CARMEN: CORDIC-Accelerated Resource-Efficient Multi-Precision Inference Engine for Deep Learning
For hardware designers of deep learning accelerators, this provides a flexible precision-scaling approach that improves efficiency without hardware redesign, though it is an incremental improvement over existing CORDIC-based designs.
CARMEN introduces a runtime-adaptive multi-precision inference engine using CORDIC-based MAC units that dynamically switch precision without hardware changes. ASIC implementation achieves 33% fewer cycles, 21% power savings, 4.83 TOPS/mm² density, and 11.67 TOPS/W efficiency; FPGA deployment shows 154.6 ms latency at 0.43 W for object detection.
This paper presents CARMEN, a runtime-adaptive, CORDIC-accelerated multi-precision vector engine for resource-efficient deep learning inference. The key insight is that CORDIC iteration depth directly governs computational accuracy, enabling dynamic switching between approximate and accurate execution modes without hardware modification. The architecture integrates a low-resource iterative CORDIC-based MAC unit with a time-multiplexed multi-activation function block, supporting flexible 8/16-bit precision and high hardware utilization. ASIC implementation in 28 nm CMOS achieves up to 33% reduction in computation cycles and 21% power savings per MAC stage; a 256-PE configuration delivers 4.83 TOPS/mm2 compute density and 11.67 TOPS/W energy efficiency. FPGA deployment on PynqZ2 validates 154.6 ms latency at 0.43 W for real-time object detection.