Im not asking 'is diversity supported in routerOS' im asking does diversity work when the cards supports it?

I couldnt find a recent radio diversity post, so I was thinking routerOS might of added a little something since.

Ive ran some tests and I cant conclude whether diversity is working or not.

Test Hardware/software:
• RB133 (2.9.46) with senao 8602+ (@18dB)
• routerOS set to 'antenna a' mode
• 18dB directional connected to MAIN on mPCI radio
• 8dB omni connected to aux
• netstumbler software.
• compaq laptop with broadcom chip, never moved! (located other side of 2 office walls.)



I started netstumbler on my laptop monitored the signal for 2 minutes. Without turning the RB133 off i unplugged the 18dBi antenna. the signal dropped a bit on the laptop but still decent. (so the RB133 switched to antenna b without me changing anything).
I then unplugged the 2nd antenna and lost signal all together on the laptop.
Reconnected the 8dB omni to MAIN port on mPCI radio and got signal back on the laptop, took it off and swapped it over to AUX port on mPCI radio. The signal dropped when it was off and come back when I connected to AUX.

so setting routerOS to 'antenna a' doesn't mean the second port on the radio is disabled.

I looked around the web a bit, and it looks like it doesn't matter what settings the operating system is set too, if the radio is receiving a better signal on its secondary port it will switch to this port. (if supported in the radio (diversity))

my live setup seems to be working, I can go a long way behind my directional antenna and as far as I can figure I am connecting to the omni. There is nothing for the directional to bounce off to come backwards. (18dBi comtenna with 18dB running on the mPCI radio)

please dont just post a 'no' or 'diversity isnt supported' Ive already read your posts in other topics. I havnt read anyone doing tests or shown the evidence that it is impossible that it will work.

Does it work or not?

thanks guys.

Nope, it doesn't work because it isn't supported.

I'm not sure how to explain your results, because if I change from antenna a to b on a tower, I am unable to scan or pick up APs that I could before, and clients do not connect...

I suspect that you are just close enough that the antenna is picking up an inductive signal and is working good enough to go through a few pieces of drywall..

As I understand it, the hardware diversity switch drops signal strength by about 20-30dB from the active port. You will still receive strong signal from the "wrong" port.

This is validated by signal strengths from a blown diversity switch.

I really wish MT would support diversity. StarOS does for example...

George

Diversity does work but you must use identical antenna specs for both the aux and main. Since only one ant. can tx/rcv at a time, there is no signal drop using the additional antenna.

Can someone please tell me what the advantage is for diversity... the only think I can see is if you want to use the same radio to build a repeater... is this correct?

Diversity antenna systems are used to deal with multipath issues such as fading and reflections. The wider the horizontal spacing the better, to a point. The dBm calculations for each antenna must be exactly the same or it defeats the purpose of diversified antennas.

Another use for diversity would be to allow long range (15+dBi) and up close (2-8dBi) traffic. This would really only work with a single client.

The lower gain antenna would have a wider beam, it would hit local clients better. I've done this in practice with a WRT54GS with an indoor antenna and an outdoor antenna. Works fine as long as I'm the only one using it. If there is a client in each area, the ping is erratic. Something like a 30ms loss, likely due to how it controls the diversity switch, or possibly just the time it takes for the change to be made.

In my experience, diversity would not work well for a WISP. Even if both antennas are matched, outdoors with ideal spacing, one customer would get better signal at the expense of another, vice versa, and the time it takes for diversity to swap would really drop performance.

One application of diversity is for long Point-to-point links. Over extremely long distances, a signal can be diminished with k-factor (atmospheric refraction) and snow/rain. The reliability of the length can be improved be vertically spacing two identical antennas on both ends. Sometimes, you will see large microwave towers with this type of configuration.

Eric

i am new to Mikrotik, i have tried to connect the Aux and interchanged with main, while also trying Ant-a and switching to ant-b. (4 ways trial) my results were the same.

I couldnt notice any difference with the modes or antena position. I'd like 2 know which is normal. ant-a with main connection or ant-b with aux.

I am using a Mikrotik RB433ah and Engenius Artheros 600mw.

diversity is not supported in RouterOS, and it doesn't matter which antenna port you use.

diversity is not supported in RouterOS, and it doesn't matter which antenna port you use.

my problem is that the network does not go beyond 1km (thats if it gets to 1km) i have changed the miniPCI, jumper cable, N bulkhead connector, Antenna from Hyperlink15dbi to Dlink 12dbi brand new. I still get the same problem. signal doesnt go far.

Infact i am already thinking of using a SENAO ECB 3220 Outdoor which has worked for me before now. but I beleive there could be something I can do to improve the coverage. I'l appreciate any suggestion.

Solution:
Mikrotik RB433ah
40ft Mast, 18v POE, Engenius 600mw miniPCI
configured as AP Bridge

Pls help with any suggestions.

what is the "wireless client" device that doesn't reach it?

Thank you for replying so fast. Dlink 524 internal PCI for desktop. with 24dbi Grid Antenna.

no matter what kind of AP you will have, this client will not reach more than 200-300m because of it's own wireless card and antenna. you will have to make more APs to reach all those clients that are located far away

to be honest, I've had very bad experiances with most of d-link wireless gear.

I only have a very small wisp now. Most of my wireless projects are P2P and hotels.

I have a feeling that at the hotels I might get better coverage with diversity, but ... I usually have great coverage with a single omniV or 180h and a R52H.

I have found that if I increase my standoff to more than 2ft, I get much better coverage, but I wonder how diversity would help, any experience from someone using a similar setup with another vendor that supports diversity?

And at end: RouterOS doen't support Space Diversity or Polarization Diversity?

Both.

using the new 802.11n wireless cards you can have some kind of diversity in B/G mode if you specify to use both chains.

We have had very good results using diversity with madwifi drivers on OpenWRT. Generally we will use one antenna with vpol and one with hpol. Driver will sample SNR off both antennas over a time period and then switch to use the best antenna based on sampling. Most effective in high motion/dynamic environments. Can also help with space diversity between two vpol antennas where multipath is changing (over water, train yards, etc).

It would be great if MT supported this, but you have to pick your battles.