High isolation integrated four-port MIMO Antenna for terahertz communication

Abstract

Future technology aims to develop wireless systems that offer improved data speed capabilities. A MIMO system presents a possible solution to address this problem. Traditional protocol-based design procedures lead to reduced product prototyping speed because they require extensive computational time. In this study, we develop an artificial neural network technique to optimize THz antenna slot dimensions, thus enhancing their performance level. The ANNs are trained with simulated data where the outputs are the performances of the antenna at the two resonant frequencies (reflection coefficients, bandwidth, gain), and the inputs are the dimensions of the slots and the operating frequencies. An ANN-optimized final dimension will coagulate as one antenna design from the multiband for future development into 2-port and 4-port MIMO antennas. To ensure robust predictions with minimal deviations from the simulation results, the ANN model is evaluated using multiple error metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Variance Accounted For (VAF), Mean Absolute Error (MAE), and R 2 . Then, the performance of the suggested MIMO antenna is being evaluated using parameters: isolation, gain, bandwidth, Envelope Correlation Coefficient, Diversity Gain and Total Active Reflection Coefficient. S11 < −15, ECC < 0.0001, a mutual coupling <-20, and diversity performance is high, thus we yield excellent agreement between ANN outputs with simulation results. Moreover, the range of TARC values at the operating frequencies (0.445 THz and 0.540 THz) verifies the effectiveness of the 4-port MIMO antenna for an advanced THz communication system, achieving a maximum diversity gain of 10 dB. This study develops a new THz antenna design, which reaches a broad bandwidth together with high gain performance, while keeping standard compact dimensions suitable for THz communication.

Topics & Concepts