To super-resolve low-resolution (LR) face image suffering from strong noise and fuzzy interference, we present a novel approach for noisy face super-resolution (SR) that is based on three-level information representation constraints. To begin with, we develop a feature distillation network that focuses on extracting pertinent face information, which incorporates both statistical anti-interference models and latent contrast algorithms. Subsequently, we incorporate a face identity embedding model and a discrete wavelet transform model, which serve as additional supervision mechanisms for the reconstruction process. The face identity embedding model ensures the reconstruction of identity information in hypersphere identity metric space, while the discrete wavelet transform model operates in the wavelet domain to supervise the restoration of spatial structures. The experimental results clearly demonstrate the efficacy of our proposed method, which is evident through the lower Learned Perceptual Image Patch Similarity (LPIPS) score and Fréchet Inception Distances (FID), and overall practicability of the reconstructed images.

Radio Frequency Identification (RFID) is one of the key technologies of the Internet of Things. However, during its application, it faces a huge challenge of co-frequency interference cancellation, that is, the tag collision problem. The multi-tag anti-collision problem is modeled as a Blind Source Separation (BSS) problem from the perspective of system communication transmission layer signal processing. In order to reduce the cost of the reader antenna, this paper uses the boundedness of the tag communication signal to propose an underdetermined RFID tag anti-collision method based on Bounded Component Analysis (BCA). This algorithm converts the underdetermined tag into the signal collision model is combined with the BCA mechanism. Verification analysis was conducted using simulation data. The experimental results show that compared with the nonnegative matrix factorization (NMF) algorithm based on minimum correlation and minimum volume constraints, the bounded component analysis method proposed in this article can perform better. Solving the underdetermined collision problem greatly improves the effect of eliminating co-channel interference of tag signals, improves the system bit error rate performance, and reduces the complexity of the underdetermined model system.

In this paper, we present a framework for composing discrete-event simulation models from a large amount of airspace traffic data without using any specific waypoints. The framework consists of two parts. In the first part, abstracted route graphs that indicate representative routes in the airspace are composed. We propose two methods for extracting important routes in the form of graphs based on combination of various technologies such as space partition, trajectory clustering, and skeleton extraction. In the second part, discrete-event simulation models are composed based on statistical information on flight time along each edge of the abstracted route graph. The composed simulation models have intermediate granularity between micro models, such as multi-agent simulation, and macro models, such as queuing models, and therefore they should be classified as mesoscopic models. Finally, we show numerical results to evaluate the accuracy of the simulation model.

Recent studies on facial expression recognition mainly employs discrete category labels to represent emotion states. However, current intelligent emotion interaction systems require more diverse and precise emotion representation metrics, which has been proposed as Valence, Arousal, Dominance (VAD) multi-dimensional continuous emotion parameters. But there are still very less datasets and methods for VAD analysis, making it difficult to meet the needs of large-scale and high-precision emotion cognition. In this letter, we build multi-dimensional facial expression recognition method by using multi-task learning to improve recognition performance through exploiting the consistency between dimensional and categorial emotions. The evaluation results show that the multi-task learning approach improves the prediction accuracy for VAD multi-dimensional emotion. Furthermore, it applies the method to academic outcomes prediction which verifies that introducing the VAD multi-dimensional and multi-task facial expression recognition is effective in predicting academic outcomes. The VAD recognition code is publicly available on github.com/YeeHoran/Multi-task-Emotion-Recognition.

Deep networks are undergoing rapid development. However, as the depth of networks increases, the issue of how to fuse features from different layers becomes increasingly prominent. To address this challenge, we creatively propose a cross-layer feature fusion module based on neural dendrites, termed dendritic learning-based feature fusion (DFF). Compared to other fusion methods, DFF demonstrates superior biological interpretability due to the nonlinear capabilities of dendritic neurons. By integrating the classic ResNet architecture with DFF, we devise the ResNeFt. Benefiting from the unique structure and nonlinear processing capabilities of dendritic neurons, the fused features of ResNeFt exhibit enhanced representational power. Its effectiveness and superiority have been validated on multiple medical datasets.

Image classification is a typical computer vision task widely used in practical applications. The images used for training image classification networks are often clean, i.e., without any image degradation. However, Convolutional neural networks trained on clean images perform poorly on degraded or corrupted images in the real world. In this study, we effectively utilize robust data augmentation (DA) with knowledge distillation to improve the classification performance of degraded images. We first categorize robust data augmentations into geometric-and-color and cut-and-delete DAs. Next, we evaluate the effectual positioning of cut-and-delete DA when we apply knowledge distillation. Moreover, we also experimentally demonstrate that combining the RandAugment and Random Erasing approach for geometric-and-color and cut-and-delete DA improves the generalization of the student network during the knowledge transfer for the classification of degraded images.

Sign constraints are a handy representation of domain-specific prior knowledge that can be incorporated to machine learning. This paper presents new stochastic dual coordinate ascent (SDCA) algorithms that find the minimizer of the empirical risk under the sign constraints. Generic surrogate loss functions can be plugged into the proposed algorithm with the strong convergence guarantee inherited from the vanilla SDCA. The prediction performance is demonstrated on the classification task for microbiological water quality analysis.

SLM (Scalable Logic Module) is a fine-grained reconfigurable logic developed by Kumamoto University. Its small configuration information size characterizes it, resulting in a smaller area for logic cells. We have been developing an SoC-type FPGA called SLMLET to take advantage of SLM. It keeps multiple sets of configuration data in the memory module inside the chip in a compressed form and exchanges them quickly. This paper proposes a simple run-length compression technique called TLC (Tag Less Compression). It achieved a 1.01-3.06 compression ratio, is embedded in the prototype of the SLMLET, and is available now. Then, we propose DMC (Duplication Module Compression), which uses repeatedly appearing patterns in the SLM configuration data. The DMC achieves a better compression ratio for complicated designs that are hard to compress with TLC.

A binarized neural network (BNN) inference accelerator is designed in which weights are stores in loadless four-transistor static random access memory (4T SRAM) cells. A time-multiplexed exclusive NOR (XNOR) multiplier with switched capacitors is proposed which prevents the loadless 4T SRAM cell from being destroyed in the operation. An accumulator with current sensing scheme is also proposed to make the multiply-accumulate operation (MAC) completely linear and read-disturb free. The BNN inference accelerator is applied to the MNIST dataset recognition problem with accuracy of 96.2% for 500 data and the throughput, the energy efficiency and the area efficiency are confirmed to be 15.50 TOPS, 72.17 TOPS/W and 50.13 TOPS/mm2, respectively, by HSPICE simulation in 32 nm technology. Compared with the conventional SRAM cell based BNN inference accelerators which are scaled to 32 nm technology, the synapse cell size is reduced to less than 16% (0.235 μm2) and the cell efficiency (synapse array area/synapse array plus peripheral circuits) is 73.27% which is equivalent to the state-of-the-art of the SRAM cell based BNN accelerators.

Transcatheter renal denervation (RDN) is a treatment for resistant hypertension, which is performed by ablating the renal nerves located outside the artery using a catheter from inside the artery. Our previous studies simulated the temperature during RDN by using constant physical properties of biological tissue to validate the various catheter RDN devices. Some other studies report temperature dependence of physical properties of biological tissues. However, there are no studies that have measured the electrical properties of low water content tissues. Adipose tissue, a type of low water content tissue, is related to RDN closely. Therefore, it is important to know the temperature dependence of the electrical constants of adipose tissue. In this study, we measured the relationship between the electrical constants and the temperature of bovine adipose tissue. Next, the obtained equation of the relationship between relative permittivity of adipose tissue and temperature was introduced. In addition, the temperature dependence of the electrical constants of high water content tissues and the temperature dependence of the thermal constants of biological tissues were also introduced into the temperature analysis. After 180 seconds of heating, the temperature of the model with the temperature dependence of the physical properties was 7.25°C lower than the model without the temperature dependence of the physical properties at a certain position. From the results, it can be said that the temperature dependence of physical properties will be significant when an accurate temperature analysis is required.

The Doppler velocity enhanced 79 GHz band millimeter wave (MMW) radar imaging approach is presented here, assuming a human body imaging or recognition application. There are numerous situations in which the spatial resolution is insufficient, due to aperture angle limitations, especially for vehicle mounted MMW radar systems. As the 79 GHz band MMW radar has a definitive advantage for higher Doppler velocity resolution even with a short coherent processing interval (CPI), this study introduces the Doppler velocity decomposed imaging scheme, focusing on micro-Doppler variations of the walking human model. The real experimental data show that our proposed approach provides further improvement for accurate and high resolution radar imaging.

In order to support a person’s various activities (e.g., primary industry, elderly care), an IoT (Internet of Things) system supporting sensing technologies that observe the status of various people, things, and places in the real world is attracting attention. The existing studies have utilized the camera device for the sensing technology to observe the condition of the real world, but the use of the camera device has problems related to the limitation of the observation period and the invasion of privacy. Therefore, new IoT systems utilizing three-dimensional LiDAR (Light Detection And Ranging) have been proposed because they can obtain three-dimensional spatial information about the real world by solving problems. However, several problems exist with the use of 3D LiDAR in the deployment on the real fields. The 3D LiDAR requires much electric power for observing the three-dimensional spatial information. In addition, the annotation process of a large volume of point-cloud data for constructing a machine-learning model requires significant time and effort. Therefore, in this study, we propose IoT systems utilizing 3D LiDAR for observing the status of targets and equipping them with new technologies to improve the practicality of the use of 3D LiDAR. First, a linkage function is designed to achieve power saving for an entire system. The linkage function operates only a sensing technology with low power consumption during normal operation and activates the 3D LiDAR with high power consumption only when it is estimated that an observation target is approaching. Second, a self-learning function is built to analyze the data collected by not only the 3D LiDAR but also the camera for automatically generating a large amount of training data with the correct label which is estimated by analyzing the camera images. Through the experimental evaluations using a prototype system, it is confirmed that the sensing technologies can correctly be interconnected to reduce the total power consumption. In addition, the machine-learning model constructed by the self-learning function can accurately estimate the status of the targets.

In-band full duplex requires digital self-interference cancellations (SICs). However, digital SICs are significantly challenged due to radio-frequency (RF) impairments. We have proposed several designs and analyses of digital SIC based on parallel Hammerstein models addressing RF impairments. Our proposed designs and analyses use orthogonal frequency division multiplexing in the Hammerstein SICs. Moreover, we have developed a testbed to evaluate the Hammerstein and our proposed SICs. Our experimental results showed that the self-interference cancellation ratio was 48.1 dB for one of the Hammerstein SICs. This paper reveals that our canceler designs and theoretical analyses can improve the Hammerstein SICs.

Characteristics of rain attenuation statistics for the Ku-band satellite signals are investigated among the three earth stations at Osaka Electro-Communication University (OECU, Neyagawa, Osaka), NTT Yokosuka R&D Center (Yokosuka, Kanagawa), and satellite base station (Matsuyama, Ehime), respectively, from April 2022 to March 2023. The time difference of the attenuation occurrence among these stations is well explained by the motion of rain fronts and extratropical cyclones obtained from the weather charts. Rain attenuation characteristics such as duration time are shown to be largely affected by the speed of the rain area motion around each station. The scale of rain cells inferred from duration time and rain area speed is found to be increased up to around 10 km for the 5 dB attenuation according to the rainfall rate at each rainfall event. The time delayed diversity effects are also examined using the attenuation data observed at 1 min interval. The results are converted to the site diversity effects with the distance up to about 10 km by the rain area motion around each station. A novel method is thus proposed to estimate the site diversity effects from the 1 min attenuation data observed at only one station during at least one year. The joint time percentages agree fairly well with the ITU-R recommendations up to about 10 km distance at the original time percentages of more than 0.05%.

This study examines the threat of information leakage when digital ICs, which process sensitive information such as cryptographic operations and handling of personal and confidential information, are mounted on printed circuit boards with split ground (GND) planes. We modeled the mechanism of generating such information leakage and proposed a methodology to control it. It is known that the GND plane of a printed circuit board on which digital integrated circuits are mounted should be solid and undivided to ensure signal integrity, power integrity, and electromagnetic compatibility. However, in actual designs, printed circuit boards may have split GND planes to isolate analog and digital circuits, isolate high-voltage and low-voltage circuits, or integrate multi-function electronic control units. Such split GND planes can increase the risk of electromagnetic information leakage. We, therefore, investigated a side-channel attack standard evaluation board, SASEBO-G, which has been reported to leak cryptographic information superimposed on common-mode currents, known as one of the major causes of electromagnetic emanation. Our experimental results showed that the split GND planes were the dominant cause of common-mode (CM) information leakage. Next, we constructed an equivalent circuit model of the dominant leakage mechanism and confirmed that the behavior of side-channel information leakage superimposed in the simulated CM current was consistent with the measured results. We also confirmed that to mitigate side-channel information leakage in CM caused by the potential difference between the split GND planes, the impedance should be reduced in the information leakage band by connecting the GND planes with capacitors, and the like. In addition, the RF band coupling between cables should be weakened if the cables are connected to the split GND planes.

Radio Frequency Identification (RFID) is crucial for the Internet of Things, with a key challenge being the efficient prevention of tag collisions for quick identification. This paper presents a novel approach for rapid tag recognition in small to medium-sized warehouses, combining a tag optimization feature set with a tail code recognition mechanism. To minimize the frequency of scanning for duplicate tags and reduce the occurrence of collisions, we construct an optimization feature set based on the reader’s position. This set helps in assessing the likelihood of tag repetition through its linear variation. It also incorporates a tail code mechanism that recognizes only the last 22 digits of the tag’s EPC code, significantly speeding up identification. The tail code length is dynamically adjusted based on the number of tags to maintain uniqueness. Simulation results indicate that our approach significantly reduces the identification of duplicate tags and minimizes the instances of collisions.

We consider the Doppler ambiguity compensation problem for weak moving target detection in passive bistatic radar. Detecting an unknown high-speed weak target has a high probability of the presence of Doppler ambiguity, which will decrease the integration performance and accordingly make the target detection difficult under low signal-to-noise ratio (SNR) environments. Resorting to the well-known keystone transform (KT) method, an approach to compensate for the Doppler ambiguity within the batch is proposed for the first time. The proposed approach establishes a good coupling between the reference and echo signals by adding a frequency shift related to the Doppler frequency in the procedure of computing the cross ambiguity function (CAF). Simulation results show that the coherent integration gain of our approach is close to the theoretical upper bound even in the presence of Doppler ambiguity.

We present Ksformer, utilizing Multi-scale Key-select Routing Attention (MKRA) for intelligent selection of key areas through multi-channel, multi-scale windows with a top-k operator, and Lightweight Frequency Processing Module (LFPM) to enhance high-frequency features, outperforming other dehazing methods in tests.

Accurately simulating physical systems is essential in various fields. In recent years, deep learning has been used to automatically build models of such systems by learning from data. One such method is the neural ordinary differential equation (neural ODE), which treats the output of a neural network as the time derivative of the system states. However, while this and related methods have shown promise, their training strategies still require further development. Inspired by error analysis techniques in numerical analysis while replacing numerical errors with modeling errors, we propose the error-analytic strategy to address this issue. Therefore, our strategy can capture long-term errors and thus improve the accuracy of long-term predictions.

We proposed a procedure to pre-process data used in a vector autoregressive (VAR) modeling of a temporal point process by using kernel density estimation. Vector autoregressive modeling of point-process data, for example, is being used for causality inference. The VAR model discretizes the timeline into small windows, and creates a time series by the presence of events in each window, and then models the presence of an event at the next time step by its history. The problem is that to get a longer history with high temporal resolution required a large number of windows, and thus, model parameters. We proposed the local density estimation procedure, which, instead of using the binary presence as the input to the model, performed kernel density estimation of the event history, and discretized the estimation to be used as the input. This allowed us to reduce the number of model parameters, especially in sparse data. Our experiment on a sparse Poisson process showed that this procedure vastly increases model prediction performance.