Reference platforms

To generalize the assessment of the project outcomes to the wide landscape of emerging low energy customized and heterogeneous platforms, ALOHA will be tested considering two main platforms as reference:

a) STMicroelectronics Orlando: a low-power IoT end-nodes integrating specialized hardware blocks for specific compute-intensive data processing

G. Desoli/et al/., "14.1 A 2.9TOPS/W deep convolutional neural network SoC in FD-SOI 28nm for intelligent embedded systems", 2017 IEEE International Solid-State Circuits Conference (ISSCC), San Francisco, CA, 2017, pp. 238-239. doi: 10.1109/ISSCC.2017.7870349

b) NEURAghe: a zynq-based heterogeneous architecture accelerating convolutional neural networks

P. Meloni, A. Capotondi, G. Deriu, M. Brian, F. Conti, D. Rossi, L. Raffo, L. Benini, "NEURAghe: Exploiting CPU-FPGA Synergies for Efficient and Flexible CNN Inference Acceleration on Zynq SoCs", 2017, https://arxiv.org/abs/1712.00994