Zeyuan JU Zhipeng LIU Yu GAO Haotian LI Qianhang DU Kota YOSHIKAWA Shangce GAO
Medical imaging plays an indispensable role in precise patient diagnosis. The integration of deep learning into medical diagnostics is becoming increasingly common. However, existing deep learning models face performance and efficiency challenges, especially in resource-constrained scenarios. To overcome these challenges, we introduce a novel dendritic neural efficientnet model called DEN, inspired by the function of brain neurons, which efficiently extracts image features and enhances image classification performance. Assessments on a diabetic retinopathy fundus image dataset reveal DEN’s superior performance compared to EfficientNet and other classical neural network models.
We consider the problem of finding the best subset of sensors in wireless sensor networks where linear Bayesian parameter estimation is conducted from the selected measurements corrupted by correlated noise. We aim to directly minimize the estimation error which is manipulated by using the QR and LU factorizations. We derive an analytic result which expedites the sensor selection in a greedy manner. We also provide the complexity of the proposed algorithm in comparison with previous selection methods. We evaluate the performance through numerical experiments using random measurements under correlated noise and demonstrate a competitive estimation accuracy of the proposed algorithm with a reasonable increase in complexity as compared with the previous selection methods.
Software refactoring is an important process in software development. During software refactoring, code smell is a popular research topic that refers to design or implementation flaws in the software. Large class is one of the most concerning code smells in software refactoring. Detecting and refactoring such problem has a profound impact on software quality. In past years, software metrics and clustering techniques have commonly been used for the large class detection. However, deep-learning-based approaches have also received considerable attention in recent studies. In this study, we apply graph neural networks (GNNs), an important division of deep learning, to address the problem of large class detection. First, to support the extensive data requirements of the deep learning task, we apply a semiautomatic approach to generate a substantial number of data samples. Next, we design a new type of directed heterogeneous graph (DHG) as an input graph using the methods similarity matrix and software metrics. We construct an input graph for each class sample and make the graph classification with GNNs to identify the smelly classes. In our experiments, we apply three typical GNN model architectures for large class detection and compare the results with those of previous studies. The results show that the proposed approach can achieve more accurate and stable detection performance.
To improve the recognition rate of the end-to-end modulation recognition method based on deep learning, a modulation recognition method of communication signals based on a cascade network is proposed, which is composed of two networks: Stacked Denoising Auto Encoder (SDAE) network and DCELDNN (Dilated Convolution, ECA Mechanism, Long Short-Term Memory, Deep Neural Networks) network. SDAE network is used to denoise the data, reconstruct the input data through encoding and decoding, and extract deep information from the data. DCELDNN network is constructed based on the CLDNN (Convolutional, Long Short-Term Memory, Fully Connected Deep Neural Networks) network. In the DCELDNN network, dilated convolution is used instead of normal convolution to enlarge the receptive field and extract signal features, the Efficient Channel Attention (ECA) mechanism is introduced to enhance the expression ability of the features, the feature vector information is integrated by a Global Average Pooling (GAP) layer, and signal features are extracted by the DCELDNN network efficiently. Finally, end-to-end classification recognition of communication signals is realized. The test results on the RadioML2018.01a dataset show that the average recognition accuracy of the proposed method reaches 63.1% at SNR of -10 to 15 dB, compared with CNN, LSTM, and CLDNN models, the recognition accuracy is improved by 25.8%, 12.3%, and 4.8% respectively at 10 dB SNR.
Youquan XIAN Lianghaojie ZHOU Jianyong JIANG Boyi WANG Hao HUO Peng LIU
In recent years, blockchain has been widely applied in the Internet of Things (IoT). Blockchain oracle, as a bridge for data communication between blockchain and off-chain, has also received significant attention. However, the numerous and heterogeneous devices in the IoT pose great challenges to the efficiency and security of data acquisition for oracles. We find that the matching relationship between data sources and oracle nodes greatly affects the efficiency and service quality of the entire oracle system. To address these issues, this paper proposes a distributed and efficient oracle solution tailored for the IoT, enabling fast acquisition of real-time off-chain data. Specifically, we first design a distributed oracle architecture that combines both Trusted Execution Environment (TEE) devices and ordinary devices to improve system scalability, considering the heterogeneity of IoT devices. Secondly, based on the trusted node information provided by TEE, we determine the matching relationship between nodes and data sources, assigning appropriate nodes for tasks to enhance system efficiency. Through simulation experiments, our proposed solution has been shown to effectively improve the efficiency and service quality of the system, reducing the average response time by approximately 9.92% compared to conventional approaches.
Feng LIU Helin WANG Conggai LI Yanli XU
This letter proposes a scheme for the backward transmission of the propagation-delay based three-user X channel, which is reciprocal to the forward transmission. The given scheme successfully delivers 10 expected messages in 6 time-slots by cyclic interference alignment without loss of degrees of freedom, which supports efficient bidirectional transmission between the two ends of the three-user X channel.
This study explores adaptive output feedback leader-following in networks of linear systems utilizing switching logic. A local state observer is employed to estimate the true state of each agent within the network. The proposed protocol is based on the estimated states obtained from neighboring agents and employs a switching logic to tune its adaptive gain by utilizing only local neighboring information. The proposed leader-following protocol is fully distributed because it has a distributed adaptive gain and relies on only local information from its neighbors. Consequently, compared to conventional adaptive protocols, the proposed design method provides the advantages of a very simple adaptive law and dynamics with a low dimension.
Izumi TSUNOKUNI Gen SATO Yusuke IKEDA Yasuhiro OIKAWA
This paper reports a spatial extrapolation of the sound field with a physics-informed neural network. We investigate the spatial extrapolation of the room impulse responses with physics-informed SIREN architecture. Furthermore, we proposed a noise-robust extrapolation method by introducing a tolerance term to the loss function.
Shi BAO Xiaoyan SONG Xufei ZHUANG Min LU Gao LE
Images with rich color information are an important source of information that people obtain from the objective world. Occasionally, it is difficult for people with red-green color vision deficiencies to obtain color information from color images. We propose a method of color correction for dichromats based on the physiological characteristics of dichromats, considering hue information. First, the hue loss of color pairs under normal color vision was defined, an objective function was constructed on its basis, and the resultant image was obtained by minimizing it. Finally, the effectiveness of the proposed method is verified through comparison tests. Red-green color vision deficient people fail to distinguish between partial red and green colors. When the red and green connecting lines are parallel to the a* axis of CIE L*a*b*, red and green perception defectives cannot distinguish the color pair, but can distinguish the color pair parallel to the b* axis. Therefore, when two colors are parallel to the a* axis, their color correction yields good results. When color correction is performed on a color, the hue loss between the two colors under normal color vision is supplemented with b* so that red-green color vision-deficient individuals can distinguish the color difference between the color pairs. The magnitude of the correction is greatest when the connecting lines of the color pairs are parallel to the a* axis, and no color correction is applied when the connecting lines are parallel to the b* axis. The objective evaluation results show that the method achieves a higher score, indicating that the proposed method can maintain the naturalness of the image while reducing confusing colors.
Hongliang FU Qianqian LI Huawei TAO Chunhua ZHU Yue XIE Ruxue GUO
Speech emotion recognition (SER) is a key research technology to realize the third generation of artificial intelligence, which is widely used in human-computer interaction, emotion diagnosis, interpersonal communication and other fields. However, the aliasing of language and semantic information in speech tends to distort the alignment of emotion features, which affects the performance of cross-corpus SER system. This paper proposes a cross-corpus SER model based on causal emotion information representation (CEIR). The model uses the reconstruction loss of the deep autoencoder network and the source domain label information to realize the preliminary separation of causal features. Then, the causal correlation matrix is constructed, and the local maximum mean difference (LMMD) feature alignment technology is combined to make the causal features of different dimensions jointly distributed independent. Finally, the supervised fine-tuning of labeled data is used to achieve effective extraction of causal emotion information. The experimental results show that the average unweighted average recall (UAR) of the proposed algorithm is increased by 3.4% to 7.01% compared with the latest partial algorithms in the field.
Ji XI Yue XIE Pengxu JIANG Wei JIANG
Currently, a significant portion of acoustic scene categorization (ASC) research is centered around utilizing Convolutional Neural Network (CNN) models. This preference is primarily due to CNN’s ability to effectively extract time-frequency information from audio recordings of scenes by employing spectrum data as input. The expression of many dimensions can be achieved by utilizing 2D spectrum characteristics. Nevertheless, the diverse interpretations of the same object’s existence in different positions on the spectrum map can be attributed to the discrepancies between spectrum properties and picture qualities. The lack of distinction between different aspects of input information in ASC-based CNN networks may result in a decline in system performance. Considering this, a feature pyramid segmentation (FPS) approach based on CNN is proposed. The proposed approach involves utilizing spectrum features as the input for the model. These features are split based on a preset scale, and each segment-level feature is then fed into the CNN network for learning. The SoftMax classifier will receive the output of all feature scales, and these high-level features will be fused and fed to it to categorize different scenarios. The experiment provides evidence to support the efficacy of the FPS strategy and its potential to enhance the performance of the ASC system.
At Crypto 2019, Gohr first adopted the neural distinguisher for differential cryptanalysis, and since then, this work received increasing attention. However, most of the existing work focuses on improving and applying the neural distinguisher, the studies delving into the intrinsic principles of neural distinguishers are finite. At Eurocrypt 2021, Benamira et al. conducted a study on Gohr’s neural distinguisher. But for the neural distinguishers proposed later, such as the r-round neural distinguishers trained with k ciphertext pairs or ciphertext differences, denoted as NDcpk_r (Gohr’s neural distinguisher is the special NDcpk_r with K = 1) and NDcdk_r , such research is lacking. In this work, we devote ourselves to study the intrinsic principles and relationship between NDcdk_r and NDcpk_r. Firstly, we explore the working principle of NDcd1_r through a series of experiments and find that it strongly relies on the probability distribution of ciphertext differences. Its operational mechanism bears a strong resemblance to that of NDcp1_r given by Benamira et al.. Therefore, we further compare them from the perspective of differential cryptanalysis and sample features, demonstrating the superior performance of NDcp1_r can be attributed to the relationships between certain ciphertext bits, especially the significant bits. We then extend our investigation to NDcpk_r, and show that its ability to recognize samples heavily relies on the average differential probability of k ciphertext pairs and some relationships in the ciphertext itself, but the reliance between k ciphertext pairs is very weak. Finally, in light of the findings of our research, we introduce a strategy to enhance the accuracy of the neural distinguisher by using a fixed difference to generate the negative samples instead of the random one. Through the implementation of this approach, we manage to improve the accuracy of the neural distinguishers by approximately 2% to 8% for 7-round Speck32/64 and 9-round Simon32/64.
Ken ASANO Masanori NATSUI Takahiro HANYU
The development of energy-efficient neural network hardware using magnetic tunnel junction (MTJ) devices has been widely investigated. One of the issues in the use of MTJ devices is large write energy. Since MTJ devices show stochastic behaviors, a large write current with enough time length is required to guarantee the certainty of the information held in MTJ devices. This paper demonstrates that quantized neural networks (QNNs) exhibit high tolerance to bit errors in weights and an output feature map. Since probabilistic switching errors in MTJ devices do not have always a serious effect on the performance of QNNs, large write energy is not required for reliable switching operations of MTJ devices. Based on the evaluation results, we achieve about 80% write-energy reduction on buffer memory compared to the conventional method. In addition, it is demonstrated that binary representation exhibits higher bit-error tolerance than the other data representations in the range of large error rates.
Hajime MIGITA Yuki NAKAGOSHI Patrick FINNERTY Chikara OHTA Makoto OKUHARA
To enhance fuel efficiency and lower manufacturing and maintenance costs, in-vehicle wireless networks can facilitate the weight reduction of vehicle wire harnesses. In this paper, we utilize the Impulse Radio-Ultra Wideband (IR-UWB) of IEEE 802.15.4a/z for in-vehicle wireless networks because of its excellent signal penetration and robustness in multipath environments. Since clear channel assessment is optional in this standard, we employ polling control as a multiple access control to prevent interference within the system. Therein, the preamble overhead is large in IR-UWB of IEEE 802.15.4a/z. Hence, aggregating as much sensor data as possible within each frame is more efficient. In this paper, we assume that reading out data from sensors and sending data to actuators is periodical and that their respective phases can be adjusted. Therefore, this paper proposes an integer linear programming-based scheduling algorithm that minimizes the number of transmitted frames by adjusting the read and write phases. Furthermore, we provide a heuristic algorithm that computes a sub-optimal but acceptable solution in a shorter time. Experimental validation shows that the data aggregation of the proposed algorithms is robust against interference.
Terahertz (THz) ultra-massive multiple-input multiple-output (UM-MIMO) is envisioned as a key enabling technology of 6G wireless communication. In UM-MIMO systems, downlink channel state information (CSI) has to be fed to the base station for beamforming. However, the feedback overhead becomes unacceptable because of the large antenna array. In this letter, the characteristic of CSI is explored from the perspective of data distribution. Based on this characteristic, a novel network named Attention-GRU Net (AGNet) is proposed for CSI feedback. Simulation results show that the proposed AGNet outperforms other advanced methods in the quality of CSI feedback in UM-MIMO systems.
Akira KITAYAMA Goichi ONO Hiroaki ITO
Edge devices with strict safety and reliability requirements, such as autonomous driving cars, industrial robots, and drones, necessitate software verification on such devices before operation. The human cost and time required for this analysis constitute a barrier in the cycle of software development and updating. In particular, the final verification at the edge device should at least strictly confirm that the updated software is not degraded from the current it. Since the edge device does not have the correct data, it is necessary for a human to judge whether the difference between the updated software and the operating it is due to degradation or improvement. Therefore, this verification is very costly. This paper proposes a novel automated method for efficient verification on edge devices of an object detection AI, which has found practical use in various applications. In the proposed method, a target object existence detector (TOED) (a simple binary classifier) judges whether an object in the recognition target class exists in the region of a prediction difference between the AI’s operating and updated versions. Using the results of this TOED judgement and the predicted difference, an automated verification system for the updated AI was constructed. TOED was designed as a simple binary classifier with four convolutional layers, and the accuracy of object existence judgment was evaluated for the difference between the predictions of the YOLOv5 L and X models using the Cityscapes dataset. The results showed judgement with more than 99.5% accuracy and 8.6% over detection, thus indicating that a verification system adopting this method would be more efficient than simple analysis of the prediction differences.
Daxiu ZHANG Xianwei LI Bo WEI Yukun SHI
With the increase of the number of Mobile User Equipments (MUEs), numerous tasks that with high requirements of resources are generated. However, the MUEs have limited computational resources, computing power and storage space. In this paper, a joint coverage constrained task offloading and resource allocation method based on deep reinforcement learning is proposed. The aim is offloading the tasks that cannot be processed locally to the edge servers to alleviate the conflict between the resource constraints of MUEs and the high performance task processing. The studied problem considers the dynamic variability and complexity of the system model, coverage, offloading decisions, communication relationships and resource constraints. An entropy weight method is used to optimize the resource allocation process and balance the energy consumption and execution time. The results of the study show that the number of tasks and MUEs affects the execution time and energy consumption of the task offloading and resource allocation processes in the interest of the service provider, and enhances the user experience.
Dongjae LEE Deukjo HONG Jaechul SUNG Seokhie HONG
In this study, we focus on evaluating the false-positive probability of the Demirci-Selçuk meet-in-the-middle attack, particularly within the context of configuring precomputed tables with multisets. During the attack, the adversary effectively reduces the size of the key space by filtering out the wrong keys, subsequently recovering the master key from the reduced key space. The false-positive probability is defined as the probability that a wrong key will pass through the filtering process. Due to its direct impact on the post-filtering key space size, the false-positive probability is an important factor that influences the complexity and feasibility of the attack. However, despite its significance, the false-positive probability of the multiset-based Demirci-Selçuk meet-in-the-middle attack has not been thoroughly discussed, to the best of our knowledge. We generalize the Demirci-Selçuk meet-in-the-middle attack and present a sophisticated method for accurately calculating the false-positive probability. We validate our methodology through toy experiments, demonstrating its high precision. Additionally, we propose a method to optimize an attack by determining the optimal format of precomputed data, which requires the precise false-positive probability. Applying our approach to previous attacks on AES and ARIA, we have achieved modest improvements. Specifically, we enhance the memory complexity and time complexity of the offline phase of previous attacks on 7-round AES-128/192/256, 7-round ARIA-192/256, and 8-round ARIA-256 by factors ranging from 20.56 to 23. Additionally, we have improved the overall time complexity of attacks on 7-round ARIA-192/256 by factors of 20.13 and 20.42, respectively.
Takahiro IINUMA Yudai EBATO Sou NOBUKAWA Nobuhiko WAGATSUMA Keiichiro INAGAKI Hirotaka DOHO Teruya YAMANISHI Haruhiko NISHIMURA
Stochastic resonance is a representative phenomenon in which the degree of synchronization with a weak input signal is enhanced using additive stochastic noise. In systems with multiple chaotic attractors, the chaos-chaos intermittent behavior in attractor-merging bifurcation induces chaotic resonance, which is similar to the stochastic resonance and has high sensitivity. However, controlling chaotic resonance is difficult because it requires adjusting the internal parameters from the outside. The reduced-region-of-orbit (RRO) method, which controls the attractor-merging bifurcation using an external feedback signal, is employed to overcome this issue. However, the lower perturbation of the feedback signal requires further improvement for engineering applications. This study proposed an RRO method with more sophisticated and less perturbed feedback signals, called the double-Gaussian-filtered RRO (DG-RRO) method. The inverse sign of the map function and double Gaussian filters were used to improve the local specification, i.e., the concentration around the local maximum/minimum in the feedback signals, called the DG-RRO feedback signals. Owing to their fine local specification, these signals achieved the attractor-merging bifurcation with significantly smaller feedback perturbation than that in the conventional RRO method. Consequently, chaotic resonance was induced through weak feedback perturbation. It exhibited greater synchronization against weak input signals than that induced by the conventional RRO feedback signal and sustained the same level of response frequency range as that of the conventional RRO method. These advantages may pave the way for utilizing chaotic resonance in engineering scenarios where the stochastic resonance has been applied.
Zhimin SHAO Chunxiu LIU Cong WANG Longtan LI Yimin LIU Zaiyan ZHOU
Data resource sharing can guarantee the reliable and safe operation of distribution power grid. However, it faces the challenges of low security and high delay in the sharing process. Consortium blockchain can ensure the security and efficiency of data resource sharing, but it still faces problems such as arbitrary master node selection and high consensus delay. In this paper, we propose an improved practical Byzantine fault tolerance (PBFT) consensus algorithm based on intelligent consensus node selection to realize high-security and real-time data resource sharing for distribution power grid. Firstly, a blockchain-based data resource sharing model is constructed to realize secure data resource storage by combining the consortium blockchain and interplanetary file system (IPFS). Then, the improved PBFT consensus algorithm is proposed to optimize the consensus node selection based on the upper confidence bound of node performance. It prevents Byzantine nodes from participating in the consensus process, reduces the consensus delay, and improves the security of data resource sharing. The simulation results verify the effectiveness of the proposed algorithm.