Signals & systems

poles and zeros

Poles and zeros are the handful of special complex-frequency points that completely define a transfer function — and reading them is like reading a system's palm. Zeros are values of s that drive the output to zero (frequencies the system blocks); poles are values of s that drive the output to infinity (the system's natural resonances, where it wants to oscillate). Plot them as ✕'s (poles) and ○'s (zeros) on the complex s-plane and you have a map of the system's entire personality.

The geometry is wonderfully predictive. A pole's distance from the imaginary axis sets how fast its part of the response decays (closer to the axis = slower, more sluggish); a pole's height up the axis sets the frequency at which it rings. Poles in the left half-plane mean a stable system that settles; a pole crossing into the right half-plane means instability — the output explodes. This single picture lets engineers design filters and tune control loops by literally dragging poles and zeros to new locations.

Reading decay rate and ringing frequency off a pole's coordinates

For discrete-time systems (the z-transform) the stability boundary isn't the imaginary axis but the unit circle: poles must lie inside |z| = 1. A pole and a zero sitting on top of each other cancel, which is how you can simplify a transfer function.

Also called
pole-zero plot極零點