Power impact in a wireless sensor network architecture

Question

this is not a real question but a way to talk about the following topic. I was thinking on a monitoring system based on a Sensor part, a Computing part and a Radio Transmission part.

My question is how much is the power impact of the radio part compared to the power consumption of the sensor part or the computing part? If I want to answer myself I think it depends on the specific Radio tecnology consumption in idle state or active, the sampling rate, the type of sensor I want to use, if I use a microcontroller based unit or a microprocessor one.

Am I right? What can you tell me more about this? Any good book or article to read about this topic?

Thanks ;)

Answer 1

This question (oh wait, it's not really one?) have multiple possible answers. It greatly depend of what you're device is doing compaired to how it will communicate.

For example, the power consumed by a smartphone is much higher than the one used for its wifi chip, but a wifi chip will consume much higher power than a lower power, low frequency, optimized microcontroller.

A good way to have a fair idea about that is reading the power specs of your components in your application, and on a general manner the complexity of your device. I'd say that often, it's the computing part (including display generation such as graphical processing units) which take most of the power. But if your device is about sending one value each second (assuming your computing unit is small and doesn't do quick&dirty polling), you gotta be sure most of the power will be consumed by RF unit.

It would be also good to have a general knowledge of power comsumption (something like power consumed : Wifi > Bluetooth > Bluetooth LE > Lora)