She has been involved in the design, implementation and evaluation of data communications, internetworking, security and network management products for more than 20 years and has advised companies large and small regarding security needs, product assessment and the use of emerging technologies and best practices.