Nilesh Kumar Tiwari

Computationally Intelligent Sensor System for Microwave Characterization of Dielectric Sheets

An automated RF sensor system for accurate determination of complex permittivity of thin samples and dielectric sheets is proposed. The proposed system is computationally intelligent making use of the machine learning algorithm along with the artificial neural network (ANN) architecture and employs a coplanar waveguide sensor for the measurement of scattering coefficients of test specimens in order to obtain their dielectric properties. Different heuristics are followed for the design of the ANN-based system. The applicability of the proposed sensor system for practical usage is increased by taking into account the effect of possible air gap between the device and test specimen. This is facilitated by developing a multilayered analytical model, and combining it with the ANN-based system. The complete procedure, comprising of ANN algorithms and the analytical formulation, is implemented in MATLAB. The proposed automated standalone sensor system is validated by testing a number of standard samples in the designated frequency bands.

Determination of Effective Constitutive Properties of Metal Powders at 2.45 GHz for Microwave Processing Applications

Abstract A reflection-transmission based rectangular waveguide approach for measuring the effective constitutive properties of metal powders at 2.45 GHz is presented. The measured effective dielectric properties of these metal powders are quite important for sintering of metal powders using microwaves, which is a new area of research. The proposed method is based on placing the metal powders of varying density into a standard glass tube, which is then positioned inside a specially designed waveguide holder for measuring the scattering coefficients at 2.45 GHz using a vector network analyser. The effective constitutive properties of the metal powders are obtained in terms of the measured scattering coefficients using the proposed two step approach. In order to validate the accuracy of the measured material properties, an electromagnetic core-shell model of the metal powders is developed. Finally, the effective dielectric properties of Cu and stainless steel powders are measured and compared with the values obtained using the electromagnetic model.

Microwave planar resonant sensor for detection of oil spills

In this work, a simple microwave resonant sensor is proposed for detection of small amount of oil spill in sea water. The sensor is realized using the microstrip technology, and the geometry of the resonant structure is optimized in order to improve the sensitivity. The proposed sensor is designed to operate at 5.85 GHz in the industrial, scientific, and medical (ISM) band. The structure of the sensor is modeled in full wave electromagnetic solver, CST Microwave Studio and the dimensions of the sensor are chosen to get the desired operating frequency. The designed sensor is finally fabricated, and is employed for detection of small amount of petroleum products present in the water. The measured results show that the proposed sensor is capable of detecting up to 5 % of petrol in water.

Quasi-optimization approach for millimeter wave imaging of multilayered dielectric media

ABSTRACT In this paper, a quasi-optimization approach involving a two-step method for the millimeter wave imaging of the stratified media is proposed. The proposed method makes use of reflection data of the test media in the mm-wave band of 26-40 GHz, whose equivalent time domain data are also available using the standard IFFT routines. In the first step, a simple analytical method is applied to extract the real permittivity and the physical thickness of each layer from time domain reflection data. In the second step, an efficient optimization method is proposed to extract material parameters of each layer more accurately from the spectral domain reflection coefficients. The extracted real permittivity and thickness of each layer from the first step provide a quite good estimated value for the next optimization process and thus overall computation time is considerably reduced. The proposed method is experimentally tested for a number of multilayered media comprising various materials.

Near field planar microwave probe sensor for nondestructive condition assessment of wood products

In this work, the unified methodology based on the newly designed electrically small planar resonant microwave sensor to detect the subsurface defect in wood products is presented. The proposed planar sensor involves loading of the specially designed coupled microstrip line with a novel small resonating element at its end. The novel design topology of the proposed near field sensor substantially increases the overall resolution and sensitivity of the microwave scanning system due to the strong localization of the electric field in the electrically small sensing region. A detailed electromagnetic and quasi static analysis of the near field scanning mechanism is also described in this work, which helps to understand the physics involved in the proposed scanning mechanism. The prototype of the designed sensor is fabricated on a 0.8 mm Roger 5880 substrate, and accordingly, the scattering parameters of the sensor under both loaded and unloaded conditions are measured. The measured and simulated scattering parameters under the unloaded condition are compared to validate the fabricated sensor, and a closed match between the simulated and measured resonance frequencies is observed. The fabricated sensor is used here for two potential applications, viz., the dielectric sensing of various low permittivity contrast dielectric materials and subsurface imaging of wood products to trace concealed defects and moisture content under the thin paint layer. The proposed resonant sensor can potentially be used to develop the low profile, low cost, non-destructive, and non-invasive quality monitoring system for inspecting various types of wood products without peeling off the upper paint coating.In this work, the unified methodology based on the newly designed electrically small planar resonant microwave sensor to detect the subsurface defect in wood products is presented. The proposed planar sensor involves loading of the specially designed coupled microstrip line with a novel small resonating element at its end. The novel design topology of the proposed near field sensor substantially increases the overall resolution and sensitivity of the microwave scanning system due to the strong localization of the electric field in the electrically small sensing region. A detailed electromagnetic and quasi static analysis of the near field scanning mechanism is also described in this work, which helps to understand the physics involved in the proposed scanning mechanism. The prototype of the designed sensor is fabricated on a 0.8 mm Roger 5880 substrate, and accordingly, the scattering parameters of the sensor under both loaded and unloaded conditions are measured. The measured and simulated scattering par...

Nondestructive Extraction of Parameters of Multilayered Media Using Terahertz Pulse Technique

Abstract In this work, an efficient non-invasive terahertz pulse technique is proposed and investigated to determine the thickness and refractive index of each layer in an optically thick stratified media. A closed form formulations are derived for simultaneous extraction of the thickness and complex refractive index of each layer with the help of primary reflected signals from the multilayered structure. The proposed technique is numerically tested using a full wave electromagnetic simulator and is experimental verified in the millimeter wave frequency range by utilizing the power peaks corresponding to the primary reflected signals. The numerical and measured results of multilayered samples under test are in good agreement with the reference data. The proposed terahertz pulse technique can be used for non-destructive testing of the multilayered system existing in various industries.

Novel spoof plasmonic based compact slow wave structure for terahertz and microwave applications

A novel compact slow wave structure (SWS) based on spoof plasmonic waveguide with meander lines in the terahertz (THz) and microwave frequency range is proposed. The use of the meander lines increases the effective length of the proposed structure thereby providing more path length for efficient trapping of the spoof surface plasmon polaritrons (SPPs). The meander line based SPPs structure is designed in THz and microwave frequency range using the full wave electromagnetic field simulator, the CST Studio. The area of proposed SWS is found to be reduced by 75% as compared to that of the conventional rectangular based surface plasmonic waveguide structure. The experimental verification of the proposed SWS is carried out in the microwave frequency band by fabricating the scaled up version, and measuring the transmission coefficient data using the network analyzer.