View Full Version : Will MIMO wireless G routers have an advantage with the SB ?

2007-08-03, 09:39
I have read several wireless router reviews and I can't tell if MIMO needs to be "supported" by the wireless card/device, in this case the squeezebox, for it to be of an advantage.

Smallnetbuilder found that a Belkin MIMO wireless G had the best combination of throughput and connectivity.

Mark Lanctot
2007-08-03, 10:00
How I believed MIMO works is that it recognizes "where" the clients are and optimizes which antennas it uses from its multiple antenna array in order to get maximum throughput around obstacles.

In that case, the smarts are all in the router and the receiver doesn't need to know about this, so yes, a MIMO 802.11g router would have advantages working with a Squeezebox that has a regular 802.11g card in it.

But my impression doesn't match the description in Wikipedia, which states that both devices must have a multiple antenna array:


so if the Wikipedia article is right, the receiver needs to be MIMO as well. However, there sure aren't many MIMO receivers out there, so I'm not sure if there will be improvements with a non-MIMO receiver, i.e. whether the router can get any improvement without a MIMO receiver. A MIMO router with a non-MIMO receiver would be the typical use case for sure - I would think since MIMO is out of infancy now there must surely be some optimizations?

2007-08-03, 10:02
All I can say is that what I have heard from others does largely match my own experience. Yes, you do need the same technology to be implemented on both devices to reap the full benefits.

I recently 'upgraded' from a Speedtouch 802.11G wireless router to a Netgear draft 'N' MIMO router & matching wireless adapter. The aim was to improve the signal that my son's PC (which is situated a good distance from the router) receives. Out of interest I initially just substituted in the new router and although everything seemed to work fine it was not noticeably better than before. But, after adding in the matching draft 'N' USB wireless adapter, the signal increased from dramatically, from approx 50% to a very healthy 90 to 100%.

Although I hoped my SB3 connectivity would benefit too, I can't really say it has made a big difference. Oddly, the signal strength when the SB3 is idle actually reports as being slightly lower than it was with my previous router, but the level seems to increase as soon as streaming begins. Part of how MIMO works maybe?

However, (and I'd need to perform a few more tests before I could say this with 100% confidence), my SB3 does seem to fair somewhat better in the Slimserver network tests. previously it would not reach 100% on the 3000kbps test, whereas it does now.

2007-08-03, 10:53
I got a performance improvement on my SB3 when I upgraded to a Linksys SRX400 MIMO G router. I had lots of dropouts before and now have almost none (except for Rhapsody, which has its own issues). The router also has a Quality Of Service function where I can set the SB3 to be the traffic priority so that when the network is heavily used the music stream is not interrupted.

2007-08-04, 15:15

The MIMO tag is really misleading. What the 'MIMO' routers with ordinary clients does for you is realizing a MISO (Multiple Input, Single Output) channel and provide antenna diversity. One can view this as multiple antennas (2 or 3) at the router will send multiple streams to the client. The client with one antenna will recieve them all and since we have created redundancy since the data is in the air in several copies we have better shot at decoding it. Then how data is mapped onto the antenna streams can be more or less clever. To send the data in three copies is one shot, but better performance is achieved with some coding where the data is cleverly spread over all antennas, but not necessarily in identical copies.

Even better can be achieved if you can recieve more of the wireless energy that's in the air, i.e. more recieving antennas, and now we get something like a MIMO channel multiple antennas transmitting and recieving. If we're lucky we could view this as multiple independent paths over pairwise antennas. Reality plays us some tricks, and we essentially never are that lucky.

We could however help luck a bit by informing the transmitter about how the reciever see things on its side and adapt the transmission accordingly. Now we're getting where draft-N is aiming.

However, the most certain benefit from multiple antennas without too much clever processing is mutliple antennas on the receiver side (compare how much better a laptop with an internal WLAN and two recieving antennas in the lid performs compared to a PCMCIA card with only one antenna). I do not know too much about the slim antenna configuration - I guess it has some sort of patch antenna, but more antennas on the router side would then at least improve the uplink to the router - something that the slim has only slight benefit from.

End of class ...