Narrow Results

One of the important parameters in the brain–computer interface (BCI) system is speed. Therefore, it is always desirable to design a high-speed system that has an acceptable performance, simultaneously. The main idea of this paper is the use of evolutionary algorithms (EAs) to select the optimal features for epilepsy diagnosis by processing the electroencephalogram (EEG) signals. The lesser the number of features is, the higher will be the usefulness of accuracy of the system to us. Therefore, here, using EAs, some of the features that are redundant in the data and do not contain a lot of information and only increase the complexity of the system are eliminated and the best features are chosen. We select this choice by EAs. Running the feature selection step is after the feature extraction step. In fact, the features were extracted using the common spatial pattern (CSP) algorithm, and then the optimal features were selected from the extracted feature set. This can save a lot of system complexity and reduce system execution time considerably. Finally, at the diagnostic stage, these selected features are given to a simple neural network (NN). The results showed that when the combination of EA and CSP is used, the precision of the system is much higher than when the CSP method is only used, although it contributes significantly to the complexity of the system.

Inertial navigation system (INS) is often integrated with satellite navigation systems to achieve the required precision at high-speed applications. In global navigation system (GPS)/INS integration systems, GPS outages are unavoidable and a severe challenge. Moreover, because of the usage of low-cost microelectromechanical sensors (MEMS) with noisy outputs, the INS will get diverged during GPS outages, and that is why navigation precision severely decreases in commercial applications. In this paper, we improve GPS/INS integration system during GPS outages using extended Kalman filter (EKF) and artificial intelligence (AI) together. In this integration algorithm, the AI receives the angular rates and specific forces from the inertial measurement unit (IMU) and velocity from the INS at $t$ and $t-1$. Therefore, the AI has positioning and timing data of the INS. While the GPS signals are available, the output of the AI is compared with the GPS increment; so that the AI is trained. During GPS outages, the AI will practically play the GPS role. Thus, it can prevent the divergence of the GPS/INS integration system in GPS-denied environments. Furthermore, we utilize neural networks (NNs) as an AI module in five different types: multi-layer perceptron (MLP) NN, radial basis function (RBF) NN, wavelet NN, support vector regression (SVR) and adaptive neuro-fuzzy inference system (ANFIS). To evaluate the proposed approach, we utilize a real dataset that has been gathered by a mini-airplane. The results demonstrate that the proposed approach outperforms the INS and GPS/INS integration systems with the EKF during GPS outages. Meanwhile, the ANFIS also reached more than 47.77% precision compared to the traditional method.

The brain–computer interface (BCI) database’s motor assessment depends heavily on the motor imagery (MI) signal classification. By examining the multiple patterns of different creative tasks in the electroencephalogram (EEG) signals, the intention of humans is translated into computer-based commands in the MI-based signals of BCI data. Nevertheless, low accuracy and efficiency are issues with MI–EEG signals’ classification because of the low signal-to-noise ratio, huge individual differences, overall volatility, and complexity in the signal. To overcome these problems, this research proposes a rider optimization algorithm-based neural network (ROA-based NN) to classify the MI signals effectively. Pre-processing is done after collecting the dataset of raw EEG signals. The suitable electrodes, such as C3, C4, and Cz, are subsequently chosen from the MI signals. Using the holo-entropy-based WPD feature extractor, tunable Q-factor wavelet transform (T-QWT), and common spatial patterns in the model, the pertinent features are extracted from the chosen electrodes. The developed holo-entropy-based WPD feature examines the electrode structure’s association. As a result, the most diverse signals are removed from the chosen electrodes before being input into the proposed RideNN classifier, where the proposed ride optimization algorithm optimizes classification performance and correctly predicts and classes the output from the MI signals that have been analyzed. The developed RideNN classifier recognizes the patterns more accurately processes more data and tackles the noise and incomplete data effectively. Utilizing the parameters of accuracy, sensitivity, and specificity, the results are evaluated. The PROA-based RideNN and the combined features obtain the maximum accuracy of 92.24%, sensitivity of 92.26%, and specificity of 92.14% for the BCI competition-IV 2a database. The qPROA-based RideNN and the combined features obtain the maximum accuracy of 92.11%, sensitivity of 91.98%, and specificity of 92.35% for the BCI competition-IV 2b database.

Urban computing could create win-win-win solutions to big issues, such as energy consumption, faced by cities. As a huge source of energy consumption, air conditioning has been drawn much attention recently. The temperature control of air conditioning based on big data for reducing energy consumption is a promising research area in urban computing. In this paper, an online learning framework for temperature control of air conditioning is proposed. Then, k-nearest neighbor (KNN), neural network (NN) and support vector machine (SVM) are embedded into this framework to obtain the favorable air condition controlling policy. Extensive computational experiments are conducted. The results show that NN outperforms KNN and SVM for indoor temperature control.