CC8520

Hi Yohan,

we have done extensive co-existence testing on the PurePath Wireless platform. This includes coexistence with multiple PPW networks, wlan (a,b,g and n) plus a number of proprietary 2.4GHz interferers. All in all we are very satisfied with the performance, and you can expect our solution to be one of the better ones currently on the market.

The challenge in answering your question is obviously the notion "close proximity". If you think about several systems in different rooms of an apartment, the number can be huge since walls and the environment attenuate the RF signal. Even if two systems use the same RF channel(s), the audio link can still work just fine if either the RF transmissions occur on different points in time or if the wanted RF signal is somewhat stronger than the interfering signal.

To give you a more "hands on" answer we just taped a video showing 4 PurePath Wireless networks running in a very limited space: 4 PPW master devices configured for audio transmission on a table within 50cm of each other.  On the other side of the link, 4 PPW receiving devices within a meter or so (some of our summer interns helping out with real world testing, standing as close as practically possible wearing PPW enabled headphones) - hence 4 independent 44.1kHz/16bit uncompressed stereo audio streams. In addition we have wireless LAN in the room and one audio source was controlled by our RF4CE remote control kit.

Why does it work?

We didn't actually put up any test equipment to see what was happening on the air, but the intelligent RF protocols in play here most likely contribute a great deal. I have no doubt that cases of saturation and/or retransmissions happened during our ad-hoc test, but the audio and the wlan didn't break. Both PPW and wlan uses a "listen-before-talk" scheme to be more "polite" in a crowded 2.4GHz band, so they will both back off for a short period of time if RF transmission with a certain signal strength is ongoing. In addition, our PurePath Wireless solution uses diversity both in frequency and time ("frequency hopping with agility"), and is continuously monitoring the whole 2.4GHz band for activity during audio transmission. PPW will stay on a set of channels, and replace these one-by-one if they are no longer suited for transmission (several explanations are possible here: RF signal too weak due to multipath environment, interfering signal occupying the channel). With 5Mbps raw data rate of the CC8520 there is also room for "catch-up" (re-transmissions) when data packets are lost. (For more details on the PPW RF protocol, see PurePath Wireless User Guide p. 38 and onwards: http://focus.ti.com/lit/ug/swru250/swru250.pdf )

I'll post a link to this video as soon as it goes on the web, it will likely take a week or two to get the post-processing done. Looking forward to your comments!

Erling

Hi Erling,
two follow-up questions:
1. Just to make sure - if these 4 PPW master devices were within 5-10cm instead of 50cm would it change anything or it should also work the same ?
2. What is the range that can be safely assumed in e.g. multiflat buildings where there are densely located walls - e.g 10cm thick wall every 3m - how many walls can the signal pass and still work ? When I tested the CC85xxDK with the 2510 range extender it was losing signal after 3 brick walls and approx. 10 meters - is your data more less the same ? How much would have the antenna to be bigger if I wanted to increase the range by say 50% ?

Thanks

Pawel

Hi Pawel,

any RF system will be more vulnerable when interfering RF signal strength increases, so moving units closer when signals are strong is not a good idea. Having several units "on top of each other" will most likely result in the radio going into saturation more often - and the intelligence in the protocol will then have to cope with that (not being able to receive at some points in time because an interferer is saturating the radio). Real-world coexistence results in such environments will also prove difficult to repeat, since antenna placement/orientation greatly affects radiation (take a look at the radiation pattern from this antenna, for example: http://focus.ti.com/lit/an/swra117d/swra117d.pdf ).

For your second question - take a look at the range measurements from our office premises in app.note089: http://focus.ti.com/lit/an/swra320/swra320.pdf

Here is a snapshot of a 30 second audio streaming test streaming uncompressed PCM16 stereo audio with 23ms latency (1024 samples)  - green means "no audio dropouts".