What is the best wireless network architecture for 1000 nodes...?

Go To Last Post
31 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi All..
I have to design a wireless network as follows:
I have 1000 sensors.. Each sensor, on detecting something has to post the "detected info" on the cloud.

Imagine that these sensors are distributed in a square shaped area where the distance between each sensor is 2.5 meters..
What is the best, simplest, no packet loss, cost effective, congestion free, interference free wireless network..?
Xbee is bit costly and needs a co-ordinator additionally..
WiFi .... May/May not get the signal for such a long range ..(If 50 sensors are in series)
Lora... 1000 LoRas...?????

Any Hopping network?

This topic has a solution.
Last Edited: Thu. May 9, 2019 - 04:27 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Intruder195 wrote:

What is the best, simplest, no packet loss, cost effective, congestion free, interference free wireless network..?>

Only in fairytales do those requirements exist. Reduce your expectations andyou’re more likely to find something in reality.

My initial thought is bluetooth 5. The nordic chips are small and reasonably priced.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How often do they have to send?

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

... and how many bytes in the "detected info"?

Ross McKenzie ValuSoft Melbourne Australia

This reply has been marked as the solution. 
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Intruder195 wrote:
Xbee is bit costly and needs a co-ordinator additionally..

 

But it WORKS!

 

YOu really are not providing much information.  1000 sensors 2.5 metres apart is a very dense coverage.  How far away is the host(coordinator)?

 

XBEE modules can also act as repeaters, so not only can the module service your node, it can regenerate the signal as well.  I am using both the 2.4Ghz for some local stuff similar to what you are doing, and the 900Mhz for my long range (2 miles) links.

 

Intruder195 wrote:
What is the best, simplest, no packet loss, cost effective, congestion free, interference free wireless network..?

Easy answer pick two requirements, and throw the rest out.

 

The ideal for your situation would be the XBEE as the modules do all the heavy work, AND they already carry FCC/CE certification.  THey are not the cheapest solution, but if you buy 1000 I am sure you can get better pricing. 

 

As King Samperi says "There is no free lunch"

 

Jim

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jgmdesign wrote:
Easy answer pick two requirements, and throw the rest out.

Good, Cheap, Fast, pick any two, you can not have all three.....    if you picked Cheap + Fast you did not choose wisely!

 

Jim

 

 

(Possum Lodge oath) Quando omni flunkus, moritati.

"I thought growing old would take longer"

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For every 1 min

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Oh ..
Thanks..
I'll try Bluetooth 5

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The frame it has to send is..

FN04H234S1T

That's it..
That too for every 10 min or so

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hahaha..
Yeah..
Nice...
Xbee.. again .. I have to set many coordinators..
Like 10 nodes one coordinator...
Etc..
Yeah..
Xbee really works well.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Intruder195 wrote:
Hahaha.. Yeah.. Nice... Xbee.. again .. I have to set many coordinators.. Like 10 nodes one coordinator... Etc.. Yeah.. Xbee really works well.

 

Um, no.  Thats not how it works..

 

But then again with your expertise in RS-422 you would know better right:

https://www.avrfreaks.net/forum/...

 

I'm out.  All the best on your quest.

 

Jim

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:
My initial thought is bluetooth 5.
A subset of Bluetooth 5 are mesh capable.

Qualified Mesh Products | Bluetooth Technology Website

the MCU in that list :

  • Cypress Semiconductor
  • Espressif Sytems ESP32
  • Nordic Semiconductor nRF5
  • ON RSL10
  • Samsung ?
  • Silicon Labs
  • STMicroelectronics
  • Toshiba

 

BLE Mesh - Cypress Semiconductor

ESP32 Overview | Espressif Systems

Bluetooth mesh - nordicsemi.com

RSL10: Radio SoC, Bluetooth® 5 Certified (QFN, SiP, automotive in preview)

Bluetooth 5.0 Low Energy Certified Modules, ICs & Software - Silicon Labs

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Zig-Bee but not necessarily Xbee. There are plenty of offerings. Yes, you need a coordinator. In the end, these messages must go out to cloud through a single pipe, so you need a such approach anyway. You may want to split your 1000 nodes in few distinct Zig-Bee networks. And plenty of testing.

 

Intruder195 wrote:
Hahaha.. Yeah.. Nice... Xbee.. again .. I have to set many coordinators.. Like 10 nodes one coordinator... Etc.. Yeah.. Xbee really works well.

I don't know about Xbee. I know that a Zig-Bee coordinator can easily handle at lest 100 nodes.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

And by the way, what's about those tags in your fist post ?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Haha..
Yeah that's the exact reason why we want to change the topology..
Want to make it so simple..

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

We came to know that the main Atmega chips are cheap quality ones..
And laying cables takes a hell lot of time..
We want to avoid it..

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Intruder195 wrote:
We came to know that the main Atmega chips are cheap quality ones..

 

Please explain more about how you came to this conclusion.

 

--Mike

 

EDIT: or maybe I misread your statement -- did you mean the

ATmegas are of "poor quality" or that they are "inexpensive

but of good quality"?

 

Last Edited: Thu. May 9, 2019 - 05:37 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Before the thread goes off topic note that the "mad tag" thing in #1 seems to be a forum bug that has struck a couple of times recently and peppered a post with just about every conceivable tag available. Just try to pretend it never happened.

 

It's much more interesting to know what the "best" wireless network system is. (if that superlative can ever be justified?).

 

Moderator

 

PS I also wonder about the use of "cost effective" as opposed to "cheap". The implication is presumably that as long as it is a "good" ("best"?) solution it does not actually matter about hardware costs - is it possible to actually quantify that - like what is the projected budget for these 1,000 nodes?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

xBee comes to mind, but with that small a packet and only transmitting once a minute, or once in 10 minutes depending on which post, counterfeit NRF24L01's should do nicely.

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The PCB manufacturer bought Atmega chips from two vendors..
1 . Vendor 1( I don't want to mention name)
2 element14.

The element 14 chips are working fine once the code is flashed ..
Where as the vendor1 chips, either UART doesn't work or whole chip itself doesn't respond after two or three power cycles..

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Intruder195 wrote:
The PCB manufacturer bought Atmega chips from two vendors.. 1 . Vendor 1( I don't want to mention name) 2 element14. The element 14 chips are working fine once the code is flashed .. Where as the vendor1 chips, either UART doesn't work or whole chip itself doesn't respond after two or three power cycles..

 

Sounds to me like you got some counterfeit parts from Vendor1. 

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sound more like tolerance on the internal clk  from  1. was 10% chips and 2. was 2% chips.

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

sparrow2 wrote:

Sound more like tolerance on the internal clk  from  1. was 10% chips and 2. was 2% chips.

 

 

 

Let's see: UART. Some work, some don't.I'll bet you're right. Need a crystal to make uarts work.

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes..

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

So there was was nothing wrong with the chips from vendor 1. It was just the wrong chips that was ordered (the devil is in the last letters after the number)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0
  • ON RSL10

new arrival today :

RSL10 Solar Cell Multi-Sensor Platform - ON Semi | Mouser

note its long lead-time; IC and module have a somewhat reasonable lead-time.

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Intruder195 wrote:
Lora... 1000 LoRas...?????

Arduino Blog » LoRa made easy: Connect your devices to the Arduino IoT Cloud

[next to last paragraph]

...

All the available solutions for LoRa® currently focus on collecting data, but they do not address it from the other way round i.e. sending data from a centralized application to the LoRa® device(s). Arduino IoT Cloud now lets you do this — you’ll be able to control actuators connected to your device by sending messages via LoRa®, with no coding needed.

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There are significant restrictions in sending data to the remote nodes over LoRaWAN - you cannot, typically, just send unsolicited.

 

EDIT: Especially if you're looking for the "power consumption to go as low as 104uA" mentioned in that article!

 

This makes LoRaWAN unlikely to be suitable for real-time remote control of actuators...

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
Last Edited: Tue. Feb 11, 2020 - 06:06 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Can the sensors be linked in groups? So 160 meter runs of twisted pair (RS485 or CAT5) with 64 sensors (160 = 64*2.5) connected to an SBC with WiFi. Strategically locate the SBC's to reduce the number of WiFi access points needed. Then you need a plan for the access points; I guess Starlink will be starting soon.

my projects: https://github.com/epccs

Debugging is harder than programming - don’t write code you can’t debug! https://www.avrfreaks.net/forum/help-it-doesnt-work

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:

no packet loss

 

You want to achieve something on wireless, which even doesn't exist on wired networks. TCP was invented for reason. If you would be serious about making it into reality you wouldn't have such requirements. First I was thinking some mesh network. But maybe own protocol over some more primitive protocol could work, like 433MHz or something similar. And time slice protocol? On WiFi there are some non-standard tweaks which can for example on mikrotik devices increase the bandwith a lot, one of them is to give clients really specific slots to talk. Have a master to broadcast frequently clock info to everybody and each node generate their own unique time slot. Have a lot of dead margin time, so even when there will be drift, have low chances for overlap. Has resend/fail mechanisms anyway, start smaller and slower to do a proof of concepts, even 5 x 5 would probably be a good point to start.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

1000 node != simple

 

1000 node != low cost

 

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!