Hybrid convolution neural network with channel attention mechanism for sensor-based human activity recognition

1, the explored HAR framework consists of four key process steps: data acquisition, data pre-processing, model training, and model assessment. UCI-HAR datasetThis paper utilizes the ”UCI Human Activity Recognition Using Smartphone Dataset (UCI-HAR)”27 as the public activity dataset for the proposed approach. The proposed hybrid convolutional neural networkThis research proposes an effective biometric recognition model called ResNet-BiGRU-SE for utilizing motion signal data captured from smartphone sensors. A residual layer can be represented as follows:$$\begin{aligned} \text {ELU}(x) = {\left{ \begin{array}{ll} x &{} \quad \text {if } x \ge 0\ \alpha (e^x - 1) &{} \quad \text {if } x < 0 \end{array}\right. } 4, this residual block consists of Conv1D layers, BN layers, ELU layers, SE modules, and shortcut connections with BiGRU.