Design and Performance Evaluation of a Graphene Biosensor for Protein Detection with Two, Three Bit Encoding and Machine Learning Optimization
摘要
The increasing demand for advanced disease diagnostic solutions, particularly in developing regions, underscores the need for innovative biosensing technologies. This study introduces a protein detection biosensor that integrates graphene-based metasurfaces operating within the terahertz frequency range. The design exploits the unique optical properties of graphene to enhance detection capabilities through plasmonic surface effects. Comprehensive numerical simulations, conducted using COMSOL Multiphysics, facilitated the optimization of the sensor's electromagnetic behaviour and key design parameters, including graphene's chemical potential and resonator geometries. The results demonstrated exceptional sensing performance, with detection thresholds as low as 0.259 RIU and a sensitivity of 339 GHzRIU−1. The sensor exhibits dual-band operation, offering superior performance in the higher frequency range. In addition to its primary sensing function, the system showcases digital signal processing capabilities, operating as both 2-bit and 3-bit encoding systems. Polynomial regression analysis demonstrates remarkable performance reducing simulation time and resources.