Gibbs Phenomenon: Impact on kth Harmonics for Bandwidth Selection

  • Thread starter Thread starter goldfronts1
  • Start date Start date
  • Tags Tags
    Gibbs Phenomenon
AI Thread Summary
The Gibbs Phenomenon affects the selection of kth harmonics by causing poor convergence near discontinuities, resulting in persistent overshoot errors regardless of the number of harmonics used. As the number of harmonics increases, the waveform approximates the original square wave more closely in a mean square sense, but the maximum error near edges remains constant. Tapering the Fourier series with a windowing technique, such as a Hamming window, can significantly reduce ringing artifacts. The choice between using fewer harmonics for lower bandwidth versus more for better accuracy involves a trade-off, as both approaches still exhibit Gibbs Phenomenon effects. Ultimately, the decision on the optimal number of harmonics depends on the specific requirements for signal fidelity and bandwidth efficiency.
goldfronts1
Messages
31
Reaction score
0
How does the Gibbs Phenomenon effect the choice of the kth harmonics for bandwidth selection?

Basically, I have plotted a square wave using the Fourier series analysis for choosing the kth harmonics. As, I increase the kth harmonics the oscillations increase. What is the smallest harmonic i can go to that is still an accurate representation of the signal. Or does it not matter.

Confused

Thanks
 
Engineering news on Phys.org
As the number of harmonics (N) increases the resultant waveform becomes progressively closer (in the sense of mean square difference) to the original square wave. Gibbs phenomenon refers to the fact that, despite getting ever closer in a mean squared sense, the convergence is very poor near the discontinuity (the edge) and in fact the maximum error at the "ringing overshoot" doesn't actually get any smaller as you increase N.

Tapering (instead of abruptly truncating) the truncated Fourier series (also called "windowing") is a very useful method for reducing these ringing artefacts. For example you could taper off the series with a "Hamming" type of window using something like : w(k) = 0.54 + 0.46 \cos( (k-1)\pi/(N-1) )

In other words, instead of using \sin(2 \pi (2k-1) t)/(2k-1) for the k-th term in the series you would use w(k) \sin(2 \pi (2k-1) t)/(2k-1) instead. Give it a try, you'll be surprised what a big improvement it makes.
 
The attachment shows a 20 term (sine) Fourier series for a square wave, both with and without windowing being used. The window I used was w(k) as defined in the previous post.
 

Attachments

  • windowed_fourier.png
    windowed_fourier.png
    9.4 KB · Views: 490
Ok, so what's the difference if I use 60 terms, which gives me about 99% of the power in the signal versus using 20 terms which gives me about 95% of the power in the signal? When the ultimate goal is to have the smallest bandwidth. I guess the gibbs phenomenon is still present in either case, but the signal is closer to the original when you use more k-terms, but that uses up more bandwidth. Is this just a judgment call? I guess which one is best?
 
Last edited:
Very basic question. Consider a 3-terminal device with terminals say A,B,C. Kirchhoff Current Law (KCL) and Kirchhoff Voltage Law (KVL) establish two relationships between the 3 currents entering the terminals and the 3 terminal's voltage pairs respectively. So we have 2 equations in 6 unknowns. To proceed further we need two more (independent) equations in order to solve the circuit the 3-terminal device is connected to (basically one treats such a device as an unbalanced two-port...
suppose you have two capacitors with a 0.1 Farad value and 12 VDC rating. label these as A and B. label the terminals of each as 1 and 2. you also have a voltmeter with a 40 volt linear range for DC. you also have a 9 volt DC power supply fed by mains. you charge each capacitor to 9 volts with terminal 1 being - (negative) and terminal 2 being + (positive). you connect the voltmeter to terminal A2 and to terminal B1. does it read any voltage? can - of one capacitor discharge + of the...
Thread 'Weird near-field phenomenon I get in my EM simulation'
I recently made a basic simulation of wire antennas and I am not sure if the near field in my simulation is modeled correctly. One of the things that worry me is the fact that sometimes I see in my simulation "movements" in the near field that seems to be faster than the speed of wave propagation I defined (the speed of light in the simulation). Specifically I see "nodes" of low amplitude in the E field that are quickly "emitted" from the antenna and then slow down as they approach the far...
Back
Top