I'm trying to find out what the following signal/line code is known as.

The communication is performed using a single wire, other wires are ground and power (5V). The signal wire is kept high (5V) and pulled low (0V) to send a signal.

How I understand it:

  • 0: long low, short high
  • 1: short low, long high

Pulseview diagram displaying single frame

Signals seem to consist of frame groups where each frame is 12 bits and followed by the next frame after 30ms. The next frame group is preceded by a longer pause.

Pulseview diagram displaying two frame groups

It does not seem to be any well-known line code that I could find.

The device generating these signals is a VELUX WLI 130, they are received by a Velux WLC 100.

Edit: If there is no specific name/protocol for the whole signal, I'm interested in whether long low, short high to represent a zero is known under a specific term.

  • \$\begingroup\$ looks like a simple remote control IR code \$\endgroup\$
    – jsotola
    Commented Nov 29, 2023 at 15:58
  • 1
    \$\begingroup\$ you identified it in the first picture and in the last paragraph ... what more is there to know? \$\endgroup\$
    – jsotola
    Commented Nov 29, 2023 at 16:06
  • 1
    \$\begingroup\$ @jsotola How to perform the decoding is not really what I'm seeking an answer on. It would be a PulseView decoder but that's not important to the question. I have tried to clarify my question. \$\endgroup\$
    – Waylander
    Commented Nov 29, 2023 at 16:41
  • 1
    \$\begingroup\$ It also seems to be the same as HIT caseette tape format from 1975. As noted in linked article, you don't have to know the clock speed: time the low, then the high, and compare. I still don't know what the encoding is called. \$\endgroup\$
    – jonathanjo
    Commented Nov 29, 2023 at 16:59
  • 1
    \$\begingroup\$ @jonathanjo thank you for that link, a very interesting read. It does indeed look like HIT with the low time moved to the front because of the signal's default high state. If you add it as an answer I can mark it as such unless a more fitting other answer comes in. \$\endgroup\$
    – Waylander
    Commented Nov 29, 2023 at 21:35

2 Answers 2


With constant bit time and variable on-off phases, this is a variety of pulse width modulation. PWM is most commonly used for encoding an analogue signal; however there's no reason we can't use it for encoding binary data.

You have PWM with bit time = 1.75 ms, low phase first, Zero = duty cycle < 50%, One = duty cycle > 50%.

A line code pretty much exactly as you describe was used as a cassette tape format, published in 1975 in "Hobbyist Interchange Tape System" Popular Electronics, Sept. 1975. This is inverted from yours: bit is from rising edge to rising edge, with a bit time of between 1.25 and 30 ms: a zero is represented by duty cycle < 50%, a one by duty cycle > 50%. They appear to use 33% and 66% approximately.

Note 1: the line code for WS2812 LEDs is similar, but not really PWM in that the bit time isn't constant. But it does use short-long and long-short phases to encode the bits.

Note 2: At fist glance this looks similar to Manchester encoding. But we can see it's not because two adjacent short phases are not the same length as a long phase.


This does not immediately resemble any known protocol, not even IR protocol, so it must be a vendor specific custom protocol, likely with their own internal name for it.

Looks like pulse length (width) encodes the bit, so not the space (distance) between pulses.

Thus, it is called pulse length or pulse width modulation.

The opposite would be pulse distance modulation.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.