1
$\begingroup$

I have just started my work in compressive sensing. the measurement vectors are obtain by multiplying the sensing matrix with input signal. the thing i cant figure out adaptive compressive sampling.

can anyone please explain it as practical ?

thank you

$\endgroup$
1
$\begingroup$

In "ordinary" compressive sensing you take measurements $y = Ax$ and try to reconstruct $x$ from $y$ where $A$ - of dimension $(M \times N)$ - has been generated with no knowledge of x, except that it is sparse. $A$ is typically generated randomly.

In adaptive compressed sensing, you in principle take each individual measurement $y_m,\ m = 0, 1,\ldots M$, trying to adapt you next measurement $y_m$ cleverly so you get the best possible measurement for reconstruction of $x$, given what you have already observed in $y_0, y_1,\ldots, y_{m-1}$. Effectively, this means determining the rows $A_m$ of $A$ one at a time, taking into account the measurements that previous rows $A_0, A_1,\ldots, A_{m-1}$ gave you.

One example of adaptive compressive sensing could be Malloy & Nowak, 2013.

$\endgroup$
  • $\begingroup$ Thank you for your answer .. that means we will select the measurements based on some threshold? could please provide a trail matlab code? $\endgroup$ – Hebah May 21 '18 at 3:34
  • $\begingroup$ Unfortunately, I do not have anything lying around. I suggest you do some literature search and try to identify examples where the authors published Matlab code of their experiments. $\endgroup$ – Thomas Arildsen May 22 '18 at 7:44

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.