[jrockway]

Well, they don't tell you the bitrate. WSPR is a protocol that can make worldwide contact on the HF bands (3-30MHz) with 0.01mW of power. The catch is that it takes 2 minutes to send a message like "KD2DTW/FN30".

900MHz signals don't propagate like HF (by refracting off the top of the ionosphere), but they do have interesting propagation characteristics like bouncing off of passing airplanes. If you have a protocol that can extract signals from below the noise floor, and enough erasure coding (reed-solomon, etc.) to handle bursts without connectivity, you can build an urban network fairly easily. The only problem is that it's not that useful.

My guess is that they are like FitBit and just plan to blanket every home with one of these things, and allow devices to roam to whatever access point is closest (like when you walk past someone's house with a FitBit, and notice yours has synced).

[amscanne]

The site says specifically that one bridge can cover tens of thousands of devices and 50 square miles.

It also seems to be based on ZigBee (that is, ZigBee devices can be retrofitted) so it's 2.4ghz, 915mhz or 868mhz.

Like you said, there are plenty of ways to go one way (from a powerful transmitter). But those are noisy frequencies for a weak transmitter to make it several miles...

Anyways, agreed that the plan is probably fitbit-style. But the crazy specs make it seem a bit like snake oil to me...

[jrockway]

This doesn't sound that unbelievable to me. You don't need a lot of power to send a receivable radio signal. (GPS is a good example, the satellites are 20,000km away and transmit at 25W, but GPS still works just fine!)

The basic equation is:

  channel capacity in bits/second = bandwidth in Hz * log2(1 + signal power / noise power).  

So let's say they are using 1MHz of spectrum (WiFi uses 20-80MHz), are transmitting at 1mW EIRP, and have an omnidirectional receiving antenna (there is no such thing, but "assume a spherical cow", it's the worst-case anyway). At the claimed range of 4 miles, the signal will lose 100dB because of "path loss" (actually spreading out), giving you ~ 1e-10mW (-10dBm) of signal at the antenna. Let's set the noise floor at -60dBm which I have not really measured, but sounds good. That's 1e-6mW. Plug this into our formula and you get a bitrate of about 1 bit every 7ms. That's ~20 bytes per second!

So let's say that we want to collect a sample once a second from 256 devices. We get 20 bytes per second, so can sample every 256 seconds (let's say that's 5 minutes). There are 256 devices, so each one gets a 1 byte ID. Then you have 19 bytes of payload, which is fine for things like thermometers or your FitBit or whatever. Fill the rest with an error-correction code.

Now you have 256 devices, each using 0.01W of power for 1/256 of the time. With a 5Wh cell-phone battery, that's enough for 14 years of transmissions.

Now, to get thousands of devices, you can use more spectrum; there is plenty more in the ISM band. You can transmit with more power, 0.01W is what a Raspberry Pi IO pin can transmit connected to a long wire. You can get a directional antenna, since you probably aren't listening for signals from the sky or underground. You can also send less data.

Anyway, I ran the numbers and I don't think this company is claiming to violate any laws of physics! It all sounds quite possible, actually, with the right engineering work. I'm looking quite forward to purchasing an eval board!

[amscanne]

Isn't your math off?

Due to path loss, rx signal is 1e-10mW => -100dBm, not -10dBm.

Plug into the formula and we have a maximum theoretical bound of ~ 0.14 bits/second, or one byte every 42 seconds.

[jrockway]

I switched from W to dBm at the last minute, so there might be mistakes. Anyway, I think it ends up being feasible, because you have some play with transmit power, the noise floor, reflections, and a better antenna, which will all change your number by a few 10s of dB in whatever direction you choose.

Definitely not the worst startup I've seen on HN. I think they can make something useful, since their goal is quite modest.

[bryanlarsen]

Even 1 byte every 42 seconds would be very useful for many types of sensors.

[amscanne]

That's with the transmitter on 100% of the time.

[gonzo]

Assuming the path loss exponent as '2' is a mistake.

[pharkmillups]

jrockway, would love for you to test with us.

[pharkmillups]

Hey Amscanne -

There are two things we're trying to convey here (but rest-assured we're not trying to be misleading):

1. We can support a lot of connections. I mean a lot. This is based on our design, but truly limited by the receiving radio.

2. 50 square miles is optimal conditions. It's hard to quantify how bridges and radio enabled devices will work in populated metropolitan areas. These numbers are solely based on field testing, not in metro areas.

[amscanne]

I'm commenting because I'm genuinely curious about the technology, I'm not trying to tear anything down.

That said, the technology page is not very clear.

The wording doesn't currently indicate this is idealized or limit conditions. Because you say it covers tens of thousands of devices over such a large area, I assumed this would be expected (like a cell tower). Your description actually fits very closely with a cell network. Why would a device that covers only my home need to support connections with tens of thousands of devices? Would it cover me and all my co-workers at the office 1 mile away (well within it's 4 mile range)? Or would I be expected to have a bridge at home and in the office? Is it like a Femtocell in a mall or like the ZigBee bridge sitting in my living room?

Given that picture, I'm genuinely curious about the capabilities. Immediately below that paragraph are bars that show a 700+ day battery life and a transmission power of 0.025mA. Are these for a totally different use case? If so, I'm confused. If not, it's amazing and I'm very exited. How is a transmission power of 0.025mA @ 915MHZ possible at a range of 4 miles? I would have assumed 1000x that (at least) would be necessary on unlicensed ZigBee spectrum. Similarly power would need to be amped up on the receive side,... meaning receive windows would be costly.

[pharkmillups]

You should probably work here. -> [e-mail]

[jrockway]

Thanks! I'm happy with my current job, but would love to play with your eval boards (in my copious spare time :) when they're available. Send me an email ([e-mail]) and I'll buy a couple to try when they're ready.

It's rare when I see a startup on HN where I initially think "this is impossible", but do a little math and find that it's actually quite possible. Easy? No. But you're going to have a much better time working with the laws of physics in your favor than you would if you came up with something theoretically impossible :)

In the interim, check out http://wsprnet.org/. There are lots of people on the forums that are interested in weak signal work. If you have (or get) a ham radio license, you can try sending WSPR from a raspberry pi with a long wire connected to an I/O pin. It's pretty cool! (I've been meaning to set this up at work to see if I can get the signal at home on various bands, especially VHF around 144MHz or 1.2GHz. If that works, I have no doubt that you guys will be successful. And I think it will work.)

[simonvc]

:-)

phark!