Summary

In this 2013 blog, Kyle introduces polyphase filtering as a computationally efficient structure for applying resampling and filtering, and shows that most digital filters can be expressed in polyphase form. The article explains practical implementations for decimation, interpolation, and FFT-based filterbanks, and highlights complexity savings and real-time implementation considerations.

Key Takeaways

• Describe polyphase decomposition and how to rewrite FIR filters in polyphase form.
• Implement decimation and interpolation using polyphase structures to reduce computation.
• Design and analyze analysis/synthesis and FFT-based polyphase filterbanks for near-perfect reconstruction.
• Quantify computational savings and trade-offs compared to direct filtering or FFT-only approaches.
• Adapt polyphase implementations for real-time and fixed-point hardware constraints.

Who Should Read This

Intermediate DSP engineers, researchers, and graduate students working on resampling, filterbank design, or efficient multirate implementations who want practical theory and implementation guidance.

TimelessIntermediate

Topics