I want to make a network in a forest area (2 acres) to monitor CO2 and send the data to the cloud.
The sensors will be distributed along the 2 acres.
1.- What HW do you recommend us LoRA, Xbee or Xenon?
2.- What is the line of sight of the Xenon?
3.- Is there routing process in Xenon like Xbee modules ?
You can get started now using the long Range RFM95w Radios or the XBee 900mhz HP radios and an Electron or Photon to collect the data and push it to the web.
The MESH range between nodes will not get anywhere close to the range the RFM95 and XBee Radios can give you at up to 1-mile range in the city without any special directional antennas at least.
What is the total number and density of sensors you need for your application?
If less than 10, maybe each could be an electron. >100 mesh networking generally becomes a better solution for both cost and device management. 10-100 is somewhat of a gray zone where star networks are more feasible.
Do you need device to device communication or is connecting to the cloud sufficient?
If each device just needs to send/receive messages to the cloud, maybe get OTAs, than the problem is simpler. We’ve seen community members like @RWB interface LoRA radios with Particle devices. Local communication will require either Particle Mesh or rolling something like MQTT or a custom solution.
What forms of cloud connectivity are available?
Is 2g/3g, LTE or Wi-Fi available somewhere that you can tap into? Can every device be a 2g Electron, for instance?
Are there other constraints, such as cost or power, you need to consider?
Is there power near each device? Is solar feasible? How much can you spend on each node?
When do you plan to pilot/deploy?
Particle Mesh development kits aren’t shipping until this summer and the mass production modules later this year. Do you need to pilot now? At what scale will these be produced and do they need modules instead of devkits?
Do the devices need to be updated in the field?
If you use Particle devices, like an Electron or Xenon, they will be able to updated in the field. If you use a LoRa or Xbee module, you’ll need to figure out how to update them. Figuring out how to OTA reliably and securely is a fairly complex challenge. If you don’t expect to need to update them than this isn’t a concern, though I personally think OTA is critical, especially if you would have to find them among 2 acres for a patch.
Sounds like an exciting project! Please share your progress
I’ll add that you may want to consider your Project’s requirements at both Ends first.
Start with the bare requirements of your sensor nodes. What Voltage and Current do your CO2 sensors require? How often do you want to sample ? I assume battery power is required for the sensor nodes? What’s the required life before recharging or replacing sensor node batteries ?
Your Power Budget for your sensors will likely control a project like this, not the MCU reading them.
For your Gateway, you will have less headaches if you can recharge your Electron Gateway with Solar, even if the Gateway needs to be located outside the 2 acre project area for decent sunlight hours.
Then you will know the required range for RF communication…and can select the best for this application, IMHO.
But keep an open mind. For example: If you only need a few readings a day, and can recharge or replace batteries every 6+ months, just go with Electrons with properly sized battery as your sensor nodes, and let them operate independently.
I agree with those above, there are a lot of questions and will really depend on what you want to do.
In addition to some of the questions posted above: Couldn’t you use any kind of mesh technology? I know Particle released several devices for this, but there are also several other out there that are available now. You can use LoRa, but it wouldn’t be natively mesh-able (https://www.link-labs.com/blog/lora-faqs).
I just did some quick google-fu, and you could even do this with ESP8266 modules (https://www.espressif.com/sites/default/files/30a-esp8266_mesh_user_guide_en.pdf). Take a look at page 6, I think it explains the Idea of a meshing to a gateway pretty well. Essentially, instead of each device connecting to the Internet, the device connects to the nearest device in range and sends its data to that, and continues down the line until a root device that communicates with a gateway (particle? Wifi Gateway?) is reached.
I guess it really depends on what you need. If you use a mesh topology, then you get the benefit of not having each individual device connecting to a wireless connection, potentially reducing power depending on your needs. The frequency you use would define your range (915Mhz > 2.4Ghz).
Is LORAS and Xbee Pro 900mhz comparable with LORAS range?
I am looking for a solution for a HAB (High Altitude Balloon)… I have a sat modem but it is not super reliable, and wish to have a faster telemetry option beside the sat modem (RockBlock)… XBee seems like a good solution… But Wonder which technology can give the greater range… It is going to be line of sight after all.
The XBee Pro 900 with rubber duck high gain antenna will provide the same range as Lora RF95W radios with 3-inch wire antenna in my suburban environment with hundreds of houses in the way which will get me 1-mile range.
If you’re putting the LORA up in the air on a Balloon then you could get 20-mile range going from the sky down to you on the ground since there are no obstructions in the way.
@frlobo, Have you considered using a cheap autopilot board that we use for UAV's ("drones") ?
A Pixhawk or APM would give you GPS, Mag, Baro, Accel, etc. A 3DR radio provides the Real Time Telemetry Data, plus the board will log to SD card. You can use many different free apps for Ground Control, such as Mission Planner.
I am actually mentoring on this middle-high school project and decided that doing our board will be better learning experience for students. I am no expert but I am teaching them how to do boards in eagle and then carve their prototypes and then send manufacturing to oshpark.
We are working on our own open source solution. We are currently finishing the last iteration of our board with XBEe footprint but can change to LORAS. Current version used Iridium and Cellular but cellular is not reliable where we live and need another primary means of communication since the iridium is also not super reliable. So we are going to use iridium+XBee or loras. Cellular will still be an option since we are using an electron but will stop transmitting when leaving 8,000ft and restart bellow 8,000 ft again.
The ground control uses a server that gets all data from particle and iridium, we will use a raspi to get XBee and send the telemetry string to the server. (Also open sourced).
From the server we have an iOS app that gets the data from the server.
Lots of learning opportunities for the kids… never thought I would be teaching node to this kids. Heck never thought that I was going to be able to teach anything… but thus kids are smarter than I am
This is the current board (will push newer version with XBee soon).
The LORA’s are cheaper, the XBee 900HP will cost more but they are FCC certified.
We have working LORA code to get you going, not sure about Xbee code that will work with the Xbee 900HP radios since some of the XBee code is different depending on XBee radio version that are not compitable with the 900HP models.
Yagi’s have the highest gain, for your ground station.
Take a Look at Digi’s User Guide for S3B hardware for all the Approved Antennas.
For maximum range, use XCTU.exe to configure both Xbee’s with XBP9B-DP , Xbee PRO 900HP 10K, latest firmware version, highest power level, API mode 2.
Resist the urge to use Transparent Mode.
Select your antenna with a cable as short as possible, and don’t use any additional connectors/converters on the coax…(select antennas with RP-SMA)