Range, Not Precision: Block-Floating-Point Half-Precision FFT and SAR Imaging on Apple Silicon
For radar signal processing practitioners, this work overturns the assumption that FP16 is unsuitable for SAR, showing that range management rather than mantissa precision is the key bottleneck.
The authors demonstrate that FP16 half-precision can be used for radar-grade FFT and SAR imaging on Apple Silicon by addressing dynamic range overflow with a block-floating-point schedule, achieving 2.2× speedup over FP32 with less than 0.1 dB quality loss and 42 dB SQNR.
Half precision (FP16) promises to double FFT throughput on GPUs, but the prevailing view is that its 10-bit mantissa makes it unsuitable for radar-grade signal processing. We show this framing is wrong on Apple Silicon: the binding constraint for FFT and Synthetic Aperture Radar (SAR) is not mantissa \emph{precision} but the 5-bit exponent's \emph{dynamic range}. We first measure that an FP16 FFT is mantissa-limited at 56--61~dB signal-to-quantization-noise ratio (SQNR) -- comfortably radar-usable -- yet a naïve FP16 SAR pipeline produces \emph{only} \texttt{NaN}, because the conjugate--FFT--conjugate inverse transform grows magnitudes by a factor of $N$, and the matched-filter product ($\sim\!5\times10^6$ at $N\!=\!4096$) overflows FP16's 65{,}504 ceiling. We resolve this with a fixed-shift \emph{block-floating-point} (BFP) schedule: a single $1/N$ scale applied before each inverse transform bounds every intermediate below 4096. A cascade follows: range-compression output becomes $O(1)$ instead of $O(N)$, which in turn keeps the downstream azimuth-FFT output FP16-loadable instead of overflowing at $O(N^2)$. The result is the first quality-preserving FP16 SAR pipeline: peak/integrated sidelobe ratios, target SNR, and resolution match the FP32 reference to within $0.1$~dB at $42$~dB end-to-end SQNR, while a radix-8 FP16 FFT reaches 306~GFLOPS -- $2.2\times$ over the 139~GFLOPS FP32 baseline -- on a fanless Apple~M1. Finally, we measure that FP8 (E4M3/E5M2) collapses to 14--20~dB SQNR, making FP16 \emph{today's} precision floor for FFT-based radar -- one that future precision-recovery methods may yet lower -- and showing that the lever for low precision here is range management, not mantissa bits.