# What is an anti-causal signal

## Forum: Digital Signal Processing / DSP / Machine Learning Causal or anti-causal, that is the question

Hello, I'm currently working on signal theory and control engineering. It is known that signals can be classified differently, a signal can be a power signal or an energy signal, for example. I am already familiar with these definitions. However, I would be interested in what is causal or anti-causal. According to the definition, a signal is causal if it does not exist or is 0 for all times t <0. Anti-causal would then have to be the opposite. Now a question arises when I look at the tri function (the triangle function), this is defined for -T and T, the triangle is symmetrical about the Y-axis. Is such a signal now neither causal nor anti-causal? Or does it have to be causal or anti-causal? Thanks for coming answers!

of Self-proclaimed do-gooder (Guest)

Hello, see "Signal and System Theory" by Thomas Frey, Martin Bossert (2nd edition), page 14 https://books.google.de/books?id=xVa3Y85zBTYC&pg=PA14&dq=Antikausal&hl=de&sa=X&ved=0ahUKEwihhaDxnt_ZAhXD8RQVAR7R7R7 = onepage & q = anti-causal & f = false What amazes me, I always thought that systems are causal or something like that. I think it's almost pointless to relate this to signals. Correct me if I missed something here. Sincerely, Self-proclaimed do-gooders

Hi, thanks for your reply. I already know the definition. Only I would be interested in what that means in relation to the triangle function? Then this would have to be neither causal nor anti-causal, because it delivers function values ​​for both t <0 and t> 0.

of Self-proclaimed do-gooder (Guest)

Hello, I don't think anyone can give you a satisfactory answer to this question. Ultimately, systems are causal, i. H. only when a signal is present at the input of the system does the response to the signal appear at the output of the system. This can already be after the time t = 0. Looking at a single signal and claiming it is "causal", "anti-causal" or "not causal", I do not think it makes sense. With a coordinate transformation (t1 = t2 - DeltaT) you can get every finite signal "causal" or "anti-causal"! Sincerely, Self-proclaimed do-gooders

Since linear time-invariant (LTI) systems are fully characterized by their impulse response, the causality of an LTI system can be discussed equivalently through the causality of the impulse response. I only know "causal" and "not causal". The triangle function corresponds to a non-causal system because it is not zero for t <0.

of Self-proclaimed do-gooder (Guest)

Hello, Martin O. wrote:> The triangle function corresponds> to a non-causal system because it is not zero for t <0. if an LTI system has a triangular function as impulse response (impulse response), the system is "non-causal". The triangle function itself is neither "causal" nor "non-causal". Sincerely, Self-proclaimed do-gooders

Arcus wrote:> Is such a signal now neither causal nor anti-causal? In an anti-causal system, the current output signal and the internal current state depend exclusively on future input values. I would call it hypothetical describe.

### Important rules - read first, then post!

• Use upper and lower case
• Do not insert longer source code in the text, but as a file attachment