Automatic modulation recognition (AMR) has become increasingly important in the field of signal processing, especially with the advancements of intelligent communication systems. Deep Learning (DL) technologies have been incorporated into the AMR field and they have shown outstanding performances against conventional AMR methods. The robustness of DL-based AMR methods under varying noise regimes is one of major concerns for the widespread utilization of this technology. Furthermore, most existing works have neglected the contributions of hand-crafted features (HCFs) in boosting the classification performances of DL-based AMR methods. In order to address the aforementioned technical challenges, a novel and robust DL-AMR method is proposed by leveraging the benefits of both contextual features (CFs) and HCFs for a specific range of signal-to-noise ratio (SNR). A novel feature selection algorithm is also proposed to search for the optimal sets of HCFs in order to reduce the dimensions of feature vectors without losing any important and relevant features. Simulation studies are performed to investigate the feasibility of proposed method in classifying 11 types of modulation schemes. Extensive performance analyses revealed the superiority of proposed method over baseline method in terms of the classification performance as well as the excellent capability of proposed feature selection algorithm in determining an optimal subset of HCFs.INDEX TERMS Automatic modulation recognition (AMR), convolutional neural network (CNN), deep learning (DL), wireless signal classification.
Wafer defect inspection is one of the crucial semiconductor processing technologies because it can help to identify the surface defects in the process and eventually improve the yield. Manual inspection using human eye is subjective and long-term fatigue can lead to erroneous classification. Deep learning technology such as convolutional neural network (CNN) is a promising way to achieve automated wafer defect classification. The training of CNN is time consuming and it is nontrivial to fine tune its hyperparameters to achieve good classification performance. In this study, Arithmetic Optimization Algorithm (AOA) is proposed to optimize the CNN hyperparameters, such as momentum, initial learn rate, maximum epochs, L2 regularization, to reduce the burden brought by trial-and-error methods. The hyperparameters of a well-known pretrained model, i.e., GoogleNet, are optimized using AOA to perform wafer defects classification task. Simulation studies report that the AOA-optimized GoogleNet achieves promising accuracy of 91.32% in classifying wafer defects.
Feature selection is a crucial pre-processing step used to remove redundant information from original datasets while preserving the accuracy and processing time of classifier. The feasibility of using metaheuristic search algorithms (MSAs) such as Flow Directional Algorithm (FDA) to solve feature selection problems is one of the active research topics. Similar with other MSAs, FDA also employs conventional initialization scheme that generates initial solutions in random basis. The absence of intelligent mechanisms in conventional initialize scheme tends to generate initial populations in local optima, hence compromising the performance of algorithm to handle datasets with complex features. In this paper, a modified algorithm known as Multi Chaotic Flow Directional Algorithm (MCFDA) is proposed to solve feature selection problems with enhanced performances by leveraging the strengths of multiple chaotic maps for population initialization. A total of 12 datasets from UCI Machine Learning Repository are selected for performance evaluation of MCFDA and another four peer algorithms to solve feature selection problems. The proposed MCFFA is revealed to deliver best performances by solving 7 out of 12 datasets with the best mean classification accuracy and 6 out of 12 datasets with the least numbers of selected features.
Feature selection is a widely used technique to remove the undesirable, noisy and inaccurate information from raw input dataset while maintaining the accuracy and efficiency of classifier. Tremendous researches have explored the feasibility of metaheuristic search algorithms (MSAs) such as African Vultures Optimization Algorithm (AVOA) to solve feature selection problem. Similar with many original MSAs, the conventional initialization scheme of AVOA has undesirable drawbacks that can lead to entrapment of local optima, especially when dealing with complex dataset. In this paper, a new variant known as Chaotic African Vultures Optimization Algorithm (CAVOA) is proposed to solve feature selection problem with enhanced classification accuracy by incorporating the chaotic map concept into the initialization scheme. Twelve datasets obtained from UCI Machine Learning Repository are used to investigate the capability of CAVOA in feature selection and compared with four other peer algorithms. Simulation results show that CAVOA can produce the best classification accuracies and lowest feature numbers in most datasets.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.