Floating- to fixed-point MATLAB algorithm conversion for FPGAs
By Tom Hill, Xilinx
June 04, 2007 -- dspdesignline.com
In a recent survey conducted by AccelChip Inc. (recently acquired by Xilinx), 53% of the respondents identified floating- to fixed-point conversion as the most difficult aspect of implementing an algorithm on an FPGA (Figure 1).
Figure 1. AccelChip DSP design challenges survey.
June 04, 2007 -- dspdesignline.com
In a recent survey conducted by AccelChip Inc. (recently acquired by Xilinx), 53% of the respondents identified floating- to fixed-point conversion as the most difficult aspect of implementing an algorithm on an FPGA (Figure 1).
Figure 1. AccelChip DSP design challenges survey.
Although MATLAB is a powerful algorithm development tool, many of its benefits are reduced during the fixed-point conversion process. For example, new mathematical errors are introduced into the algorithm because of the reduced precision of the fixed-point arithmetic. You must rewrite code to replace high-level functions and operators with low-level models that reflect the actual hardware macro-architecture. And simulation run times can be as much as 50 times longer. For these reasons, MATLAB, the overwhelming choice for algorithm development, is often abandoned in favor of C/C++ for fixed-point modeling.
To read the full article, click here
Related Semiconductor IP
- HBM4 PHY IP
- Ultra-Low-Power LPDDR3/LPDDR2/DDR3L Combo Subsystem
- MIPI D-PHY and FPD-Link (LVDS) Combinational Transmitter for TSMC 22nm ULP
- HBM4 Controller IP
- IPSEC AES-256-GCM (Standalone IPsec)
Related Articles
- From a Lossless (~1.5:1) Compression Algorithm for Llama2 7B Weights to Variable Precision, Variable Range, Compressed Numeric Data Types for CNNs and LLMs
- Procrastination Is All You Need: Exponent Indexed Accumulators for Floating Point, Posits and Logarithmic Numbers
- Growing demand for high-speed data in consumer devices gives rise to new generation of low-end FPGAs
- Implementing floating-point algorithms in FPGAs or ASICs
Latest Articles
- ElfCore: A 28nm Neural Processor Enabling Dynamic Structured Sparse Training and Online Self-Supervised Learning with Activity-Dependent Weight Update
- A 14ns-Latency 9Gb/s 0.44mm² 62pJ/b Short-Blocklength LDPC Decoder ASIC in 22FDX
- Pipeline Stage Resolved Timing Characterization of FPGA and ASIC Implementations of a RISC V Processor
- Lyra: A Hardware-Accelerated RISC-V Verification Framework with Generative Model-Based Processor Fuzzing
- Leveraging FPGAs for Homomorphic Matrix-Vector Multiplication in Oblivious Message Retrieval