So I want to implement a room occupancy detection system at my institute. The idea is to create multiple sensing nodes consisting of a PIR sensor, a microcontroller, some form of wireless transmitter and a battery. Many of these nodes would be placed throughout my institute and all of the data(which area shows activity) from the nodes should be available on some central monitoring system, over the internet
The problem I am facing right now is in deciding what form of wireless should we use? We do not want to depend on the institute's WiFi network and want to create an independent network.
Here is a rough outline of a floor at my institute:
Roughly 4-5 nodes would be placed in each Lab/Classroom.
Points that I'm concerned about:
Should not cause interference with existing WiFi network/devices
Low power
Range
I've thought about using the ESP or nRF24 modules, or some kind of hybrid network, but I'm not able to decide. What do you guys suggest?
Xbee is a good option and is robust, but it will drive up the cost per node by a significant amount and I don't want to go for it unless absolutely necessary.
Whereas in case of LoRa, isn't it best for outdoor applications? Would something like this module be good in my case?
you can use Lora anywhere - make sure you have the correct frequency for your country
I think the module linked in post #2 is just a transceiver - you would require a host microcontroller
have a look at these devices which you can program using the Arduino IDE 32u4 lora
I have several 32u4 Lora devices I will try the code from
clearly the development of the Lora mesh is in its early days!
Unlike (the more expensive) XBee which works off the shelf!
the code to connect to a LoraWAN gateway is large, e.g. my ttn-otaa code shows
Sketch uses 22868 bytes (79%) of program storage space. Maximum is 28672 bytes.
this is a run talking to The Things Gateway
a simple Rx-Tx test program for a peer-to-peer connection between two 32u4 devices gives
Sketch uses 9506 bytes (33%) of program storage space. Maximum is 28672 bytes.
if you require more memory there is the Adafruit Feather M0 Lora
Thank you for all that info @horace. Since I have to create a deployable system and not merely a hobby type device in the span of about 2 months, I think I'll probably go with a tree-type network using the nRF24l01 modules. Or maybe if the range causes issues, I can use LoRa as well at some points of my network.
All the walls are fully reinforced concrete, with many cables running all around the 8-storey building, I'll first have to do some range testing. Meanwhile, I'd really appreciate any input that you guys have!
This library leaves little free space in the ATmel... Not enough for my own Arduino sketch.
A nice thing would be to have a sketch on this board that would emulate AT commands via the serial port.
The board would be considered as a modem and would simplify the use.
tk5ep:
A nice thing would be to have a sketch on this board that would emulate AT commands via the serial port.
The board would be considered as a modem and would simplify the use.
Well volunteered, do make it open sourced when its ready.
.
There is however already a module specifically designed for LoRaWAN and addressed via the serial port, the RN2483.
srnet:
Well volunteered, do make it open sourced when its ready.
.
There is however already a module specifically designed for LoRaWAN and addressed via the serial port, the RN2483.
and there is a module with onboard RN2483 LoRaGo Mote