Transporting High-Def Video Broadcasts: Peer-To-Peer Cognizance, Spectrum And Speed Analysis, And Other References
This blog post references my feature article ‘Transporting High-Def Video Broadcasts: Are Wireless Networks Up To The Task?‘ in EDN’s August 20, 2009 edition. It’s one of a series of web addendums to the print writeup.
Those of you who read my hands-on missive of two months back hopefully discerned that my difficulty in streaming high-def MPEG-2 video over 802.11n was in no small part due to the fact that I was in actuality attempting to simultaneously transport two streams through the same Wi-Fi link; one going from my Windows Media Center-based laptop to the router, and the other sourced at that same router and ending up at the Xbox 360. This cumbersome man-in-the-middle approach is a necessary outcome of the unfortunate reality that at the moment, Wi-Fi-based LAN clients are largely unable to directly communicate with each other. Granted, there is the ‘ad hoc’ mode, which is supported by (for example) the Nintendo DS and Sony PlayStation Portable handhelds. But in my experience its connections are difficult to set up, unreliable to maintain, and not pervasively supported.
Note, however, the carefully chosen words ‘at the moment’ in the above paragraph. During the research phase of the article, I was tipped off by several sources to the pending standardization (by the Wi-Fi Alliance, mind you, not the IEEE) of so-called ‘Wi-Fi Direct’, a peer-to-peer connectivity extension to 802.11 that was unveiled yesterday (and which, unlike some, I don’t see as a competitor to Bluetooth by virtue of the fundamental range-vs-power consumption tradeoff of the two approaches). Some of you might be scratching your heads at the moment; isn’t IEEE 802.11s old news at this point, and wasn’t it in fact implemented by the several-year-old OLPC? You’re right, but remember that the IEEE and Wi-Fi Alliance are partner organizations that exist to serve different functions.
The IEEE’s primary focus is on developing new standards; the Wi-Fi Alliance promotes them and ensures interoperability via ‘plugfests’ and other methods. With IEEE 802.11n, in fact, the Wi-Fi Alliance began its work more than two years prior to spec finalization. And to that point, I was happy to see at the beginning of the month that 802.11n certification has finally gotten the green light by the IEEE (formal ratification will actually occur next month). Equally important, the IEEE’s goal that ‘draft’ silicon be fully compatible with the final specification has also been met, judging from the blizzard of press releases I received from silicon and system suppliers afterward. Most products will be compatible as-is with the final spec; others will require only a firmware update.
Recall that due to the performance limitations of single-channel 802.11n, coupled with weird glitches exhibited by my router (which I still haven’t yet gotten around to replacing) when attempting to bridge between two Wi-Fi access points either internal or external to it, I’m running a hybrid 802.11n-plus-HomePlug AV topology between the laptop and game console. Last week, I noticed I was experiencing glitchy playback of all content coming from the laptop, not just ~20 Mbps ATSC material. From past experience, and particularly considering that I’d just experienced premises power loss, I suspected my powerline adapters were to blame.
Yet unplugging all of them and plugging them back in did not solve the problem. Nor did rebooting the laptop. I ran a Network Performance Monitor test within Windows Media Center, which reported no bandwidth issues. Finally, on a hunch, I power-cycled my NETGEAR WNHDE111 802.11n access point, and everything was smooth sailing from that point forward. I don’t know what went wrong with the AP, and no newer firmware release for it is currently available. Nor do I know why Microsoft’s testing utility was unable to discern the problem. But perhaps a more generic bandwidth measurement program would have worked better. I’ve written about Iperf, which I used in-depth back in the summer of 2007, many times before. And perhaps the following short list of links I’ve been collecting over the past few days will give you some other ideas:
How about another list, this one more extensive? As mentioned in the writeup, I did as-extensive-as-possible analysis of my testing environment to ensure that there were no other notable contending spectrum interference sources, Wi-Fi or otherwise, in the 2.4 GHz and 5 GHz ISM bands. For more on the topic, hit up the following links:
- HeatMapper Helps You Create a Detailed Signal Strength Map
- The Definitive Guide to Finding Free Wi-Fi
- AirPort Menu Improves in Snow Leopard
- What’s your wireless signal strength?
Conversely, if you want to restrict the effective range of your wireless network, check out:
Regarding the single-versus-dual channel (and single-versus-dual band) experiments I ran in my study, the following writeups may be of interest to you:
- Understanding Wi-Fi’s two spectrum bands
- Ask Engadget: What’s the best dual-band router?
- Notes on a Dual-Mode AirPort Extreme Network
More generally, to optimize your router’s priority processing of packets you care about from bandwidth and latency standpoints, see:
And finally, I’m glad to read that Microsoft’s finally coming out with an 802.11n-cognizant wireless adapter for the Xbox 360. Its number-of-streams support is still unknown, however, and its $100 price tag is mind-blowing (not in a good way, either).