In recent years, the use of WiFi fingerprints for indoor positioning has grown in popularity, largely due to the widespread availability of WiFi and the proliferation of mobile communication devices. However, many existing methods for constructing fingerprint datasets rely on labor-intensive and time-consuming processes of collecting large amounts of data. Additionally, these methods often focus on ideal laboratory environments, rather than considering the practical challenges of large multi-floor buildings. To address these issues, we present a novel WiDAGCN model that can be trained using a small number of labeled site survey data and large amounts of unlabeled crowdsensed WiFi fingerprints. By constructing heterogeneous graphs based on received signal strength indicators (RSSIs) between waypoints and WiFi access points (APs), our model is able to effectively capture the topological structure of the data. We also incorporate graph convolutional networks (GCNs) to extract graph-level embeddings, a feature that has been largely overlooked in previous WiFi indoor localization studies. To deal with the challenges of large amounts of unlabeled data and multiple data domains, we employ a semi-supervised domain adversarial training scheme to effectively utilize unlabeled data and align the data distributions across domains. Our system is evaluated using a public indoor localization dataset that includes multiple buildings, and the results show that it performs competitively in terms of localization accuracy in large buildings.

Functional magnetic resonance imaging (fMRI) is a commonly used method of brain research. However, due to the complexity and particularity of the fMRI task, it is difficult to find enough subjects, resulting in a small and, often, imbalanced dataset. A dataset with small samples causes overfitting of the learning model, and the imbalance will make the model insensitive to the minority class, which has been a problem in classification. It is of great significance to classify fMRI data with small and imbalanced samples. In the present study, we propose a 3-step method on a small and imbalanced fMRI dataset from a word-scene memory task. The steps of the method are as follows: (1) An independent component analysis is performed to reduce the dimension of data; (2) The synthetic minority oversampling technique is used to generate new samples of the minority class to balance data; (3) A convolution-Gated Recurrent Unit (GRU) network is used to classify the independent component signals, indicating whether the subjects are performing episodic memory tasks. The accuracy of the proposed method is 72.2%, which improves the classification performance compared with traditional classifiers such as support vector machines (SVM), logistic regression (LGR), linear discriminant analysis (LDA) and k-nearest neighbor (KNN), and this study gives a biomarker for evaluating the reactivation of episodic memory.

Vibration rendering is essential for creating realistic tactile experiences in human-virtual object interactions, such as in video game controllers and VR devices. By dynamically adjusting vibration parameters based on user actions, these systems can convey spatial features and contribute to texture representation. However, generating arbitrary vibrations to replicate real-world material textures is challenging due to the large parameter space. This study proposes a human-in-the-loop vibration generation model based on user preferences. To enable users to easily control the generation of vibration samples with large parameter spaces, we introduced an optimization model based on Differential Subspace Search (DSS) and Generative Adversarial Network (GAN). With DSS, users can employ a one-dimensional slider to easily modify the high-dimensional latent space to ensure that the GAN can generate desired vibrations. We trained the generative model using an open dataset of tactile vibration data and selected five types of vibrations as target samples for the generation experiment. Extensive user experiments were conducted using the generated and real samples. The results indicated that our system could generate distinguishable samples that matched the target characteristics. Moreover, we established a correlation between subjects' ability to distinguish real samples and their ability to distinguish generated samples

The physical stimuli provided by ultrasonic mid-air haptic interfaces are relatively limited in their ability to represent compliance and stiffness. In our previous research, we proposed to apply spatiotemporal modulation (STM) for the ultrasonic focus to simulate the contact process between a finger and a virtual surface, extending the perceived range of softness by varying the rate of contact area change and focal point speed. In the present study, we proposed to combine STM with amplitude modulation (AM), further investigating the effects of two aforementioned factors and output power variation, on perceived softness by simulating the contact process between a palm and a virtual small sphere. Based on the results of psychophysical experiments, we found that haptic stimuli with a high contact area change rate, fast focal point speed, and low output power amplitude were perceived as softer. Consequently, we covered a wide perception range, i.e., 5.3-fold softer and 4-fold harder than the modulus by altering the parameters. Our research contributes to the understanding and design of soft object simulation, and thus extends the rendering capabilities of ultrasound haptic displays in VR and AR.

Episodic memory allows a person to recall and mentally reexperience specific episodes from one’s personal past. Studies of episodic memory are of great significance for the diagnosis and the exploration of the mechanism of memory generation. Most of the current studies focus on certain brain regions and pay less attention to the interrelationship between multiple brain regions. To explore the interrelationship in the brain network, we use an open fMRI dataset to construct the brain functional connectivity and effective connectivity network. We establish a binary directed network of the memory when it is reactivated. The binary directed network shows that the occipital lobe and parietal lobe have the most causal connections. The number of edges starting from the superior parietal lobule is the highest, with 49 edges, and 31 of which are connected to the occipital cortex. This means that the interaction between the superior parietal lobule and the occipital lobe plays the most important role in episodic memory, and the superior parietal lobule plays a more causal role in causality. In addition, memory regions such as the precuneus and fusiform also have some edges. The results show that the posterior parietal cortex plays an important role of hub node in the episodic memory network. From the brain network model, more information can be obtained, which is conducive to exploring the brain’s changing pattern in the whole memory process.