Encoding and Tuning of THz Metasurface-Based Refractive Index Sensor With Behavior Prediction Using XGBoost Regressor

Abstract

We have investigated graphene-based three various refractive index sensors (split ring resonator (SRR), split ring resonator with thin wire (SRRTW), and thin wire (TW) refractive index sensors) for the encoding and sensing-based applications. The sensors are designed to detect the presence of hemoglobin biomolecules with high sensitivity. The results are analyzed in the form of transmittance, and electric field, and detailed sensitivity analysis is also carried out for the proposed graphene-based refractive index sensors for four various concentrations of hemoglobin biomolecules. We have also investigated the sensor's performance in terms of quality factor, Q, and figure of merit (FOM). The encoding of '0' and '1' is attained by varying the graphene chemical potential fulfilling the one-digit coding. An array of these sensors can then be used for encoding-based applications. The detailed analysis of reported sensors is also carried out by checking the effect of varying physical parameters such as substrate thickness, split ring gap, and thin wire width on tunability. These sensors can be applied in biomedical or encoding-based applications. Experiments are performed using XGBoost regressor to determine, whether simulation time and resources can be reduced by using regression analysis to predict the transmittance values of intermediate frequency or not. Experimental results prove that regression analysis using XGBoost Regressor can reduce the simulation time and resources by at least 70 percent.

Optimization of Metamaterial‐Based Solar Energy Absorber for Enhancing Solar Thermal Energy Conversion Using Artificial Intelligence

“…[36][37][38] The novel application of machine learning in the field of photonics devices is reported by. [39][40][41] We still believe that the absorption rate and corresponding bandwidth can be enhanced and we have implemented a titanium-based solar energy absorber design that can achieve a higher absorption rate for enhanced bandwidth. Zheng et al proposed a simple, perfect, and polarizationinsensitive absorber for ultrasensitive sensing applications that achieved a triple-band absorption response with the highest sensitivity of 1194 nm/RIU.…”

“…[ 36–38 ] The novel application of machine learning in the field of photonics devices is reported by. [ 39–41 ] We still believe that the absorption rate and corresponding bandwidth can be enhanced and we have implemented a titanium‐based solar energy absorber design that can achieve a higher absorption rate for enhanced bandwidth. Zheng et al.…”

Machine learning assisted metamaterial-based reconfigurable antenna for low-cost portable electronic devices

“…Machine learning (ML) can be applied to predict or forecast the behavior of different parameters of electronic and photonic systems. ML can also be used to observe, predict, and forecast the reflectance response of the photonic devices 31 – 34 . Misilmani reviewed ML applications for antenna designs utilizing regression models 35 .…”

Graphene‐based tunable terahertz electromagnetically induced transparency using metamaterial structure

“…Recently, metamaterials have received tremendous attention since they can manipulate the interaction between light and matter in artificial structures, making it possible to realize the analogue EIT effect in classical optical systems. [4][5][6][7][8][9][10][11][12][13][14] Moreover, a classical method was proposed to realize the analogue EIT phenomenon through the coupling between the resonators of bright and dark modes, and the strength of the analogue EIT effect could be adjusted by controlling the near field coupling and resonant loss of the two modes. 15,16 However, for most previous metamaterialbased EIT studies, the operating frequency of the transparent window can be passively adjusted, that is, the frequency of the transparent window can only be tuned by changing the internal EIT-like structural parameters, such as the geometric size and coupling distance, which severely limits the application of EIT-like in tunable slow light devices.…”

“…Recently, metamaterials have received tremendous attention since they can manipulate the interaction between light and matter in artificial structures, making it possible to realize the analogue EIT effect in classical optical systems 4–14 . Moreover, a classical method was proposed to realize the analogue EIT phenomenon through the coupling between the resonators of bright and dark modes, and the strength of the analogue EIT effect could be adjusted by controlling the near field coupling and resonant loss of the two modes 15,16 .…”