The answer is:

$\cos(1250\pi t)$ and $\cos(21250\pi t)$

I’m not able to verify.

My understanding is that when you periodically increase $t$ by $1/10$ ($10$ here is the frequency, so $1/10$ is time period) from $0$ to the signal $\cos(1250\pi t)$, you should get $x(n)=\cos(n\pi /8)$.

However, $\cos(1250\pi (1/10))$ is not same as $\cos((1)\pi /8)$.

So I would like the verification and what's wrong with my reasoning? Any help is appreciated.

  • $\begingroup$ These are two of infinitely many signals that produce the same sequence! The effect you're looking for is aliasing. $\endgroup$ – Marcus Müller May 9 '17 at 8:45

Considering the aliasing, here's the simple derivation: Given a cosine-signal $x(t)$ by

$$x(t)=\cos(2\pi f t)$$

which we sample with sampling frequency $F$, i.e. $t=n/F$, we get

$$x[n]=\cos(2\pi f/F n)$$

and if we now factor $f=kF+f_0$ with $k\in\mathbb{N}$ you'll see that

$$x[n]=\cos(2\pi (kF+f_0)/F n)=\cos(2\pi f_0/F n)$$

For an illustration of aliasing, you can have a look at one of my articles on aliasing.

So, cosines that are an integer multiple apart in frequency, yield the same sampled signal.

Now, to have

$$cos(2\pi f_0/F n)=\cos(\pi n/8)$$

You need $2\pi f_0/F = \pi/8$ and hence $f_0 = F/16$. Hence, your proposed solutions are not correct in my opinion, because $1250\pi t=2\pi 625=2\pi(62*F+5)$.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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