Data compression is a useful method to reduce the communication energy consumption in wireless sensor networks (WSNs). Most existing neural network compression methods focus on improving the compression and reconstruction accuracy (i.e., increasing parameters and layers), ignoring the computation consumption of the network and its application ability in WSNs. In contrast, we pay attention to the computation consumption and application of neural networks, and propose an extremely simple and efficient neural network data compression model. The model combines the feature extraction advantages of Convolutional Neural Network (CNN) with the data generation ability of Variational Autoencoder (VAE) and Restricted Boltzmann Machine (RBM), we call it CBN-VAE. In particular, we propose a new efficient convolutional structure: Downsampling-Convolutional RBM (D-CRBM), and use it to replace the standard convolution to reduce parameters and computational consumption. Specifically, we use the VAE model composed of multiple D-CRBM layers to learn the hidden mathematical features of the sensing data, and use this feature to compress and reconstruct the sensing data. We test the performance of the model by using various real-world WSN datasets. Under the same network size, compared with the CNN, the parameters of CBN-VAE model are reduced by 73.88% and the floating-point operations (FLOPs) are reduced by 96.43% with negligible accuracy loss. Compared with the traditional neural networks, the proposed model is more suitable for application on nodes in WSNs. For the Intel Lab temperature data, the average Signal-to-Noise Ratio (SNR) value of the model can reach 32.51 dB, the average reconstruction error value is 0.0678 °C. The node communication energy consumption can be reduced by 95.83%. Compared with the traditional compression methods, the proposed model has better compression and reconstruction accuracy. At the same time, the experimental results show that the model has good fault detection performance and anti-noise ability. When reconstructing data, the model can effectively avoid fault and noise data.
This letter proposes a simple and effective EEG classification method using convolutional neural networks (CNN). The EEG signals are first converted to character sequences in order to simplify feature extraction and meanwhile reduce data dimensionality. The character sequences are then transformed according to binary encoding and finally fed into a specific CNN, which gives consideration to both classification accuracy and computation efficiency. Experimental simulations show the proposed method has the advantages of simple preprocessing and lightweight network structure, the average accuracy on two BCI Competition datasets can reach 94.36% and 81.8%, while the average accuracy on the self-made dataset is 71.67%, outperforming traditional methods. Furthermore, the proposed algorithm has been successfully applied to control the robotic arm of an EEG-based manipulator system, further validating its feasibility.
21 January 2020 in Environmental Science and Pollution Research
The recognition of steady-state visual evoked potential (SSVEP) signals in brain-computer interface (BCI) systems is challenging due to the lack of training data and significant inter-subject variability. To address this, we propose a novel unsupervised transfer learning framework that enhances SSVEP recognition without requiring any subject-specific calibration. Our method employs a three-stage pipeline: (1) preprocessing with similarity-aware subject selection and Euclidean alignment to mitigate domain shifts; (2) hybrid feature extraction combining canonical correlation analysis (CCA) and task-related component analysis (TRCA) to enhance signal-to-noise ratio and phase sensitivity; and (3) weighted correlation fusion for robust classification. Extensive evaluations on the Benchmark and BETA datasets demonstrate that our approach achieves state-of-the-art performance, with average accuracies of 83.20% and 69.08% at 1 s data length, respectively—significantly outperforming existing methods like ttCCA and Ensemble-DNN. The highest information transfer rate reaches 157.53 bits min −1 , underscoring the framework’s practical potential for plug-and-play SSVEP-based BCIs.