In this letter, we propose a feature-based knowledge distillation scheme which transfers knowledge between intermediate blocks of teacher and student with flow-based architecture, specifically Normalizing flow in our implementation. In addition to the knowledge transfer scheme, we examine how configuration of the distillation positions impacts on the knowledge transfer performance. To evaluate the proposed ideas, we choose two knowledge distillation baseline models which are based on Normalizing flow on different domains: CS-Flow for anomaly detection and SRFlow-DA for super-resolution. A set of performance comparison to the baseline models with popular benchmark datasets shows promising results along with improved inference speed. The comparison includes performance analysis based on various configurations of the distillation positions in the proposed scheme.

Involving higher-order thinking in learning activities can produce meaningful learning. It impacts the student's ability to solve problems in new situations. Concept mapping is a learning strategy that has been proven to promote higher-order thinking. Concept map recomposition (KB-mapping) in the Kit-Build system is a closed concept mapping where learners are given concepts and links to build a concept map, and it has advantage that the recomposed map can be automatically diagnosed. It has been proven that KB-mapping improves the students' learning achievement similar to the traditional concept mapping called scratch concept map composition (SC-mapping). However, the study on the effect of KB-mapping in fostering students' higher-order thinking has yet to be evaluated. This study designed and conducted an experiment to compare the impact of KB-mapping and SC-mapping on promoting students' ability in higher-order thinking. Fifty-four undergraduate students were assigned to either KB-Mapping or SC-Mapping for learning activities. The result of this study suggested that students who learn with KB-mapping had better abilities to solve questions of higher-order thinking than those who applied SC-mapping. The findings also suggested that the quality of students' concept maps affected their performance in solving higher-order thinking questions.

Binary Decision Diagrams (BDDs) are an important data structure for the design of digital circuits using VLSI CAD tools. The ordering of variables affects the total number of nodes and path length in the BDDs. Finding a good variable ordering is an optimization problem and previously many optimization approaches have been implemented for BDDs in a number of research works. In this paper, an optimization approach based on Spider Monkey Optimization (SMO) algorithm is proposed for the BDD variable ordering problem targeting number of nodes and longest path length. SMO is a well-known swarm intelligence-based optimization approach based on spider monkeys foraging behavior. The proposed work has been compared with other latest BDD reordering approaches using Particle Swarm Optimization (PSO) algorithm. The results obtained show significant improvement over the Particle Swarm Optimization method. The proposed SMO-based method is applied to different benchmark digital circuits having different levels of complexities. The node count and longest path length for the maximum number of tested circuits are found to be better in SMO than PSO.

A super-resolution algorithm is proposed to solve the problem of measuring the thin thickness of oil slick using compressed sensing theory. First, a mathematical model of a single pulse underwater ultrasonic echo is established. Then, the estimation model of the transmit time of flight (TOF) of ultrasonic echo within oil slick is given based on the sparsity of echo signals. At last, the super-resolution TOF value can be obtained by solving the sparse convex optimization problem. Simulations and experiments are conducted to validate the performance of the proposed method.

The estimation problem of structured clutter covariance matrix (CCM) in space-time adaptive processing (STAP) for airborne radar systems is studied in this letter. By employing the prior knowledge and the persymmetric covariance structure, a new estimation algorithm is proposed based on the whitening ability of the covariance matrix. The proposed algorithm is robust to prior knowledge of different accuracy, and can whiten the observed interference data to obtain the optimal solution. In addition, the extended factored approach (EFA) is used in the optimization for dimensionality reduction, which reduces the computational burden. Simulation results show that the proposed algorithm can effectively improve STAP performance even under the condition of some errors in prior knowledge.

Let f be a Boolean function in n variables. The Möbius transform and its converse of f can describe the transformation behaviors between the truth table of f and the coefficients of the monomials in the algebraic normal form representation of f. In this letter, we develop the Möbius transform and its converse into a more generalized form, which also includes the known result given by Reed in 1954. We hope that our new result can be used in the design of decoding schemes for linear codes and the cryptanalysis for symmetric cryptography. We also apply our new result to verify the basic idea of the cube attack in a very simple way, in which the cube attack is a powerful technique on the cryptanalysis for symmetric cryptography.

Fractional Frequency Reuse (FFR), which was introduced by 3GPP is considered the powerful technique to improve user performance. However, implementation of FFR is a challenge due to strong dependence between base stations (BSs) in terms of resource allocations. This paper studies a modified and flexible FFR scheme that allows all BSs works independently. The analytical and simulation results prove that the modified FFR scheme outperforms the conventional FFR.

In order to improve the anomaly detection efficiency of network traffic, firstly, the model is established for network flows based on complex networks. Aiming at the uncertainty and fuzziness between network traffic characteristics and network states, the deviation extent is measured from the normal network state using deviation interval uniformly, and the intuitionistic fuzzy sets (IFSs) are established for the various characteristics on the network model that the membership degree, non-membership degree and hesitation margin of the IFSs are used to quantify the ownership of values to be tested and the corresponding network state. Then, the knowledge measure (KM) is introduced into the intuitionistic fuzzy weighted geometry (IFWGω) to weight the results of IFSs corresponding to the same network state with different characteristics together to detect network anomaly comprehensively. Finally, experiments are carried out on different network traffic datasets to analyze the evaluation indicators of network characteristics by our method, and compare with other existing anomaly detection methods. The experimental results demonstrate that the changes of various network characteristics are inconsistent under abnormal attack, and the accuracy of anomaly detection results obtained by our method is higher, verifying our method has a better detection performance.

This paper introduces a deep reinforcement learning approach to solve the virtual network function scheduling problem in dynamic scenarios. We formulate an integer linear programming model for the problem in static scenarios. In dynamic scenarios, we define the state, action, and reward to form the learning approach. The learning agents are applied with the asynchronous advantage actor-critic algorithm. We assign a master agent and several worker agents to each network function virtualization node in the problem. The worker agents work in parallel to help the master agent make decision. We compare the introduced approach with existing approaches by applying them in simulated environments. The existing approaches include three greedy approaches, a simulated annealing approach, and an integer linear programming approach. The numerical results show that the introduced deep reinforcement learning approach improves the performance by 6-27% in our examined cases.

Autonomous driving technology is expected to be applied to various applications with unmanned vehicles (UVs), such as small delivery vehicles for office supplies and smart wheelchairs. UV remote control by a cloud server (CS) would achieve cost-effective applications with a large number of UVs. In general, dead time in real-time feedback control reduces the control accuracy. On remote path tracking control by the CS, UV control accuracy deteriorates due to transmission delay and jitter through the Internet. Digital twin computing (DTC) and jitter buffer are effective to solve this problem. In our previous study, we clarified effectiveness of them in UV remote control by CS. The jitter buffer absorbs the transmission delay jitter of control signals. This is effective to achieve accurate UV remote control. Adaptive buffering time optimization according to real-time transmission characteristics is necessary to achieve more accurate UV control in CS-based remote control system with DTC and jitter buffer. In this study, we proposed a method for the adaptive optimization according to real-time transmission delay characteristics. To quantitatively evaluate the effectiveness of the method, we created a UV remote control simulator of the control system. The results of simulations quantitatively clarify that the adaptive optimization by the proposed method improves the UV control accuracy.

Hardware oriented security and trust of semiconductor integrated circuit (IC) chips have been highly demanded. This paper outlines the requirements and recent developments in circuits and packaging systems of IC chips for security applications, with the particular emphasis on protections against physical implementation attacks. Power side channels are of undesired presence to crypto circuits once a crypto algorithm is implemented in Silicon, over power delivery networks (PDNs) on the frontside of a chip or even through the backside of a Si substrate, in the form of power voltage variation and electromagnetic wave emanation. Preventive measures have been exploited with circuit design and packaging technologies, and partly demonstrated with Si test vehicles.

This paper presents a Ka-band high-efficiency power amplifier (PA) with a novel adaptively controlled gate capacitor circuit and a two-step adaptive bias circuit for 5th generation (5G) mobile terminal applications fabricated using a 45-nm silicon on insulator (SOI) CMOS process. The PA adopts a stacked FET structure to increase the output power because of the low breakdown voltage issue of scaled MOSFETs. The novel adaptive gate capacitor circuit properly controls the RF swing for each stacked FET to achieve high efficiency in the several-dB back-off region. Further, the novel two-step adaptive bias circuit effectively controls the gate voltage for each stacked FET for high linearity and high back-off efficiency. At a supply voltage of 4 V, the fabricated PA has exhibited a saturated output power of 20.0 dBm, a peak power added efficiency (PAE) of 42.7%, a 3dB back-off efficiency of 32.7%, a 6dB back-off efficiency of 22.7%, and a gain of 15.6 dB. The effective PA area was 0.82 mm by 0.74 mm.

This paper presents a hippocampal cognitive prosthesis chip designed for restoring the ability to form new long-term memories due to hippocampal system damage. The system-on-chip (SOC) consists of a 16-channel micro-power low-noise amplifier (LNA), high-pass filters, analog-digital converters (ADCs), a 16-channel spike-sorter, a generalized Laguerre-Volterra model multi-input, multi-output (GLVM-MIMO) hippocampal processor, an 8-channel neural stimulator and peripheral circuits. The proposed LNA achieved a voltage gain of 50dB, input-referred noise of 3.95µVrms, and noise efficiency factor (NEF) of 3.45 with the power consumption of 3.3µW. High-pass filters with a 300-Hz bandwidth are used to filter out the unwanted local field potential (LFP). 4 12-bit successive approximation register (SAR) ADCs with a signal-to-noise-and-distortion ratio (SNDR) of 63.37dB are designed for the digitization of the neural signals. A 16-channel spike-sorter has been integrated in the chip enabling a detection accuracy of 98.3% and a classification accuracy of 93.4% with power consumption of 19µW/ch. The MIMO hippocampal model processor predict output spatio-temporal patterns in CA1 according to the recorded input spatio-temporal patterns in CA3. The neural stimulator performs bipolar, symmetrical charge-balanced stimulation with a maximum current of 310µA, triggered by the processor output. The chip has been fabricated in 40nm standard CMOS technology, occupying a silicon area of 3mm2.

The total number of solar power-producing facilities whose Feed-in Tariff (FIT) Program-based ten-year contracts will expire by 2023 is expected to reach approximately 1.65 million in Japan. If the facilities that produce or consume renewable energy would increase to reach a large number, e.g., two million, blockchain would not be capable of processing all the transactions. In this work, we propose a blockchain-based electricity-tracking platform for renewable energy, called ‘ZGridBC,’ which consists of mutually cooperative two novel decentralized schemes to solve scalability, storage cost, and privacy issues at the same time. One is the electricity production resource management, which is an efficient data management scheme that manages electricity production resources (EPRs) on the blockchain by using UTXO tokens extended to two-dimension (period and electricity amount) to prevent double-spending. The other is the electricity-tracking proof, which is a massive data aggregation scheme that significantly reduces the amount of data managed on the blockchain by using zero-knowledge proof (ZKP). Thereafter, we illustrate the architecture of ZGridBC, consider its scalability, security, and privacy, and illustrate the implementation of ZGridBC. Finally, we evaluate the scalability of ZGridBC, which handles two million electricity facilities with far less cost per environmental value compared with the price of the environmental value proposed by METI (=0.3 yen/kWh).

In CRYPTO 2019, Gohr first introduced the deep learning method to cryptanalysis for SPECK32/64. A differential-neural distinguisher was obtained using ResNet neural network. Zhang et al. used multiple parallel convolutional layers with different kernel sizes to capture information from multiple dimensions, thus improving the accuracy or obtaining a more round of distinguisher for SPECK32/64 and SIMON32/64. Inspired by Zhang's work, we apply the network structure to other ciphers. We not only improve the accuracy of the distinguisher, but also increase the number of rounds of the distinguisher, that is, distinguish more rounds of ciphertext and random number for DES, Chaskey and PRESENT.

Since the dark channel prior (DCP)-based dehazing method is ineffective in the sky area and will cause the problem of too dark and color distortion of the image, we propose a novel dehazing method based on sky area segmentation and image fusion. We first segment the image according to the characteristics of the sky area and non-sky area of the image, then estimate the atmospheric light and transmission map according to the DCP and correct them, and then fuse the original image after the contrast adaptive histogram equalization to improve the details information of the image. Experiments illustrate that our method performs well in dehazing and can reduce image distortion.

Single image deraining is an ill-posed problem which also has been a long-standing issue. In past few years, convolutional neural network (CNN) methods almost dominated the computer vision and achieved considerable success in image deraining. Recently the Swin Transformer-based model also showed impressive performance, even surpassed the CNN-based methods and became the state-of-the-art on high-level vision tasks. Therefore, we attempt to introduce Swin Transformer to deraining tasks. In this paper, we propose a deraining model with two sub-networks. The first sub-network includes two branches. Rain Recognition Network is a Unet with the Swin Transformer layer, which works as preliminarily restoring the background especially for the location where rain streaks appear. Detail Complement Network can extract the background detail beneath the rain streak. The second sub-network which called Refine-Unet utilizes the output of the previous one to further restore the image. Through experiments, our network achieves improvements on single image deraining compared with the previous Transformer research.

Convolutional Neural Networks (CNNs) have shown remarkable performance in image recognition tasks. In this letter, we propose a new CNN model called the EnsNet which is composed of one base CNN and multiple Fully Connected SubNetworks (FCSNs). In this model, the set of feature maps generated by the last convolutional layer in the base CNN is divided along channels into disjoint subsets, and these subsets are assigned to the FCSNs. Each of the FCSNs is trained independent of others so that it can predict the class label of each feature map in the subset assigned to it. The output of the overall model is determined by majority vote of the base CNN and the FCSNs. Experimental results using the MNIST, Fashion-MNIST and CIFAR-10 datasets show that the proposed approach further improves the performance of CNNs. In particular, an EnsNet achieves a state-of-the-art error rate of 0.16% on MNIST.

Speech emotion recognition (SER) has been a complex and difficult task for a long time due to emotional complexity. In this paper, we propose a multitask deep learning approach based on cascaded attention network and self-adaption loss for SER. First, non-personalized features are extracted to represent the process of emotion change while reducing external variables' influence. Second, to highlight salient speech emotion features, a cascade attention network is proposed, where spatial temporal attention can effectively locate the regions of speech that express emotion, while self-attention reduces the dependence on external information. Finally, the influence brought by the differences in gender and human perception of external information is alleviated by using a multitask learning strategy, where a self-adaption loss is introduced to determine the weights of different tasks dynamically. Experimental results on IEMOCAP dataset demonstrate that our method gains an absolute improvement of 1.97% and 0.91% over state-of-the-art strategies in terms of weighted accuracy (WA) and unweighted accuracy (UA), respectively.

The application of time-series prediction is very extensive, and it is an important problem across many fields, such as stock prediction, sales prediction, and loan prediction and so on, which play a great value in production and life. It requires that the model can effectively capture the long-term feature dependence between the output and input. Recent studies show that Transformer can improve the prediction ability of time-series. However, Transformer has some problems that make it unable to be directly applied to time-series prediction, such as: (1) Local agnosticism: Self-attention in Transformer is not sensitive to short-term feature dependence, which leads to model anomalies in time-series; (2) Memory bottleneck: The spatial complexity of regular transformation increases twice with the sequence length, making direct modeling of long time-series infeasible. In order to solve these problems, this paper designs an efficient model for long time-series prediction. It is a double pyramid bidirectional feature fusion mechanism network with parallel Temporal Convolution Network (TCN) and FastFormer. This network structure can combine the time series fine-grained information captured by the Temporal Convolution Network with the global interactive information captured by FastFormer, it can well handle the time series prediction problem.