A Josephson parametric oscillator (JPO) is an interesting system from the viewpoint of quantum optics because it has two stable self-oscillating states and can deterministically generate quantum cat states. A theoretical proposal has been made to operate a network of multiple JPOs as a quantum annealer, which can solve adiabatically combinatorial optimization problems at high speed. Proof-of-concept experiments have been actively conducted for application to quantum computations. This article provides a review of the mechanism of JPOs and their application as a quantum annealer.

This paper addresses single image super-resolution (SR) based on convolutional neural networks (CNNs). It is known that recovery of high-frequency components in output SR images of CNNs learned by the least square errors or least absolute errors is insufficient. To generate realistic high-frequency components, SR methods using generative adversarial networks (GANs), composed of one generator and one discriminator, are developed. However, when the generator tries to induce the discriminator's misjudgment, not only realistic high-frequency components but also some artifacts are generated, and objective indices such as PSNR decrease. To reduce the artifacts in the GAN-based SR methods, we consider the set of all SR images whose square errors between downscaling results and the input image are within a certain range, and propose to apply the metric projection onto this consistent set in the output layers of the generators. The proposed technique guarantees the consistency between output SR images and input images, and the generators with the proposed projection can generate high-frequency components with few artifacts while keeping low-frequency ones as appropriate for the known noise level. Numerical experiments show that the proposed technique reduces artifacts included in the original SR images of a GAN-based SR method while generating realistic high-frequency components with better PSNR values in both noise-free and noisy situations. Since the proposed technique can be integrated into various generators if the downscaling process is known, we can give the consistency to existing methods with the input images without degrading other SR performance.

In this paper, we study Hermitian linear complementary dual (abbreviated Hermitian LCD) rank metric codes. A class of Hermitian LCD generalized Gabidulin codes are constructed by qm-self-dual bases of Fq2m over Fq2. Moreover, the exact number of qm-self-dual bases of Fq2m over Fq2 is derived. As a consequence, an upper bound and a lower bound of the number of the constructed Hermitian LCD generalized Gabidulin codes are determined.

Local discriminative regions play important roles in fine-grained image analysis tasks. How to locate local discriminative regions with only category label and learn discriminative representation from these regions have been hot spots. In our work, we propose Searching Discriminative Regions (SDR) and Learning Discriminative Regions (LDR) method to search and learn local discriminative regions in images. The SDR method adopts attention mechanism to iteratively search for high-response regions in images, and uses this as a clue to locate local discriminative regions. Moreover, the LDR method is proposed to learn compact within category and sparse between categories representation from the raw image and local images. Experimental results show that our proposed approach achieves excellent performance in both fine-grained image retrieval and classification tasks, which demonstrates its effectiveness.

This paper demonstrates that a 360° radio-frequency phase detector consisting of a combination of symmetrical mixers and 45° phase shifters with tunable devices can achieve a low phase-detection error over a wide frequency range. It is shown that the phase detection error does not depend on the voltage gain of the 45° phase shifter. This allows the usage of tunable devices as 45° phase shifters for a wide frequency range with low phase-detection errors. The fabricated phase detector having tunable low-pass filters as the tunable device demonstrates phase detection errors lower than 2.0° rms in the frequency range from 3.0 GHz to 10.5 GHz.

In this paper, the high-power high-efficiency asymmetric Doherty power amplifiers based on high-voltage GaN HEMT devices with internal input matching for base station applications are proposed and described. For a three-way 1:2 asymmetric Doherty structures, an exceptionally high output power of 1 kW with a peak efficiency of 83% and a linear flat power gain of about 15 dB was achieved in a frequency band of 2.11-2.17 GHz, whereas an output power of 59.5 dBm with a peak efficiency of 78% and linear power gain of 12 dB and an output power of 59.2 dBm with a peak efficiency of 65% and a linear power gain of 13 dB were obtained across 1.8-2.2 GHz. To provide a high-efficiency broadband operation, the concept of inverted Doherty structure is applied and described in detail. By using a high-power broadband inverted Doherty amplifier architecture with a 2×120-W GaN HEMT transistor, a saturated power of greater than 54 dBm, a linear power gain of greater than 13 dB and a drain efficiency of greater than 50% at 7-dB power backoff in a frequency bandwidth of 1.8-2.7 GHz were obtained.

A practical Doherty amplifier design method has been developed based on an asymmetric configuration scheme. By embedding a load modulation function into matching circuits of a carrier amplifier (CA) and a peaking amplifier (PA) in the Doherty amplifier, an issue of the Doherty amplifier design is boiled down to the CA and PA matching circuit design. The method can be applied to transistors with unknown parasitic elements if optimum termination impedance conditions for the transistor are obtained from a source-/load-pull technique in simulation or measurement. The design method was applied to GaN HEMT Doherty amplifier MMICs. The fabricated 4.5-GHz-band GaN HEMT Doherty amplifier MMIC exhibited a maximum drain efficiency of 66% and a maximum power-added efficiency (PAE) of 62% at 4.1GHz, with a saturation output power of 36dBm. In addition, PAE of 50% was achieved at 4.1GHz on a 7.2-dB output back-off (OBO) condition. The fabricated 8.5-GHz-band GaN HEMT Doherty amplifier MMIC exhibited a maximum drain efficiency of 53% and a maximum PAE of 44% at 8.6GHz, with a saturation output power of 36dBm. In addition, PAE of 35% was achieved at 8.6GHz on a 6.7-dB (OBO). And, the fabricated 12-GHz-band GaN HEMT Doherty amplifier MMIC exhibited a maximum drain efficiency of 57% and a maximum PAE of 52% at 12.4GHz, with a saturation output power of 34dBm. In addition, PAE of 32% was achieved at 12.4GHz on a 9.5-dB (OBO) condition.

Contamination of water resources with pathogenic microorganisms excreted in human feces is a worldwide public health concern. Surveillance of fecal contamination is commonly performed by routine monitoring for a single type or a few types of microorganism(s). To design a feasible routine for periodic monitoring and to control risks of exposure to pathogens, reliable statistical algorithms for inferring correlations between concentrations of microorganisms in water need to be established. Moreover, because pathogens are often present in low concentrations, some contaminations are likely to be under a detection limit. This yields a pairwise left-censored dataset and complicates computation of correlation coefficients. Errors of correlation estimation can be smaller if undetected values are imputed better. To obtain better imputations, we utilize side information and develop a new technique, the asymmetric Tobit model which is an extension of the Tobit model so that domain knowledge can be exploited effectively when fitting the model to a censored dataset. The empirical results demonstrate that imputation with domain knowledge is effective for this task.

One of the main problems in list decoding is to determine the tradeoff between the list decoding radius and the rate of the codes w.r.t. a given metric. In this paper, we first describe a “stronger-weaker” relationship between two distinct metrics of the same code, then we show that the list decodability of the stronger metric can be deduced from the weaker metric directly. In particular, when we focus on matrix codes, we can obtain list decodability of matrix code w.r.t. the cover metric from the Hamming metric and the rank metric. Moreover, we deduce a Johnson-like bound of the list decoding radius for cover metric codes, which improved the result of [20]. In addition, the condition for a metric that whether the list decoding radius w.r.t. this metric and the rate are bounded by the Singleton bound is presented.

This paper presents the average block error rate (BLER) performance of circular 32QAM and 64QAM schemes employing a frequency domain equalizer (FDE) for discrete Fourier transform (DFT)-precoded orthogonal frequency division multiplexing (OFDM) in multipath Rayleigh fading channels. The circular QAM scheme has an advantageous feature in that the fluctuation in the amplitude component is smaller than that for the cross or rectangular QAM scheme. Hence, focusing on the actual received signal-to-noise power ratio (SNR) taking into account a realistic peak-to-average power ratio (PAPR) measure called the cubic metric (CM), we compare the average BLER of the circular 32QAM and 64QAM schemes with those of cross 32QAM and rectangular 64QAM schemes, respectively. We investigate the theoretical throughput of various circular 32QAM and 64QAM schemes based on mutual information from the viewpoint of the minimum Euclidean distance. Link-level simulation results show that the circular 32QAM and 64QAM schemes with independent bit mapping for the phase and amplitude modulations achieves a lower required average received SNR considering the CM than that with the minimum Euclidean distance but with composite mapping of the phase and amplitude modulations. Through extensive link-level simulations, we show the potential benefit of the circular 32QAM and 64QAM schemes in terms of reducing the required average received SNR considering the CM that satisfies the target average BLER compared to the cross 32QAM or rectangular 64QAM scheme.

We provide a formal proof for the indifferentiability of SKINNY-HASH internal function from a random oracle. SKINNY-HASH is a family of sponge-based hash functions that use functions (instead of permutations) as primitives, and it was selected as one of the second round candidates of the NIST lightweight cryptography competition. Its internal function is constructed from the tweakable block cipher SKINNY. The construction of the internal function is very simple and the designers claim n-bit security, where n is the block length of SKINNY. However, a formal security proof of this claim is not given in the original specification of SKINNY-HASH. In this paper, we formally prove that the internal function of SKINNY-HASH has n-bit security, i.e., it is indifferentiable from a random oracle up to O(2n) queries, substantiating the security claim of the designers.

Accurate and automatic quantitative cephalometry analysis is of great importance in orthodontics. The fundamental step for cephalometry analysis is to annotate anatomic-interested landmarks on X-ray images. Computer-aided automatic method remains to be an open topic nowadays. In this paper, we propose an efficient deep learning-based coarse-to-fine approach to realize accurate landmark detection. In the coarse detection step, we train a deep learning-based deformable transformation model by using training samples. We register test images to the reference image (one training image) using the trained model to predict coarse landmarks' locations on test images. Thus, regions of interest (ROIs) which include landmarks can be located. In the fine detection step, we utilize trained deep convolutional neural networks (CNNs), to detect landmarks in ROI patches. For each landmark, there is one corresponding neural network, which directly does regression to the landmark's coordinates. The fine step can be considered as a refinement or fine-tuning step based on the coarse detection step. We validated the proposed method on public dataset from 2015 International Symposium on Biomedical Imaging (ISBI) grand challenge. Compared with the state-of-the-art method, we not only achieved the comparable detection accuracy (the mean radial error is about 1.0-1.6mm), but also largely shortened the computation time (4 seconds per image).

Content-based image retrieval has been a hot topic among computer vision researchers for a long time. There have been many advances over the years, one of the recent ones being deep metric learning, inspired by the success of deep neural networks in many machine learning tasks. The goal of metric learning is to extract good high-level features from image pixel data using neural networks. These features provide useful abstractions, which can enable algorithms to perform visual comparison between images with human-like accuracy. To learn these features, supervised information of image similarity or relative similarity is often used. One important issue in deep metric learning is how to define similarity for multi-label or multi-object scenes in images. Traditionally, pairwise similarity is defined based on the presence of a single common label between two images. However, this definition is very coarse and not suitable for multi-label or multi-object data. Another common mistake is to completely ignore the multiplicity of objects in images, hence ignoring the multi-object facet of certain types of datasets. In our work, we propose an approach for learning deep image representations based on the relative similarity of both multi-label and multi-object image data. We introduce an intuitive and effective similarity metric based on the Jaccard similarity coefficient, which is equivalent to the intersection over union of two label sets. Hence we treat similarity as a continuous, as opposed to discrete quantity. We incorporate this similarity metric into a triplet loss with an adaptive margin, and achieve good mean average precision on image retrieval tasks. We further show, using a recently proposed quantization method, that the resulting deep feature can be quantized whilst preserving similarity. We also show that our proposed similarity metric performs better for multi-object images than a previously proposed cosine similarity-based metric. Our proposed method outperforms several state-of-the-art methods on two benchmark datasets.

Deep learning is playing an increasingly important role in signal processing field due to its excellent performance on many inference problems. Parametric bilinear generalized approximate message passing (P-BiG-AMP) is a new approximate message passing based approach to a general class of structure-matrix bilinear estimation problems. In this letter, we propose a novel feed-forward neural network architecture to realize P-BiG-AMP methodology with deep learning for the inference problem of compressive sensing under matrix uncertainty. Linear transforms utilized in the recovery process and parameters involved in the input and output channels of measurement are jointly learned from training data. Simulation results show that the trained P-BiG-AMP network can achieve higher reconstruction performance than the P-BiG-AMP algorithm with parameters tuned via the expectation-maximization method.

The plain bend in a pair of differential transmission lines causes a path difference, which leads to differential-to-common mode conversion due to the phase difference. This conversion can cause serious common-mode noise issues. We previously proposed a tightly coupled asymmetrically tapered bend to suppress forward differential-to-common mode conversion and derived the constraint conditions for high-density wiring. To provide sufficient suppression of mode conversion, however, the additional correction was required to make the effective path difference vanish. This paper proposes a practical and straightforward design methodology by using a very tightly coupled bend (decreasing the line width and the line separation of the tightly coupled bend). Full-wave simulations below 20GHz demonstrated that sufficient suppression of the forward differential-to-common mode conversion is successfully achieved as designed. Measurements showed that our design methodology is effective.

This paper describes a method to control prosodic features using phonetic and prosodic symbols as input of attention-based sequence-to-sequence (seq2seq) acoustic modeling (AM) for neural text-to-speech (TTS). The method involves inserting a sequence of prosodic symbols between phonetic symbols that are then used to reproduce prosodic acoustic features, i.e. accents, pauses, accent breaks, and sentence endings, in several seq2seq AM methods. The proposed phonetic and prosodic labels have simple descriptions and a low production cost. By contrast, the labels of conventional statistical parametric speech synthesis methods are complicated, and the cost of time alignments such as aligning the boundaries of phonemes is high. The proposed method does not need the boundary positions of phonemes. We propose an automatic conversion method for conventional labels and show how to automatically reproduce pitch accents and phonemes. The results of objective and subjective evaluations show the effectiveness of our method.

In this study we propose a method to perform personal identification (PI) based on Electroencephalogram (EEG) signals, where the used network is named residual and multiscale spatio-temporal convolution neural network (RAMST-CNN). Combined with some popular techniques in deep learning, including residual learning (RL), multi-scale grouping convolution (MGC), global average pooling (GAP) and batch normalization (BN), RAMST-CNN has powerful spatio-temporal feature extraction ability as it achieves task-independence that avoids the complexity of selecting and extracting features manually. Experiments were carried out on multiple datasets, the results of which were compared with methods from other studies. The results show that the proposed method has a higher recognition accuracy even though the network it is based on is lightweight.

In this paper, a compact microwave push-push oscillator based on a resonant tunneling diode (RTD) has been fabricated and demonstrated. A symmetrical spiral inductor structure has been used in order to reduce a chip area. The designed symmetric inductor is integrated into the InP-based RTD monolithic microwave integrated circuit (MMIC) technology. The circuit occupies a compact active area of 0.088 mm2 by employing symmetric inductor. The fabricated RTD oscillator shows an extremely low DC power consumption of 87 µW at an applied voltage of 0.47 V with good figure-of-merit (FOM) of -191 dBc/Hz at an oscillation frequency of 27 GHz. This is the first implementation as the RTD push-push oscillator with the symmetrical spiral inductor.

In this study, we investigate fundamental trade-off among identification, secrecy, template, and privacy-leakage rates in biometric identification system. Ignatenko and Willems (2015) studied this system assuming that the channel in the enrollment process of the system is noiseless and they did not consider the template rate. In the enrollment process, however, it is highly considered that noise occurs when bio-data is scanned. In this paper, we impose a noisy channel in the enrollment process and characterize the capacity region of the rate tuples. The capacity region is proved by a novel technique via two auxiliary random variables, which has never been seen in previous studies. As special cases, the obtained result shows that the characterization reduces to the one given by Ignatenko and Willems (2015) where the enrollment channel is noiseless and there is no constraint on the template rate, and it also coincides with the result derived by Günlü and Kramer (2018) where there is only one individual.

This paper proposes an id-eCK secure identity-based authenticated key exchange (ID-AKE) scheme, where the id-eCK security implies that a scheme resists against leakage of all combinations of master, static, and ephemeral secret keys except ones trivially break the security. Most existing id-eCK secure ID-AKE schemes require two symmetric pairing operations or a greater number of asymmetric pairing, which is faster than symmetric one, operations to establish a session key. However, our scheme is realized with a single asymmetric pairing operation for each party, and this is an advantage in efficiency. The proposed scheme is based on the ID-AKE scheme by McCullagh and Barreto, which is vulnerable to an active attack. To achieve id-eCK security, we apply the HMQV construction and the NAXOS technique to the McCullagh-Barreto scheme. The id-eCK security is proved under the external Diffie-Hellman for target group assumption and the q-gap-bilinear collision attack assumption.