I think this is possible. Can you please explain how this solves the problem?

Well, it may not solve the problem if the supply is dropping out, but if the problem is the voltage drop across that 12 feet of cable it just might solve the issue.

It also might give you the margin to avoid the micro freezing up.

Have you measured the voltage at the micro when it stops responding as well as at the supply?

 

I am just getting into electronics, so please forgive me if I ask stupid questions.
Integrating a resistor would cause the voltage to drop, right? And because of the voltage drop, it limits the power?
What specifics are required to determine this?

The first thing to look at is what is required by the LED panel. What happens as its supply voltage is lowered? Do the LEDs just get dimmer, or does it fail to work? Since the LEDs are addressable, I'm going to guess that it's supply voltage can't drop very much before it stops functioning properly. If this is the case, then adding a series resistor won't solve the problem. But it depends on just how the addressing is done and how the panel functions. Do you have a manufacture and part number for this panel?

 

If I understand you correctly, you suggest to use a second power supply/adapter for the controller?
I understand this would solve the problem of the ESP controller not getting sufficient power. But this would mean that I need to use two outlets which is undesirable. Also, this would not solve the problem of overloading the power supply of the LED matrix, right?

A bigger power supply would be the simplest solution. Since each LED can consume max. 60mA at 5V and I have 256 LEDs, I would need a 80W power supply at least. This becomes quite heavy an bulky.
I hope to find a more elegant solution, also since the LED matrix does not constantly ask for a lot of power. I just want to prevent it from being able to ask 80W.

If all 256 LEDs can be turn on at 60 mA, that's over 15 A of current.

A switching power supply that can deliver, say, 20 A of current shouldn't be too big or bulky, but we have no idea what your size/weight constraints might be.

 

Like is mentioned in post #22, try reducing the voltage setting of the power supply to 4 1/2 volts or maybe even down to 4 volts. LEDs do not need to be run at maximum voltage and current.

 

Welcome to AAC.

Don’t forget that if you rely on current limiting to prevent overloading the supply, the matrix will vary in brightness as more or fewer LEDs are involved.

Completely separate from solving the supply sag problem that can lock up your MCU, you should algorithmically level the brightness of the LEDs in the matrix so fewer LEDs lit doesn’t mean much brighter elements than when more are on.

This means starting with the expected maximum number of LEDs on, setting a baseline maximum brightness then assuring that no matter how many LEDs are on, individual LEDs always have the same apparent brightness.