A low pass filter allows signals with a low frequency to pass. What does that mean? Well, data sources often come with noise that we don't want. An example of such a source is a thermometer, or an accelerometer.
Noise is different for thermometers and accelerometers: a thermometer's intent is to measure the ambient temperature; thus a lonely spike is bad! However, for an accelerometer, it is typically the spike itself is what you want to measure. How can you achieve that?! The answer is filters!
This repository contains a simple server that outputs a binary signal with heavy noise disrupting it, and a webpage that visualizes the signal sent to it, and tries to decode the message.
The task associated with this challenge is to implement a proxy for the server that will filter out the noise and emit a simple binary signal to the webpage.
The first step to working on this challenge is cloning the repository.
git clone git@github.com:akainth015/low-pass-filter-demo.git There will be 2 or 3 servers running on your computer once this challenge is complete. You can start the signal server first.
node noisy_server/index.mjs
PORT=1076 node noisy_server/index.mjsNext, let's see what this noisy signal looks like.
python3 -m http.server --directory decoder 3000
# open http://localhost:3000 in your browserNow, go ahead and start the denoiser!
node denoiser/index.mjs
PORT=1077 node denoiser/index.mjsYou should see the same signal if you connect to it from the webpage.
Edit denoiser/index.mjs to try and fix that! The final output from your
filter should be a 0 or a 1. Good luck!
Tip: depending on how you scope your variables, the filter can retain what it has "learned" about the noise when you press the "retransmit" button. That means if your filter starts working properly halfway through the broadcast, you can just press "retransmit" and you should be able to decode the full message.