By mapping iterative optimization algorithms into neural networks (NNs), deep unfolding networks (DUNs) exhibit well-defined and interpretable structures and achieve remarkable success in the field of compressive sens...
详细信息
By mapping iterative optimization algorithms into neural networks (NNs), deep unfolding networks (DUNs) exhibit well-defined and interpretable structures and achieve remarkable success in the field of compressive sensing (CS). However, most existing DUNs solely rely on the image-domain unfolding, which restricts the information transmission capacity and reconstruction flexibility, leading to their loss of image details and unsatisfactory performance. To overcome these limitations, this paper develops a dual-domain optimization framework that combines the priors of (1) image- and (2) convolutional-coding-domains and offers generality to CS and other inverse imaging tasks. By converting this optimization framework into deep NN structures, we present a Dual-Domain Deep convolutional coding Network ((DC2)-C-3-Net), which enjoys the ability to efficiently transmit high-capacity self-adaptive convolutional features across all its unfolded stages. Our theoretical analyses and experiments on simulated and real captured data, covering 2D and 3D natural, medical, and scientific signals, demonstrate the effectiveness, practicality, superior performance, and generalization ability of our method over other competing approaches and its significant potential in achieving a balance among accuracy, complexity, and interpretability. Code is available at https://***/lwq20020127/D3C2-Net.
Background and objective: Low-dose computed tomography (LDCT) has become increasingly important for alleviating X-ray radiation damage. However, reducing the administered radiation dose may lead to degraded CT images ...
详细信息
Background and objective: Low-dose computed tomography (LDCT) has become increasingly important for alleviating X-ray radiation damage. However, reducing the administered radiation dose may lead to degraded CT images with amplified mottle noise and nonstationary streak artifacts. Previous studies have confirmed that deep learning (DL) is promising for improving LDCT imaging. However, most DL-based frameworks are built intuitively, lack interpretability, and suffer from image detail information loss, which has become a general challenging issue. Methods: A multiscale reweighted convolutional coding neural network (MRCON-Net) is developed to address the above problems. MRCON-Net is compact and more explainable than other networks. First, inspired by the learning-based reweighted iterative soft thresholding algorithm (ISTA), we extend traditional convolutional sparse coding (CSC) to its reweighted convolutional learning form. Second, we use dilated convolution to extract multiscale image features, allowing our single model to capture the correlations between features of different scales. Finally, to automatically adjust the elements in the feature code to correct the obtained solution, a channel attention (CA) mechanism is utilized to learn appropriate weights. Results: The visual results obtained based on the American Association of Physicians in Medicine (AAPM) Challenge and United Image Healthcare (UIH) clinical datasets confirm that the proposed model significantly reduces serious artifact noise while retaining the desired structures. Quantitative results show that the average structural similarity index measurement (SSIM) and peak signal-to-noise ratio (PSNR) achieved on the AAPM Challenge dataset are 0.9491 and 40.66, respectively, and the SSIM and PSNR achieved on the UIH clinical dataset are 0.915 and 42.44, respectively;these are promising quantitative results. Conclusion: Compared with recent state-of-the-art methods, the proposed model achieves subtle struct
Distributed matrix computations - matrix-matrix or matrix-vector multiplications - are well-recognized to suffer from the problem of stragglers (slow or failed worker nodes). Much of prior work in this area is (i) eit...
详细信息
Distributed matrix computations - matrix-matrix or matrix-vector multiplications - are well-recognized to suffer from the problem of stragglers (slow or failed worker nodes). Much of prior work in this area is (i) either sub-optimal in terms of its straggler resilience, or (ii) suffers from numerical problems, i.e., there is a blow-up of round-off errors in the decoded result owing to the high condition numbers of the corresponding decoding matrices. Our work presents a convolutional coding approach to this problem that removes these limitations. It is optimal in terms of its straggler resilience, and has excellent numerical robustness as long as the workers' storage capacity is slightly higher than the fundamental lower bound. Moreover, it can be decoded using a fast peeling decoder that only involves add/subtract operations. Our second approach has marginally higher decoding complexity than the first one, but allows us to operate arbitrarily close to the storage capacity lower bound. Its numerical robustness can be theoretically quantified by deriving a computable upper bound on the worst case condition number over all possible decoding matrices by drawing connections with the properties of large block Toeplitz matrices. All above claims are backed up by extensive experiments done on the AWS cloud platform.
In this paper, we propose a noise-robust online convolutional coding model for image representation, which can use the noisy images as training data. Then an alternating algorithm is utilized to convert the model into...
详细信息
In this paper, we propose a noise-robust online convolutional coding model for image representation, which can use the noisy images as training data. Then an alternating algorithm is utilized to convert the model into two sub-problems, the image pursuit problem and the dictionary learning problem. For the image pursuit problem, the Gauss elimination method is used to solve the equation set which is derived by the Euler equation and discrete Fourier transform. For the dictionary learning problem, a gradient-descent flow is derived to solve it. Experimental results show that our method can output more meaningful feature representations compared to the related models while the training data was corrupted by Poisson noise. (c) 2021 Elsevier Inc. All rights reserved.
In this paper, a new power control system for wind turbines based on a squirrel cage induction generator linked to the mains by means of a back-to-back converter for smart grid applications is proposed. The wireless s...
详细信息
In this paper, a new power control system for wind turbines based on a squirrel cage induction generator linked to the mains by means of a back-to-back converter for smart grid applications is proposed. The wireless system employs orthogonal frequency division multiplexing, convolutional coding, and functionally weighted moving average filtering to improve the system performance against the errors due to the propagation channel in the transmitted power references. Hence, the system avoids damages in turbines and converters and also increases the power quality of the power injected into the grid. Hence, the reactive power injection can be helpful in ancillary services. The system was analyzed regarding the total harmonic distortion for SCIGs' currents and the system response to verifying the delivered power quality by the generator to the mains. The simulated results validate the proposed wireless-coded control system.
In practical applications of electromagnetic measurement while drilling (EM-MWD) in the underground coal mine, the signal-to-noise ratio (SNR) of the receiver cannot always meet the requirements of reliable communicat...
详细信息
ISBN:
(纸本)9781538634608
In practical applications of electromagnetic measurement while drilling (EM-MWD) in the underground coal mine, the signal-to-noise ratio (SNR) of the receiver cannot always meet the requirements of reliable communication conditions due to the earth-attenuation, interfering signal from a well site, etc. Traditional digital communication systems use independent design coding and modulation techniques to improve system performance. The coding is mainly achieved by introducing redundant bits and the improvement of error performance is at the expense of information rate. Aimed to solve this problem, we use a coding technique based on trellis coded modulation (TCM) to maximize the minimum distance between modulated output sequences and achieve significant coding gain. Simulation and experiments show that the system can improve the anti noise performance of the system by obtaining a coding gain without reducing the transmission rate. Compared to the uncoded quadrature phase shift keying (QPSK), TCM can achieve a coding gain of at least 3 dB under the same transmitting rate. At that time, as the number of TCM system states or trace back depth increases, the coding gain is further enhanced. The TCM code modulation method can be used in an EM-MWD system to improve the system performance.
In this paper, a Generalized Space Shift Keying modulation scheme combined with convolutional coding is proposed, and the convolutional coding structure is given. Firstly, the transmission model and the convolutional ...
详细信息
ISBN:
(纸本)9789811065712;9789811065705
In this paper, a Generalized Space Shift Keying modulation scheme combined with convolutional coding is proposed, and the convolutional coding structure is given. Firstly, the transmission model and the convolutional coding structure of the system are introduced. Secondly, the decoding algorithm of antenna detection used in the receiver is analyzed, and the complexity of the receiving algorithm is compared. Finally, the transmission performance of the scheme is simulated and analyzed in the Rician fading channel, and the advantages and disadvantages of the scheme, traditional Space Shift Keying modulation and Generalized Space Shift Keying modulation are compared. The results show that the proposed scheme can prove the anti-channel fading performance of the system.
Image denoising is the foundation of computer vision and a classic research project of computer vision. Currently, image denoising is widely used in various fields. In order to perform image denoising more effectively...
详细信息
ISBN:
(数字)9781728176475
ISBN:
(纸本)9781728176482
Image denoising is the foundation of computer vision and a classic research project of computer vision. Currently, image denoising is widely used in various fields. In order to perform image denoising more effectively, we propose a deep convolutional sparse coding based on the attention mechanism (DCSCA) denoising model. In order to identify noise and remove noise more effectively, we sparsely encode the feature map instead of encoding the original image. Our model introduces attention blocks in the network structure, which allows our model to mine the noise hidden behind complex images. Our model has the advantages of a compact network structure and does not involve complex optimization models. We did not use the traditional mean square error (MSE) as our loss function, and chose to combine the loss function L1 with ssim. Compared with the most advanced methods, our model can produce better denoising results than some of the latest methods.
Iterative decoding of orthogonal convolutional code is widely used for its excellent performance. However, the iterations lead to high complexity and long decoding delay, which is unsuitable for low-rate voice service...
详细信息
ISBN:
(纸本)9781479980925
Iterative decoding of orthogonal convolutional code is widely used for its excellent performance. However, the iterations lead to high complexity and long decoding delay, which is unsuitable for low-rate voice service in mobile satellite communications with on-board processing. In this paper, a novel scheme of orthogonal convolutional coding as well as an associated non-iterative joint decoding algorithm based on factor graph are proposed. Due to its non-iterative nature, this novel scheme has low decoding complexity and short latency, which indicates its potential on-board use in the low-rate satellite voice service, or other kinds of services with a high bit error rate (BER) tolerance and a tight delay requirement. Simulation results demonstrate the efficacy of the proposed novel coding scheme and non-iterative decoding algorithm in both AWGN and Rician channels.
In recent years, with the advent of 6G and intelligent devices, sensors, and new applications such as virtual reality and autonomous driving, user data traffic has exploded, especially video traffic and small IoT pack...
详细信息
In recent years, with the advent of 6G and intelligent devices, sensors, and new applications such as virtual reality and autonomous driving, user data traffic has exploded, especially video traffic and small IoT packets. These bandwidth-hungry applications require an increased network capacity and user access. The 6G network may use nonorthogonal multiple access (NOMA) instead of orthogonal multiple access (OMA) to maintain higher data rates, throughput, and lower latency. On the other hand, choosing the channel coding method for future 6G mobile communication is critical for maintaining the high demand for 6G. This paper proposes two-channel coding structures to achieve higher data rates with a lower error rate floor;these structures are polar convolutional serial code (PCSC) and polar convolutional parallel code (PCPC);these structures can achieve a larger channel capacity and reduced bit error rates when used with NOMA. The obtained simulation results showed that bit error rate (BER) performance improves the overall coding gain by 1.2 dB compared to polar code in fifth-generation (5G). PCSC surpasses PCPC with a 1.5dB coding gain. This performance ranged from 4 to 6.25dB with higher system settings. The obtained throughput results showed an improvement of 56-60%, in which the enhancement percentage depended on the modulation method used in a direct proportion manner.
暂无评论