Hacker News Reader

Comment by bdamm | Hacker News Reader

Don't get too excited. Low-power devices get really interesting when they start to scale up in numbers. None of the technologies listed in this article have demonstrated that capability. The coming IoT shakedown is going to leave quite a few ebullient "visionaries" crying in their martini while they consider their next startup.

* Star networks are infrastructure heavy, requiring build-outs on part with cellular networks, and having low per-base density.

* LoRa's CSS (Chirp Spread Spectrum) approach doesn't promise to scale well, as devices start to interfere with each other over its very wide spectrum scattershot. There are other problems.

Disclaimer: I work for an IoT mesh company (SSNI/WiSun) and we're busy cranking out massive networks that actually work. Sadly you can't get a dev kit at Fry's yet.

View on HN

Shawn from BeepNetworks here. Check out OnRamp (now Ingenu). Their coding scheme allows 1600 simultaneously communicating devices per spread factor per channel. That 1600 factor means their base station can handle massive numbers of clients compared to cellular. http://www.scribd.com/doc/273814075/On-Ramp-Wireless-White-P.... That paper is certainly biased in their favor, but the tech is real and they have serious deployments.

The counterpoint to mesh is that it's a) much more complex to implement b) uses more power since nodes need to relay each others' communications.

It will be interesting to see how it plays out, and who ends up crying in whose martinis as you say.

Hm, 1600 sounds like a lot. But depending on the applications it sounds like it might not be that much if it's 1600 over three square miles.

that's 1600 devices simultaneously communicating. a sensor might only spend a fraction of a percent of its time transmitting data, so the number of devices supported is orders of magnitude higher.

Hmmm. I'd pay very careful attention to the fine print on that. The way their scheme works is that each node chooses a random time to start its transmission, which means that all their chipping sequences are hopefully desynchronised. Then the AP tries decoding from all possible starting points at once. If I'm understanding spread-spectrum correctly, each parallel receiver pulls out only the signal from the node (or nodes) whose chipping sequences are synchronized with it, while all the other node's signals appear as noise to it.

Which is the catch. Run too many nodes at once, and the signal-to-noise ratio will rise and your signals will start to fail to decode. What this generally means is that nodes at the edge of the network - the ones with the lowest SNR budget - start to drop out first. I'd be interested to see what their link budget and range is with 1600 devices on the same node, or conversely how many devices they can fit on a node at the advertised link budget and range. (Also, just how well they cope with a mixture of near and far nodes.)

But didn't IBM sided with LoRa ? doesn't it mean they tested it for scaling ?

I know a guy who worked for years at an IBM research lab near Zurich on Lorawan. I posted his thing a couple of weeks ago on HN (https://news.ycombinator.com/item?id=10344956).

Reply on news.ycombinator.com