Abstract: A verification system and a verification method for a neural network accelerator hardware are provided. The verification system for a neural network accelerator hardware includes a neural network graph compiler and an execution performance estimator. The neural network graph compiler is configured to receive an assumed neural network graph and convert the assumed neural network graph into a suggested inference neural network graph according to a hardware information and an operation mode. The execution performance estimator is configured to receive the suggested inference neural network graph and calculate an estimated performance of the neural network accelerator hardware according to a hardware calculation abstract information of the suggested inference neural network graph.

Abstract: A chip-scale integrated multi-wavelength biological sensing device employing a plurality of filter-sensor assemblies is provided. The plurality of filter-sensor assemblies can include sufficient number of optical channels enabled by plasmonic filters having different nanoscale patterns and provided directly on an array of photodetectors. Combined with simultaneous illumination of light including multiple peak wavelengths with small full width at half maximum values, the independent optical channels of the plurality of filter-sensor assemblies enable correlation of detected optical signals with biological parameters. Signal processing methods for robustly extracting The PPG signals can be extracted by a robust signal processing method. Successful measurement of peripheral blood oxygen (SpO2) and blood pressure has been demonstrated.