IOT RANGE: Photon, Electron, Argon, Boron, Xenon

Hardware
1

Inspired by how the term “Open Source” got coined, https://opensource.com/article/18/2/coining-term-open-source-software

I was thinking with the Particle.io new devices: the Argon, Boron and Xenon on their way, it might be a good time to coin a new term for the the various ranges we get with different devices and call it the IOT RANGE. By Jeremy Ellis, Feb 15, 2018 @rocksetta

I did some research and found this sensible site https://boundless.aerohive.com/technology/What-Is-The-Range-Of-Wi-Fi.html

Unfortunately, all that information is not very useful to me. What I would like to know is: inside and
outside realistic ranges in meters or feet for my devices.

Both change quite a bit. Inside Wifi is effected by number of users, walls, metal framing (Faraday cage), microwaves, concrete floors etc etc. Outside Wifi is very strange as some days you get great stability and other days I can barely leave my backyard.

So I am going to try to coin the set of values called the “IOT RANGE” for any device that communicates using any method or protocol.

After two years of using 15 photons daily in a wifi busy High School, In my opinion I would say the Photons IOT RANGE is about:

The Photon:Wifi IOT RANGE 3:10:20:60 meters (10:30:70:200 feet)

This should cause lots of controversy while people argue the values :slight_smile: but in the end an average of peoples values may actually be useful.

The format would be
DEVICE:METHOD IOT RANGE inside bad day : normal inside : normal outside : fully outside including the Gateway perfect day and stable connection.

Values in both meters and feet in brackets would be polite but not needed as long as the unit is shown.

Would be interesting to see generic Bluetooth, Bluetooth5, cellular, 802.15.4 radio and Wifi IOT RANGES for various devices.

@jberi any thoughts?

2

You’re going to need some way to define “bad day/normal day, inside/outside”. Is the Line of sight, are there obstructions, what’s the transmitter strength, etc, etc…

2 Likes
3

Your missing the fun part!

This should cause lots of controversy while people argue the values :slight_smile: but in the end an average of peoples values may actually be useful.

20 people will come up with 20 different results. That is why we tend to use averages. (works as long as people don't state ridiculous values to skew the results)

Should be fun. What is your gut instinct @Moors7 for the Photon Wifi IOT RANGE?

Note: Particle.io employees might want to check there employment policies before posting IOT RANGE opinions :neutral_face:

4

Quite the interesting, albeit ambitious, project! I share @Moors7’s skepticism around the challenge of creating concrete definitions. The real trick would be coming up with a system that is reliable and repeatable. If you look at most wireless datasheets, they provide controlled experiments, with different ranges of variables and with non-committal numbers. Maybe there’s something to glean from test houses like UL or TUV? Maybe standards bodies like NIST or ISO? Curious to learn more about your thoughts!

I’d also say range is just one number that one should look at and the others are very much inter-related, such as operating voltage, current consumption, bandwidth/throughput, latency, etc. Maybe coin terms for that too :slight_smile:

1 Like
5

Yeh good point. If stating IOT RANGE you probably should include some qualifiers such as stating strength of gateway etc (although I have no idea how to estimate that). My values are a merge of School Wifi and home Wifi experience, whereas all my outside values are using an inside based gateway. I would expect much better outside values using an outside based line of sight gateway. On the flip side my results are real estimates from an experienced non-Particle employee. No reason for me to fudge the data to increase sales etc. (Not saying Particle would do that, just saying I have no reason to do that.)

I also strongly believe that lots of data, even from non-professionals, is still better than no data at all.

Of possible interest, my Raspberry PI 3 has way worse WIFI range inside the house than my Photons. I don't have enough data or experience with the RPI3 to state the whole IOT RANGE, but I know that I have got to have my RPI3 within 2 m (~7 feet) of my Wifi router (other side of a wall) at my home, whereas with the Photon, I can easily work 12 m (40 feet) or more away from the same router (and same wall).

Trying to get my school to place an order for: 2 x Borons, 15 x Argons and 15 x Xenons. Wish me luck, raising money is not my strength :grimacing:

1 Like
6

Looks like my charming personality worked. I got my order in. Nice that the billing happens in the summer I can put it towards the next school years budget.:rofl: , I also just saw and purchased this bit of sorcery. The Ethernet Featherlight shield.

I will include a diagram but this makes the IOT RANGE information even more important. I can adapt to the photon having good and bad Wifi days but with the mesh, especially with a starting good Ethernet connection the IOT RANGE information becomes really important for spacing your mesh.

So your IOT RANGE now becomes really important to determine how many Xenons are needed to cover a certain distance.

ethernet%20mesh
ethernet%20mesh1496×4776 70.4 KB

On the low end using my Photon if x = 3 m (10 feet) then with 10 devices Y can only be 30m (100 feet), however if x=my high end of 60 m (200 feet) then Y could theoretically be 600 m (2000 feet) a huge difference. And who knows the new devices may have 10x the ability of the old devices. We will just have to wait and see.

But for now I think my IOT RANGE could be quite useful to people thinking of purchasing Particle Mesh devices.

2 Likes
7

If you’re serious about this whole IOT RANGE thing, I really do think you should come up with some standards that things can be compared to. Have some kind of reference. “a good” day might be when my Router isn’t overloaded with different devices, or my neighbors are out of town, or channels have been picked better by a router, of I’m not sitting in front of the device (a legitimate issue I’ve been having recently :/). There needs to be some way of saying:“assuming these and these conditions, you can expect these and these results”.

Just my two cents :slight_smile:

2 Likes
8

I agree @Moors7 the problem is that most of us (including me) don't have the tech knowledge to know why the Wifi is bad on a certain day. A way around this issue is just to ask real users, using real devices their opinions. No one is going to be perfectly correct but the group knowledge or averages could be very useful.

The data outliers then become interesting. If most people have an outside Photon Wifi range of say 30 meters (100 feet), but Joe Somebody is consistently getting 100 meters (300 feet). What the heck is Joe Somebody doing that the rest of us might be interested in. Can we copy it or is it specific to the location.

The community can then help improve the product.

9

But in able to make an informed decision, you'd need to know the condition in which you do those tests. I can't stuff a tag on a product saying "indoor range: 50m", and conveniently forge to to mention that this was in clear line of sight, inside a Faraday cage.

Joe Somebody could very well be getting an outside range of 100m with a clear line of sight, and a directional, or amplified antennae. Without him stating that in his IOT RANGE, there's no way of knowing what he did or why he got those results.

That's the exact issue. Without stating the variables, there's no way of saying what's happening for what reason. You might as well not bother then, if there's no reference to go off.
If the packaging states "50m with clear line of sight, no interference", then I can be sure those values can only drop. There's something to go by, unlike when it would say "50m. Could be 10m, could be 250m, éh... we don't know".

If I had to mix some chemicals during a science experiment, I'd like to know the reference values before dumping everything together. "Might create pure diamonds, might acid-melt your face of. On average, 42".

I totally agree that it would be useful to have more real life data to go by, but then there'd need to be rules if you'd want it to have any meaning whatsoever.

1 Like
10

I think we need to agree to disagree on this one @Moors7

Have a great day.

1 Like
11

The question is, how far are you both willing to go with this :rofl::joy::wink:

1 Like
12

Approximately 42 :wink:

13

I don’t think you have to come up with anything new here, radio ranges have been quantified for many years – the hotspot team I worked on always had meaningful numbers to back up their ratings (other, smarter people that is, I was not an RF engineer), and certainly cellular networks have standard definitions they follow that would be similar if not directly applicable – mesh networking standards I’m not that familiar with but I’d assume there would be some meaningful measurement standards already defined out there that would be helpful – something corresponding to the definition of erlang for example that would be meaningful in a mesh network environment.

1 Like
14

This looks like a good starting point – an evaluation of a WiFi mesh network done at MIT – the referenced documents at the bottom would also probably be a good source of research material.

3 Likes
15

I am a general class ham radio operator with years of experience dealing with RF at many different wavelengths. I could go on and on and on about RF… but I’m not going to. The performance (range) of an RF device depends, more than anything, on the antenna system. Working in the WiFi frequency range where wavelengths are short, the signal doesn’t penetrate things well, and the signal is easily reflected, antenna placement becomes critical as well. Typically, if you’re going to characterize the performance of a radio, you would use an antenna that is an isotropic radiator (with no gain or loss in any direction). I question the value of the results if you do this, however, since antennas with gain, properly aimed, will out-perform the isotropic radiator hands down. I’ve been considering coming out with a line of antennas for particle devices. I believe I have some novel ideas using modern materials and methods that could produce positive results. If there’s some interest, perhaps I’ll move forward. For the purpose of this thread, however, the type of antenna you are using, the quality of that antenna (is it resonant at the desired frequencies?), the gain of the antenna, and its placement all come into play when measuring the performance of these radios… keep that in mind as you press on!

7 Likes
16

Always best to have more things for the Particle products. Best of luck. From my experience the antennae I have tried has not significantly improved the Photon range, since our issues at School are overloaded Wifi not really a problem with either the Photon or the antennae. Outside, at home the range seemed more variable on different days, but the duck antennae did seem to get about a 10-20 percent increase in range.

Most people would just connect the antennae and leave it connected, however in the classroom the students connect and disconnect the antennae. I have had three Photons have the antennae connector ripped off the Photon. For this reason I do not use the antennae any more.

P.S. All three Photons still work fine with the antennae connector removed.

17

Quite a bit of engineering probably went into the built-in antenna on the Photon; its performance is most likely difficult to beat. It is very difficult to design an external antenna system for frequencies in the wi-fi range that can be mass-produced and works reliably. A few factors that come into play are resonance, the impedance of the transmission line, and gain / directionality. I’m pretty confident that if you properly configure good antennas you’ll see a very significant increase in performance. It would be interesting to see if the built-in antenna is directional… does the physical position of the Photon effect its performance? You might also play around with parabolic reflectors… put the photon at the focal point and you should see gain… Anyhow… antennas are an interesting subject that’s worth consideration. Good luck!