I need to establish links to 2-3 moving targets equipped with Metal 5 radios @ 3-5 Mbps speeds. (no more than 3 clients)
I understand advantages of MIMO with H/V antennas but in my case, all clients are V-only Metal 5 radios.
My priority is link stability over speed.
Should I use two V sector antennas with dual-chain NetMetal or dual-polarized H/V sector antenna?
What advantage from using H/V if all clients are V-only?
Does NetMetal support diversity antennas? (spatial)
Similar question about mobile “command and control” van:
Should I use triple-chain NetMetal with three V-omni antennas or dual-chain with H/V omni?
Please, advise.
Thank you.
Thank you, jarda.
Omnitik is almost half power (26 dbm) vs. Metal (31 dbm) with so much bigger plastic case.
Due to design specifics, I would have to use Metal 5 radios with V-pol antennas.
So, what’s the deal with NetMetals? Can I connect multiple V-pol antennas for diversity?
For moving and tilting motorcycles will be dual polarised antennas maybe better than single polarisation antenna on metal. There are also legislative rules that will not allow you to transmit at full power, so if you will be in line with them, even the power of omnitik will not be fully used, most probably.
None will be able to foresee what will be working the best. Maybe using two vertically polarised antennas just slightly tilted left and right can help you a bit.
Yes, you can use how many antennas you like, with the same polarization . It will work very well .
You can use power divider 2: 1 to mount two antennas on the same connector .
Total 6 antennas !!
Few years ago, somebody mentioned that RouterOS does not support spatial diversity antennas, only H/V.
I guess, this situation changed? That’s good news!
To combine antennas with the same polarization has nothing to do with the spatial diversity . No more than it can be said that the system will become SISO . The advantage would be that it will be a single polarization , less interference , an area covered more widely .
The spatial diversity has nothing to do with antennas polarization .
Multiple, physically separated antennas with the same characteristics stands for spacial diversity technique.
Antennas are typically positioned at specific distances related to frequency (wavelength).
Spacial diversity has nothing to do with polarization.
The question is, will NetMetal be able to “combine” signals from two (or three) spatially separated antennas?
H/V MIMO is different, it utilizes two separate streams from H and V polarized antennas.
(Some people pointed that Mikrotik radios are designed for H/V MIMO only..)
For the receive part, I’m certain Mikrotik devices do feature spatial diversity gains for every receive antenna chain you activate, by making use of maximal ratio combining (MRC). You can check this happening quite easily in Winbox, Wireless → Registration → {radio name} → Signal. Here, receive signal strengths (and send strengths as well, if the client device is Mikrotik as well) are displayed per antenna chain as well as combined overall. The overall value is equal to the sum of the single strengths, e.g. two -75dBm signals will combine to one -72dB signal.
On the transmit side, I’m not completely sure how it has been implemented by Mikrotik. When I try it on an Nv2 link, It sure looks as if it’s the same situation as with receive. That is, if I limit the spatial streams to one (by disallowing HT MCS above 7), it still shows me different transmit signal strengths. My assumption here would be the driver generally chooses the one which works best, and checks back both periodically. It could also be that the same stream is just always transmitted as often as there are antennas (hopefully not at exactly the same time), which wouldn’t be ideal but probably fine.
Edit: I’ve just now re-read your original post and learned MIMO isn’t really your target at all anyway, therefore I’ve removed the MIMO discussion. You just wanted to know whether a MIMO antenna (H/V) would benefit you. My answer would be “probably not so much”, better go with real spatial diversity, i.e. two similar antennas placed apart from each other. Your option with multiple vertical sectors should be fine.
My general advice for maximum robustness in something like your scenario would be:
Use diversity on the mobile terminals as well. It will increase receive strength by several dB “for free”, by diversity and MRC. Omnitiks are a good place to start. Use single chain only if size or weight are of importance (which might indeed be the case here).
Disable 11n altogether, limit the network to 802.11g rates (probably independent of protocol, Nv2 should be fine). I don’t see AMPDU aggregation to be a very good idea for moving terminals. Maybe even consider fragmenting large packets (>>500B).
For even more robustness, restrict to only use BPSK and QPSK rates (up to 18Mbit/s). This way you avoid the “A” part (amplitude) of higher order QAM, which is way more fragile than the “Q” part (quadrature, i.e. phase shift) of QPSK (and QAM).
This would give you maximum range and lowest bit error rate, however at somewhat modest maximum/peak rates. I’d guess something like 12-13Mb/s at peak with Nv2. You would have to ask yourself, how much do you need?
Of course you are aware that since 802.11 in all variants uses ODFM, it copes very badly with Doppler shift inherent to moving clients, so don’t put your hopes to high…
Yeah well, all mobile communications with moving terminals are severely affected (give or take, depends only on speed really) by Doppler shift, in particular Doppler spread. At least OFDM subchannels are flat-fading, so something will get through often enough. Even if the “orthogonal” part won’t work at all times. That’s also why fragmentation and disabling aggregation can help a lot, in my opinion.
The alternative, 11b with DSSS, would probably fare even worse because of coherence bandwidth. Think UMTS (CDMA uses DSSS) vs. LTE (OFDM). My suggestion would be to configure the WiFi as if it was LTE.
