I had a chance to look at a wireless HDMI transmitter/receiver system the other day. My colleague had sourced it to simplify one school’s gym A/V system. The specifications all sounded quite good. The device in question is a Kramer Electronics KW-11 Wireless High Definition Transmitter/Receiver. You can find the product page here and the spec sheet (direct to pdf) here.
I was told by my partner in crime that:
- It is a 5 GHz device,
- It uses 2 wi-fi channels (so, a 40 MHz channel),
- It automatically selects the best available channel, and
- It has a range of about 40′.
I was curious to see what this would look like on a spectrum analyzer, whether or not it would select the best channel, and what kind of special prep would be needed to minimize impact if they were going to use it. My test was simple:
- Start Chanalyzer.
- Turn on the KW-11.
- Watch the colours!
- Change a nearby access point to operate on whichever channel the KW-11 came up on.
- See if the KW-11 changed channels.
- Check client performance on that local AP.
This is what the spectrum looked like after the KW-11 had been running for a few minutes.
You can see that it was jumping back and forth between channels 149+153 and 157+161. It did seem to spend most of its time on 157+161, but it didn’t sit still for more than a minute. The gym AP was running on channel 36, so it *seemed* to pick the farthest channels from that. So what happened when I moved the AP from 36 to 157 and cranked up the power?
In a word, nothing. Even after restarting the KW-11 while the AP was on channel 157, it came back up in that band. I probably don’t have to tell you that the AP may as well not have even been there, for all the good it was doing at that point. My running ping immediately saw massive packet loss and my remote desktop connection was dropped. I set my NIC to prefer 2.4 GHz and everything came back up.
I can confidently say that the KW-11 will only affect the immediate area, ie. the room that it is in. At the time of the image above, my laptop was about 20′ from the transmitter. Taking my spectrum analyzer behind a brick wall to the room where the receiver was, I saw the signal drop down almost into the noise floor. Surprisingly, the video feed was still up. Not really usable, but up.
So, this is a pretty solid little device as far as quality and robustness goes. It includes audio in the stream; we used a splitter to pull the audio out to a PA system, but before we put that in, the sound was coming out of the projector speaker. As for special network preparations, none will be necessary in this case. The gym walls keep it isolated nicely.
If anyone would like a copy of the Chanalyzer files (I have three), please let me know in the comments or on Twitter (@alfmckim). Thanks for reading!