different implementations. The first implementation uses a 3×3 network-on-chip and is validated on the DE4 development board, which uses Altera's Stratix IV GX FPGA chip. The second implementation uses a 2×2 network-on-chip and is validated on the Cyclone V SoC FPGA, which is a smaller FPGA chip available on the DE1-SoC board. Both implementations will be released to the public domain with the hope that they will foster further research, in addition to facilitating replication and comparison to our results. SECTION I.