MSP_performance

In this paper, a bidirectional machine learning (ML)-based framework is introduced to predict the performance parameters such as S11, VSWR, and radiation pattern in two planes (XZ and YZ), radiation efficiency, and total efficiency, as well as to estimate the design variables including the patch and feed line dimensions (L, W, L1, and W1) and the dielectric constant (\u03b5r ) of a quarter-wave transformer (QWT)-microstrip patch (MSP) antenna at an operating frequency of 3 GHz. A Forward Artificial Neural Network (F-ANN) is developed to predict performance parameters based on given design variables. At the same time, an Inverse Artificial Neural Network (I-ANN) is implemented to estimate design variables from the desired performance parameter. This technique involves the designing, training, and optimization of an MSP antenna in real time. The F-ANN architecture is built with five input features and one output, whereas the I-ANN is implemented with three input features and five outputs. Model evaluation is done using the R-squared (R2) and Mean Squared Error (MSE). The F-ANN and I-ANN achieve R2 values of 0.99 and minimal MSE values of 4 \u00d710-4 and 4 \u00d710-3 respectively. The estimated design variables through inverse modeling are further re-verified using retro simulation and compared with simulated and predicted results; also, the developed models S11 and VSWR parameters are validated using measured performance. Both F-ANN and I-ANN predict and estimate parameters with higher accuracy.