Trends of Microwave Devices Design Based on Artificial Neural Networks: A Review

Abstract

The usage of techniques of the artificial neural networks (ANNs) in the field of microwave devices has recently increased. The advantages of ANNs in comparison with traditional full-wave methods are that the prediction speed when the traditional time-consuming iterative calculations are not required and also the complex mathematical model of the microwave device is no longer needed. Therefore, the design of microwave device could be repeated many times in real time. However, methods of artificial neural networks still lag behind traditional full-wave methods in terms of accuracy. The prediction accuracy depends on the structure of the selected neural network and also on the obtained dataset for the training of the network. Therefore, the paper presents a systematic review of the implementation of ANNs in the field of the design and analysis of microwave devices. The guidelines for the systematic literature review and the systematic mapping research procedure, as well as the Preferred Report Items for Systematic Reviews and Meta-Analysis statements (PRISMA) are used to conduct literature search and report the results. The goal of the paper is to summarize the application areas of usage of ANNs in the field of microwave devices, the type and structure of the used artificial neural networks, the type and size of the dataset, the interpolation and the augmentation of the training dataset, the training algorithm and training errors and also to discuss the future perspectives of the usage of ANNs in the field of microwave devices.

MSCNN-LSTM Model for Predicting Return Loss of the UHF Antenna in HF-UHF RFID Tag Antenna

“…Samson Daniel used ANN to optimize the performance of cavity-backed antenna loaded with slots for RFID applications [10]. However, there are comparatively few studies on ANN for EM performance prediction of RFID tag antennas, which have remarkable achievements in other antenna designs [11]. To mention a few, Sami Khafaga employed an improved LSTM to predict the bandwidth of a metamaterial antenna, and the predictions of the LSTM were superior compared to Multilayer Perceptron (MLP), K-Nearest Neighbors (KNN) and the basic LSTM [12].…”

Cost-Efficient Two-Level Modeling of Microwave Passives Using Feature-Based Surrogates and Domain Confinement

“…There exist two primary categories of surrogates: data-driven and physics-based models. The former come in many variations (e.g., kriging [48][49][50], neural networks [51][52][53][54][55], radial basis functions [56,57], and Gaussian process regression [58,59] to name just a few). They are exploited by global search procedures [60] and multi-criterial optimization [61] and frequently combined with sequential sampling routines [62,63].…”

Daugiasluoksnių puslaidininkinių ir dielektrinių mikrobangų įtaisų tyrimas

“…Analitiniai metodai yra greiti, tačiau taikant analitinius metodus kiekvienam naujos struktūros įtaisui reikėtų paruošti specifinį matematinį modelį, todėl procesas yra ilgas ir sudėtingas. Bet to, taikant analitinius metodus galima apskaičiuoti tik dalinius mikrobangų įtaisų atvejus (Katkevičius et al, 2022).…”

“…Praktikoje, kuriant naujus mikrobangų įtaisus, jiems modeliuoti naudojamos specializuotos komercinės programos, kurių skaičiavimai yra grįsti skaitiniais metodais. Sudėtingoms mikrobangų struktūroms tokie skaičiavimai gali trukti nuo kelių valandų iki kelių dienų (Marinkovic et al, 2016;Pomarnacki et al, 2014;Katkevičius et al, 2022). Procesas dar labiau išsitęsia, jeigu apytiksliai sistemos parametrai nėra žinomi ar nėra pakankamai tikslūs.…”

“…Dirbtiniai neuronų tinklai yra naudojami mikrobangų įtaisų charakteristikų analizei ir sintezei. Tyrimai atliekami antenų, antenų gardelių, mikrojuostelinių filtrų, fazės keitiklių, rezonatorių, vėlinimo linijų ir kitų mikrobangų įtaisų charakteristikų prognozavimo srityse (Katkevičius et al, 2022). Vienas iš pavyzdžių -dirbtinių neuronų tinklo naudojimas mikrobangų MEMS komponentų charakteristikų analizei.…”