2x2 MiMo 802.11 N on 900MHz

Unfortunately, we still have a lot of customers in deeply wooded areas that have been stranded on 900MHz. Although we have made a push for more 5ghz on our network, and are making excellent ground in our plans to deploy fiber, we still have a large chunk of customers that are stranded in this spectrum, and suffering greatly from the limited amount of bandwidth we are able to provide. Unfortunately, Metal 9s are just so incremental, we can’t justify the expense for such little gain. Cambium on the other hand is also far to expensive to be justifiable. Overall, we are happy with Mikrotik’s platform for 900, just need an 802.11n 900MHz 2x2 MiMo interface. Has anyone found any such solution?

These are Atheros based cards, should work under ROS.
https://vizmonet.com/nm2-series

I have absolutely no experience with this manufacturer … so you should really test them first.

Regards,
M.

Nice find. Been looking for something like this. Curious if anyone has found anything else, or has any experience with these cards. I wonder if anyone from mikrotik could comment on this. (obviously it would be unsupported by your team, just simply if it’s feasible).

https://www.bhphotovideo.com/c/product/1077708-REG/ubiquiti_networks_locom900_loco9_mimo_cpe_airmax.html
https://guide.alibaba.com/shop/ubiquiti-nanostation-loco-m900-900mhz-2x2-mimo-airmax-locom9_30675002.html

I have always had terrible luck with ubiquiti’s M900 equipment. I have seen Metal 9s outperform these.

The problem with 900 MHz is that there is just not enough frequency channel width available.

There is enough channel space to make a B or G or even an N connection - but there is not enough space for a N Ce connection.

I am very much aware of how dirty 900mhz is. We focus on 5ghz, 2.4, and soon (beginning of September) Fiber. However, we still have about 100 customers on 900mhz that we are unable to reach using these methods, that it could take many many years before we will be able to reach through fiber (only real option given how heavily wooded the area is). We don’t expect to be able to run super high bandwidth, but even at 10MHz wide channels with 2 chains and fixed framing (I have an XR9 setup running 80/20 @ 5ms running flawlessly), we could provide a very usable connection that beats the alternative. For many in our 900mhz network the alternative is nothing. The trees block signals for satellite, and a lot of the folks out here struggle to pull cell phone signals even with repeaters with mounted external antennas on their roofs. It’s also a good chunk of cash flow that we are in danger of losing, as the speeds we are able to provide with the existing equipment are approaching unusable for some, but have been unusable for many for many years.

You always have the chainsaw option. I’m only half kidding. If having Internet (or faster Internet) is of enough concern it may be cheaper for you to get higher up in the air at points and chop trees down as needed to get line of sight. If you are planning to bring fiber to them they may just have to wait.

An alternative may be to offer to share the cost of a fiber dig to them. A user of yours may just be rich enough, bored enough or want higher speed Internet bad enough to help.

You could look at pay to play spectrum to jack up the power in other bands. That said, even in a remote area I wouldn’t be surprised if you couldn’t find a decently low MHz license for anything cheaper than your own fiber.

We are using Rohn 25g towers and already have them at 140-145’. Unfortunately, it’s not just a few trees. In a lot of cases the trees extend on to other properties and will cover over a quarter mile. Also, we are located in the poorest county in the state, not a lot of folks are capable of re-roofing their homes, let alone pay large amounts for internet service. Fiber will make it to a lot of them, but the question is when. I expect it could take 5+ years before we could reach all customers we intend on reaching, and some of them will likely end up left in the cold anyways. Just need that one last “hurrah” for 900.

Re: “… You always have the chainsaw option. …”
This is officially called the “Gasoline powered signal enhancer” and they often work very well :slight_smile:

My boss keeps talking about tethered antenna balloons and 100 percent solar powered continuous antenna plane/drones. Yea right - until the first mild breeze or lightning or heavy rain or anything.



North Idaho Tom Jones

…and what about night? How will the solar panels work?

Solar panels at night - no problem. You just use massive bright floodlights aimed at the solar panels.

I am not against solar panels. I am against there use when the alternative power design is not done correctly and does not account for things like snow on the panels and long winter nights, and fog & rain.

Up here in North Idaho, I would consider solar panels if I had ample battery run-time and automatic wiper blades on the solar panels - and solar panel defrosters - and did not have to worry about batteries operating in -30 degree temperatures - and constant battery charge/discharge cycles for the life of the batteries.

North Idaho Tom Jones

Anyone know if these cards are compatible with the XR9 and XC900M?

I sent an email to support but I haven’t heard back yet.

Being able to slowly upgrade and swap out the radios would be great. If not, upgrading is going to be a lot more interesting.

I got a reply Friday afternoon. They are compatible with the XR9 and XC900M.

Now to find a supplier.

I don’t think they’re FCC certified yet, but it is in process.

The 900MHz card uses the AR9223 chipset, the same as the Mikrotik R2SHPn among others.

I have also received replies from Vizmonet. They claim that the cards are compatible with routerOS. I would expect compatibility to be similar to Metal 9s with XR9s. It works, but their are some combination quirks that don’t always link up. It’s been awhile since I have tested, but I believe I have had issues when the AP is XR9 and the client is Metal 9. Unfortunately the cards from Vizmonet are too expensive at this time for us to consider. We need a lower cost option, or larger performance gains (I don’t anticipate enough of a gain between Metal 9 to these cards) to justify the investment. If we could nab these cards at $100ea, we would consider it. Also, cold weather may be an issue, and we are located in Wisconsin. Has anyone ever used a cheap solution to hardware that has this type of restriction?

Where are you getting them?

How much?

RBMetal 9 technically isn’t compatible with the XR9 and XC900M, at least according to Mikrotik. It wasn’t designed to be. They will link up using nv2, but I can’t use nv2 on my 900MHz. Throughput goes from ~10Mbps to 1Mbps.

Victor from Vizmonet said these were designed to be compatible with the XR9. We’ll see I suppose.

Since price is determined by MOQ through vizmonet, would any other wisps be interested in combining efforts to see what kind of pricing we could get it down to if we worked together? We would probably purchase 100 units if we could get the pricing low enough.

Any updates on this?

I still can’t find a supplier stateside.

I am revisiting this card as an option.

Trying to see if we can get enough interest to make it worth the time and money to get it FCC certified.

Anyone have any experience with it yet? Alternative branded options? They seem to be just about the only ones making these cards any more. Xagyl is effectively gone.