How to build a fast, custom FFT from C
Alan Coppola, OptNgn, and Brian Durwood, Impulse
EETimes (5/11/2011 7:58 AM EDT)
FPGAs are great vehicles for FFTs. You can get solid libraries from the manufacturer and be quickly on your way. For even faster performance, engineers are splicing their other processing code in-line with FFT libraries. This is particularly effective when a system is to be optimized for a specific frequency or profile.
Recently, C-based FPGA FFT libraries have been announced that open up more options, since working from C is a particularly natural way to describe and experiment with log, exponential, and angle iteration.
Customizing an FFT provides engineers and software developers the opportunity to embed part of their processing logic, in-line, in the actual transform. This type of digital signal processing is useful for scaling or for situations in which the coefficients might need to be tweaked or go through extensive iterations to determine the optimal configuration. Customizing FFTs like this used to require sophisticated VHDL or Verilog coding, but not anymore…
First, one asks, why bother to use C-to-FPGA to implement a custom FFT? Well, three common and one newish cases where a custom FFT makes sense are as follows:
To read the full article, click here
Related Semiconductor IP
- MACsec Engine optimized for 10Base-T1S/T1L applications
- Dual Port, Voltage Scaling Register File Compiler on GF22FDX+
- Single Port, Voltage Scaling, High Speed SRAM Memory Compiler on GF22FDX+
- I3C Full Controller / Target
- I3C HCI Host Controller
Related Articles
- From Principles to Practice: A Systematic Study of LLM Serving on Multi-core NPUs
- How to Design SmartNICs Using FPGAs to Increase Server Compute Capacity
- Retargeting IP -> Design system compiles silicon straight from C code
- How to transform video SerDes from a nightmare to a dream
Latest Articles
- PVTSizing 2.0: A Fast and Reliable Analog Circuit Sizing Operator for Agentic EDA Workflows
- An Agentic Python Framework for LLM-Driven UVM Testbench Synthesis with Closed-Loop Compiler Feedback
- Reconfiguring System-Level Protection Through Bounded Delegation in RISC-V MCUs
- SkipLane: Write-Once, Conflict-Free Accumulation for Zero-Skipping CNN Engines on FPGAs
- Priority-Aware Deficit Round Robin for AXI4 Bandwidth Fairness in ASAP-7nm Process