MmWave imaging assisted by metasurfaces is a burgeoning technique attributed to its fine-grained imaging ability by time-varying phase coding, which produces a large virtual aperture. To obtain stereoscopic 3D perception, multiple metasurfaces can be utilized for virtual multiview fusion, allowing for the capture of features that are not visible from a single viewpoint. However, the multipath imaging fusion faces huge data burden and the environment clutter especially reflection from the direct path will cause disturbance. To address these issues, this paper introduces Bayesian compressive sensing to focus on the region of interest (ROI) and design a double sparse prior for high-resolution multiview image reconstruction. First, multiple metasurfaces are utilized to generate virtual multiview imaging results. Then, the Bayesian inference method is leveraged to resist environmental noise and achieve autofocusing imaging with undersampled data. The expectation propagation (EP) is introduced to estimate the statistical parameter iteratively. Further, a double sparse prior is designed based on spike-and-slab to promote inter-sparsity and intra-sparsity simultaneously. Simulation results show that our proposed system demonstrates superior performance by fusing sparse images from multiple metasurfaces arrangements.

Metasurface-aided near-field radio imaging is emerging as an essential part of the future mmWave communication systems. Adjusting the metasurface phase shift to generate multiple measurements can significantly increase the system imaging aperture and enhance the resolution. However, this imaging technique relies heavily on precise measurements, and high-precision sampling leads to expensive hardware costs and memory burdens. To address this issue, this paper draws inspiration from binary compressive sensing and proposes a method for imaging under one-bit quantization. First, we propose a dynamic metasurface-aided mmWave imaging system with one-bit sampling, which simplifies the received signal acquisition process and significantly reduces the hardware and storage requirements. Then, a fused binary compressive sensing model is developed with an additional total-variation norm penalty to promote target continuity and suppress artifacts in mmWave images. Subsequently, we employ the proximal binary iterative hard thresholding algorithm to optimize the joint sparsity and the total variation (TV) constraints. In addition, the hybrid ℓ1-TV constraint is introduced to solve the problem of unknown a priori sparsity of image, and the alternating direction multiplication method is designed for effective reconstruction. Finally, the simulation results show that the proposed algorithms can utilize the target features under binary measurements and achieve better imaging accuracy and focusing performance than the sparsity constraint-only methods.

Network pruning is a widely studied technique of obtaining compact representations from over-parameterized deep convolutional neural networks . Existing pruning methods are based on finding an optimal combination of pruned filters in the fixed search space . However, the optimality of those methods is often questionable due to limited search space and pruning choices - e.g., the difficulty with removing the entire layer and the risk of unexpected performance degradation . Inspired by the exploration vs. exploitation trade-off in reinforcement learning, we propose to reconstruct the filter space without increasing the model capacity and prune them by exploiting group sparsity . Our approach challenges the conventional wisdom by advocating the strategy of Growing-before-Pruning (GbP), which allows us to explore more space before exploiting the power of architecture search. Meanwhile, to achieve more efficient pruning, we propose to measure the importance of filters by global group sparsity , which extends the existing Gaussian scale mixture model. Such global characterization of sparsity in the filter space leads to a novel deterministic annealing strategy for progressively pruning the filters. We have evaluated our method on several popular datasets and network architectures. Our extensive experiment results have shown that the proposed method advances the current state-of-the-art.

We study the problem of Non-line-of-sight (NLOS) mmWave imaging under inaccurate knowledge of multiangle relay surfaces. To this end, we propose a novel double sparse structure-enhanced Bayesian compressive sensing framework with dictionary parameters updating. First, the hierarchical probabilistic model with a parametric multipath dictionary is constructed, where the angles of multiple relay surface are considered as an unknown parameter. Then, a double sparse spike-and-slab (DS-SS) prior is introduced to model the intra-group and inter-group sparsity of multipath image, where the expectation propagation method is employed for posterior inference (dubbed as DS-SSEP). Moreover, the maximum likelihood solutions of the dictionary parameters are estimated iteratively by coupling the expectation maximization framework with DS-SSEP. Several experiments demonstrate the superiority of our proposed method, which significantly reduces image reconstruction errors in imprecise layout scenarios.

This article presents a 10-bit radiation-hardened-by-design (RHBD) SAR analog-to-digital converter (ADC) operating at 50 MS/s, designed for aerospace applications in high-radiation environments. The system- and circuit-level redundancy techniques are implemented to mitigate radiation-induced errors and metastability. A novel split coarse/fine asynchronous SAR ADC architecture is proposed to provide system-level redundancy. At circuits level, single-event effects (SEEs) error detection and radiation-hardened techniques are implemented. Our co-designed SEE error detection scheme includes last-bit-cycle (LBC) detection following the LSB cycle and metastability detection (MD) via a ramp generator with a threshold trigger. This approach detects and corrects radiation-induced errors using a coarse/fine redundant algorithm. The radiation-hardened latch comparators and D flip-flops (DFFs) are incorporated to further mitigate SEEs. The prototype design is fabricated using TSMC 65-nm technology, with an ADC core area of 0.0875 mm2and a power consumption of 2.79 mW at a 1.2-V power supply. Postirradiation tests confirm functionality up to 100-krad(Si) total ionizing dose (TID) and demonstrate over 90% suppression of large SEE under laser testing.

Cloud computing, as a cutting-edge computing paradigm, offers substantial data processing and storage capabilities. In a heterogeneous cloud environment, the diversity among cloud platforms results in varying task execution times, posing challenges in minimizing workflow makespan under budget constraints. On this basis, a novel meta-heuristic optimization algorithm, named snake optimizer (SO), is proposed for workflow scheduling in the cloud. Then, a negative offset mechanism is designed to dynamically guide the offset of individual positions during population update to prevent falling into local optimums, thereby optimizing the search for feasible solutions and improving the success rate. Finally, using the negative offset mechanism, a snake optimization budget-aware scheduling algorithm (NO-SO) is developed to schedule budget-constrained workflows in heterogeneous cloud computing environments and minimize the makespan. A series of comparative experiments conducted on real-world scientific workflows demonstrates that the NO-SO algorithm enhances the success rate in finding a feasible solution by 38.89% and 34.45% compared with the advanced MG-PRO algorithm and the original SO algorithm, respectively. Moreover, it achieves an average reduction in makespan of 30.30% and 32.19%.

This paper aims to enhance user fairness in unmanned aerial vehicles (UAVs) enabled cell-free wireless networks. We consider a scenario where multiple UAVs serve as access points for ground users (UEs). We jointly design the UAV deployment, power allocation, UAV-UE association, and pilot assignment to maximize the minimum downlink user rate, while taking account the impact of pilot contamination. The design is formulated as a mixed-integer non-convex optimization problem. To circumvent the problem non-convexity and facilitate the design of a computationally efficient suboptimal solution, a series of transformations and approximations are proposed based on the particle swarm optimization and successive convex approximation techniques. Simulation results are presented to demonstrate the effectiveness and advantages brought by the joint design for enhancing user fairness. Furthermore, our study provides new insights into the feasibility of using sparse association to reduce costs while maintaining performance, especially in scenarios with a large number of users.

Both network pruning and neural architecture search (NAS) can be interpreted as techniques to automate the design and optimization of artificial neural networks. In this paper, we challenge the conventional wisdom of training before pruning by proposing a joint search-and-training approach to learn a compact network directly from scratch. Using pruning as a search strategy, we advocate three new insights for network engineering: 1) to formulate adaptive search as a cold start strategy to find a compact subnetwork on the coarse scale; and 2) to automatically learn the threshold for network pruning; 3) to offer flexibility to choose between efficiency and robustness. More specifically, we propose an adaptive search algorithm in the cold start by exploiting the randomness and flexibility of filter pruning. The weights associated with the network filters will be updated by ThreshNet, a flexible coarse-to-fine pruning method inspired by reinforcement learning. In addition, we introduce a robust pruning strategy leveraging the technique of knowledge distillation through a teacher-student network. Extensive experiments on ResNet and VGGNet have shown that our proposed method can achieve a better balance in terms of efficiency and accuracy and notable advantages over current state-of-the-art pruning methods in several popular datasets, including CIFAR10, CIFAR100, and ImageNet. The code associate with this paper is available at: https://see.xidian.edu.cn/faculty/wsdong/Projects/AST-NP.htm.

Filter pruning, as an effective strategy to obtain efficient compact structures from over-parametric deep neural networks(DNN), has attracted a lot of attention. Previous pruning methods select channels for pruning by developing different criteria, yet little attention has been devoted to whether these criteria can represent correlations between channels. Meanwhile, most existing methods generally ignore the parameters being pruned and only perform additional training on the retained network to reduce accuracy loss. In this paper, we present a novel perspective of re-parametric pruning by Bayesian estimation. First, we estimate the probability distribution of different channels based on Bayesian estimation and indicate the importance of the channels by the discrepancy in the distribution before and after channel pruning. Second, to minimize the variation in distribution after pruning, we re-parameterize the pruned network based on the probability distribution to pursue optimal pruning. We evaluate our approach on popular datasets with some typical network architectures, and comprehensive experimental results validate that this method illustrates better performance compared to the state-of-the-art approaches.

Network pruning has been proposed as a remedy for alleviating the over-parameterization problem of deep neural networks. However, its value has been recently challenged especially from the perspective of neural architecture search (NAS). We challenge the conventional wisdom of pruning-after-training by proposing a joint search-and-training approach that directly learns a compact network from the scratch. By treating pruning as a search strategy, we present two new insights in this paper: 1) it is possible to expand the search space of networking pruning by associating each filter with a learnable weight; 2) joint search-and-training can be conducted iteratively to maximize the learning efficiency. More specifically, we propose a coarse-to-fine tuning strategy to iteratively sample and update compact sub-network to approximate the target network. The weights associated with network filters will be accordingly updated by joint search-and-training to reflect learned knowledge in NAS space. Moreover, we introduce strategies of random perturbation (inspired by Monte Carlo) and flexible thresholding (inspired by Reinforcement Learning) to adjust the weight and size of each layer. Extensive experiments on ResNet and VGGNet demonstrate the superior performance of our proposed method on popular datasets including CIFAR10, CIFAR100 and ImageNet.

Deep neural networks (DNNs) have shown very promising results for various image restoration (IR) tasks. However, the design of network architectures remains a major challenging for achieving further improvements. While most existing DNN-based methods solve the IR problems by directly mapping low quality images to desirable high-quality images, the observation models characterizing the image degradation processes have been largely ignored. In this paper, we first propose a denoising-based IR algorithm, whose iterative steps can be computed efficiently. Then, the iterative process is unfolded into a deep neural network, which is composed of multiple denoisers modules interleaved with back-projection (BP) modules that ensure the observation consistencies. A convolutional neural network (CNN) based denoiser that can exploit the multi-scale redundancies of natural images is proposed. As such, the proposed network not only exploits the powerful denoising ability of DNNs, but also leverages the prior of the observation model. Through end-to-end training, both the denoisers and the BP modules can be jointly optimized. Experimental results on several IR tasks, e.g., image denoisig, super-resolution and deblurring show that the proposed method can lead to very competitive and often state-of-the-art results on several IR tasks, including image denoising, deblurring, and super-resolution.