No abstract

This paper presents two microstrip reflectarray designs based on variable size unit cells for 10 GHz. One design uses a 3-layer unit cell with polygon shaped patch and the other uses a unit cell with 1-layer square loop patch. Both arrays have a size of 10λ × 10λ at 10 GHz, can reflect the incoming wave as a high gain pencil beam in the desired direction of θ = 30°. Gains at 10 GHz are 23.6 dB and 26.1 dB for the 3-layer and one-layer structures, respectively. The multi-layer structure resulted in a wider gain banwdith. Simulation results show that the 3-dB gain bandwdith is about 22% for the 3-layer structure reflectarray, as compared to 12% for the one-layer structure.

This paper presents microstrip reflectarray design using variable size unit cell approach for 26 GHz, and investigates the effect of feed antenna distance to the surface. The unit cell structure is made up of a 1.5 mm foam layer (εr=1.37) and a 0.787 mm substrate layer (εr=2.2) between the ground and reflective layers. Two arrays of 14λx14λ in size were designed to reflect the incoming wave in the θ=25º direction; the distance to the feed was 160 mm in one design and 320 mm in the other design. Simulation results show gain of 30.2 dBi for the former and 29.8 dBi for the latter. 3-dB gain bandwidths were 11 % and 14 %, respectively

Purpose: This paper compares five supervised learning algorithms (support vector machines, k-nearest neighbor, decision tree, random forest, and AdaBoost) for predicting heart disease and examines the impact of normalization and GridSearch hyper-parameter tuning on model performance. Methods: The study utilizes the Cleveland database from the University of California-Irvine (UCI) repository, comprising data on 918 instances of heart disease patients with 12 attributes. Eleven attributes serve as predictors, while one attribute represents the target class. Models are built and tested using this dataset. Results: Comparing the algorithm performances with existing literature, accuracies range from 89.13–91.85%. AdaBoost exhibits the highest performance, whereas the decision tree performs the least effectively. Results surpass those reported in the literature. Normalization improves prediction performance by 17% for Support Vector Machines (SVM) and 14% for k-nearest neighbor (kNN). SVM does not benefit from GridSearch, while GridSearch enhances the decision tree and AdaBoost by 7% and 4% respectively. Normalization combined with GridSearch improves kNN and random forest by 2–3%. Conclusion: This study compares supervised learning algorithms for heart disease prediction. AdaBoost emerges as the top-performing algorithm, while the decision tree performs relatively poorly. The findings surpass those in the literature. Normalization significantly improves performance for SVM and kNN, while GridSearch enhances the decision tree and AdaBoost. Combined, normalization and GridSearch yield performance improvements for kNN and random forest. These results contribute to the field of heart disease prediction, offering valuable insights for algorithm selection and guiding future research.

No abstract