Cambium needs LTE

Just got off road testing the Telrad 4G stuff with a friend in a nearby WISP.

Gota say it knocked my socks off how well it worked. Particuarly with its NLOS and at 3.65ghz.

When comparing price Telrad vs Cambium Networks, the cost vs benifit was huge and Telrad won.

Giving you the heads up Cambium Networks management, to stay sucsesful in this game you need to adapt and move on. The market is quickly moving away from these under performaning WIFI clones as customers needs grow and require more speed. :)

There's a lot to like about Telrad's LTE system, but I think the major drawback is that it's at least initially too expensive and complex for smaller or new WISP's to deploy. It really doesn't make sense to deploy Telrad unless you're planning on a very dense or urban deployment where your on the fast track to pack 100+ customers on each AP. If you're not packing that many customers onto an AP, then your fastest ROI is going to be an ePMP, or PMP450 solution... especially now considering they just released their 'lite' PMP450 AP at reduced pricing (IIRC it's like $1100). In addition, if you're in an exclusion zone, or deploying in an area where 3.65 is already being used, then Cambium's your best bet for deploying something hardy enough to operate effectively in the unlicensed bands. They're also hard at work bringing a 900MHz flavor of PMP450 to the table, which should bring some good non-LOS performance to their portfolio, and they also have PMP450i on the way, which will introduce a new lower cost SM form factor (imagine Force110 + PMP450 SM) and new AP's and SM's that will have access to the lower UNII-1 band. Anyway, point being, Cambium isn't sitting idly by, I think they're trying to listen to customer requests and bring some cool tech to the table, but there's only so much time and money for R/D... they have to pick and choose their battles.

2 Likes

We looked into deploying some Telrad. It's not worth the cost considering the number of users per site. We'd never make any money. For the cost of one Telrad BTS, we can deploy a full 4 sector cluster of 450. There's a lot more back-end devices and software required to run LTE on more than a small scale as well, which adds up to a lot, A WHOLE LOT.

We tested Wimax as a replacement for 900MHz. It didn't work at all with the foliage we have. And as we all know, Wimax is basically dead now. We also tested 3.6GHz 450 at a few 900MHz customers. It also didn't work. There's no reason to believe LTE will work through a 1/2 mile of trees either. Physics is physics, GHz do not "penetrate" at all, at least at the power levels we get with unlicensed and even 3650, I don't care what fancy technology you have, or claim to have.

You also have to consider interference. LTE and interference do not go together well. I have a lot more confidence in the Canopy scheduler to get bits through some noise.

I bet Cambium could do Canopy 4x4 MIMO (maybe even beam steering?) that is backwards compatible with existing 2x2 MIMO PMP450 SMs. Maybe that's the next logical step after the PMP450i.

2 Likes

Not familiar with Telrad, but after all of my experiences I cannot take 3.65 GHz anything seriously.

We're deploying massive amounts of PMP-450 in the 2.4 GHz band. It is, simply, the nicest, most cost-effective platform we've ever used or demoed.

By "underperforming WiFi clones" I assume you mean ePMP

1 Like

Telrad is way to expensive to be competitive. With a lot of required backend stuff it is to complicated. We need a simple Platform. We like ePMP a lot as it is very cost effective and simple. The SM has a reasonable size and gain. We would buy 3.5 GHZ 450 when there is a higher gain SM without this big reflector. A 15db panel instead of this 8db SM would increase coverage a lot.

LTE in 3,5 is unusual so there is no price scaling effect. Cambium should improve 450 further.

Matt asked if we would want a 3GHz 450 SM board in a 450i type integrated panel version (about 19dBi, and tilted 45 degrees of course). Me want... NOW! I'm going to move my house from UBNT 3.65 to 3.6 450 and I'd really like to test such a thing. I have a bare CPE now shooting through a tree between me and the tower that is a mile away. The UBNT CPE is 13dBi, but the bare 450 SM is only 8dBi and that's probably not going to work. I need my QAMs!

The 450d SM is also probably what we're going to move to for 5GHz. It's sexy, especially with the radome.