Could this function be approximated by Dirac delta function?

Click For Summary
SUMMARY

The discussion centers on approximating the function $$N(k)=\frac {-1} {|k|} e^{-c|k|}$$ with the Dirac delta function $$\delta(k)$$ for frequency analysis. While $$N(k)$$ exhibits the desired behavior of becoming very large as $$k$$ approaches zero and decreasing rapidly for non-zero $$k$$, it fails to satisfy the integral condition $$\int_{-\infty}^{+\infty} N(k) dk = 1$$, which is crucial for delta function approximation. The proposed modification involves redefining $$N(k)$$ as $$\tilde{N}(k)= e^{-c|k|}/\sqrt{|k|}$$ and adjusting $$P(k)$$ accordingly to facilitate integration.

PREREQUISITES
  • Understanding of frequency analysis in signal processing
  • Familiarity with the properties of the Dirac delta function
  • Knowledge of integration techniques in mathematical physics
  • Experience with exponential decay functions
NEXT STEPS
  • Research the properties and applications of the Dirac delta function in physics
  • Learn about the implications of modifying functions in frequency space
  • Explore integration techniques for singular functions in mathematical analysis
  • Investigate the behavior of exponential functions as parameters approach infinity
USEFUL FOR

Physicists, mathematicians, and engineers involved in frequency analysis and those interested in the mathematical foundations of signal processing.

Haorong Wu
Messages
419
Reaction score
90
TL;DR
Under what conditions, could a function be replaced by a delta function?
hi, there. I am doing some frequency analysis. Suppose I have a function defined in frequency space $$N(k)=\frac {-1} {|k|} e^{-c|k|}$$ where ##c## is some very large positive number, and another function in frequency space ##P(k)##. Now I need integrate them as $$ \int \frac {dk}{2 \pi} N(k) P(k).$$However, the integration wil be too complicated to be solved.

Meanwhile, I notice that, ##N(k)## is infinite when ##k\rightarrow 0##, and decreases to zero rapidly when ##k \ne 0##. However, the integral of $$ \int dk N(k)$$ is also infinite. I am not sure whether I could approximate ##N(k)## by ##\delta(k)## or not. Would it yield problems in physics?

Thanks!
 
Last edited:
Physics news on Phys.org
No. There are three facts that you want for a function that acts like a delta function are:

1. ##f(k) \rightarrow \text{very large}## when ##k \rightarrow 0##
2. ##f(k) \rightarrow 0## when ##k \rightarrow \infty##
3. ##\int_{-\infty}^{+\infty} f(k) dk = 1##

Your function has properties 1 and 2, but not 3.
 
You could shuffle some of N onto P. Like if you said ##\tilde{N}(k)= e^{-c|k|}/\sqrt{|k|}## and ##\tilde{P}(k)=P(k)/\sqrt{|k|}## you might be able to do something.

As c goes to infinity the integral of ##\tilde{N}## is not constant, so you would also need to shuffle around some ##c## stuff in order to get this to totally work.
 

Similar threads

  • · Replies 25 ·
Replies
25
Views
4K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 20 ·
Replies
20
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 32 ·
2
Replies
32
Views
5K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 9 ·
Replies
9
Views
4K