GALS NoC architectures on FPGA dedicated to multispectral image applications

An efficient Network on Chip (NoC) implemented on Field Programmable Gate Arrays (FPGA) is proposed for the data communication of multispectral image analysis algorithms in an adaptive architecture. The architecture design is based on the linear effort property and reusable IPs. Multispectral image data are several types of data, different length values. According to the requirement of the multispectral image data communication, this NoC used Virtual Channels (VC) with packet-switching. Implementations on FPGA show that a NoC-based communication scheme provides a superior solution than classical point-to-point communication.