HF-UHF Dual-band RFID Tag Antenna Designs Based On Intelligence Algorithms

Radio Frequency Identification(RFID),as an important part of the Internet of Things sensing layer technology,has been widely used in recent years.Among them,high-frequency-Ultrahigh-frequency(HF-UHF)dual-band RFID tags are favored for their ability to both near-field communication and long-range group reading and writing.However,the design of dual-band RFID tag antennas solely using electromagnetic simulators such as High Frequency Structure Simulator(HFSS)is time-consuming and labor-intensive and depends on professional electromagnetic knowledge.Intelligent algorithms can provide low-cost solutions for complex problems.Therefore,this paper studies the design methods of HF-UHF dual-band RFID tag antenna based on intelligent algorithms.Three design methods of the antenna are proposed,which include HFSS-assisted multi-population genetic algorithm(MGA-(HFSS)),multi-scale convolutional neural network stacked with long and short-term memory(MSCNN-LSTM),and their combination MGA-(MSCNN-LSTM).In the method of designing dual-band RFID tag antennas based on MGA-(HFSS),this paper uses an improved Wheeler formula to design the HF antenna,and the new dipole antenna with the added inductive ring structure as the UHF antenna to complete the initial structural design of the dual-band tag antenna.The proposed MGA-(HFSS)is used to optimize the dimensions of the tag antenna.Simulation results show that the optimized UHF antenna has the input impedance of(16.75+j349.9Ω)at 915MHz,which is close to the ideal value of(22.5+j349Ω),achieving good impedance matching with the chip.The size of the optimized dual-band tag antenna is 42×42×0.075mm~3,with a maximum gain of 1.4d B and a maximum read distance of 7.17m.Compared with existing dual-band tag antennas,it has significant advantages in miniaturization.In the design method of dual-band RFID tag antenna based on MSCNN-LSTM,the MSCNN-LSTM model is proposed to predict the reflection coefficient of UHF antennas instead of HFSS for the problem of long simulation time for calculating the performance of UHF RFID antennas.In the proposed MSCNN-LSTM,the MSCNN has three branches,which include three convolution layers with different kernel sizes and numbers.Therefore,MSCNN can extract fine-grain localized information of the antenna and overall features.The LSTM can effectively learn the electromagnetic characteristics of different structures of the antenna to improve the prediction accuracy of the model.Experimental results show that the mean absolute error(0.0073),mean square error(0.00032),and root mean square error(0.01814)of the MSCNN-LSTM are better than those of other prediction methods.In predicting the reflection coefficient of100 UHF antennas,compared with the simulation time of 4800 seconds for HFSS,MSCNN-LSTM takes only 0.927519 seconds under the premise of ensuring prediction accuracy,significantly reducing the calculation time,which provides a basis for the rapid design of HF-UHF RFID tag antenna.Then MSCNN-LSTM is used to determine the dimensions of the UHF antenna quickly.The reflection coefficient of the designed dual-band RFID tag antenna is-58.76d B and-22.63d B at 13.56 MHz and 915 MHz,respectively,achieving the desired goal.Finally,the paper combines MGA with MSCNN-LSTM,and uses MSCNN-LSTM to calculate the fitness of each individual in the MGA optimization process,i.e.,the reflection coefficient of the antenna,and only after selecting the optimal individual,HFSS is used to determine if it meets the design requirements,greatly reducing the computational cost of antenna size optimization and achieving rapid intelligent design of the dual-band RFID tag antenna.The research work presented in this paper provides feasible methods for the rapid intelligent design of HF-UHF dual-band RFID tag antennas,which has important practical significance and theoretical value.