Multifunctional bionic devices have widespread applications in neuromorphic computing, intelligent sensors, and robotics. The inherent properties of memristors make them suitable for these emerging applications, but different require either volatile or nonvolatile operations a unique device. In this work, we developed novel reconfigurable Ag/Al2O3/ITO memristor, which achieves adjustable switching behavior between by modulating the compliance current. A proposed mechanism controls state...

Multiple-input multiple-output-synthetic aperture radar (MIMO-SAR) can realize high-resolution imaging by its predominance of parallel sampling. However, the motion compensation and large data amount are inevitable problems to be solved. In this paper, a novel compensating method for MIMO-SAR with under-sampled echo signal is put forward. First, model system error analyzed Nyquist-sampled proposed. Using technique two-step compensation, compensated result obtained. Second, compensate signal,...

Raptor codes are state-of-the-art forward error correction (FEC) solutions for multimedia transmission, which have been applied to unequal protection (UEP) of multi-layered media such as scalable video coding. In this paper, we address the problem UEP single-layered over packet erasure channels. By exploiting different priorities packets inside a group pictures (GOP) and making full use good characteristics standardized at large block length, propose an optimized framework develop efficient...

Semantically searching and navigating products (e.g., on Taobao.com or Amazon.com) with professional metadata user-generated content from social media is a hot topic in information retrieval recommendation systems, while most existing methods are specifically designed as purely system. In this paper, taking Social Book Search an example, we propose general search-recommendation hybrid system for topic. Firstly, Generalized Content-Based Filtering (GCF) model. model, preference value, which...

In recent years, comprehensive learning particle swarm optimization (CLPSO) has attracted the attention of many scholars for using in solving multimodal problems, as it is excellent preserving particles' diversity and thus preventing premature convergence. However, CLPSO exhibits low solution accuracy. Aiming to address this issue, we proposed a novel algorithm called LILPSO. First, introduced Lagrange interpolation method perform local search global best point (gbest). Second, gain better...

Elephant herding optimization (EHO) has received widespread attention due to its few control parameters and simple operation but still suffers from slow convergence low solution accuracy. In this paper, an improved algorithm solve the above shortcomings, called Gaussian perturbation specular reflection learning golden‐sine‐mechanism‐based EHO (SRGS‐EHO), is proposed. First, introduced into enhance diversity ergodicity of initial population improve speed. Meanwhile, used further increase...

In this paper, we study performance optimization for scalable video coding and multicast over networks. Multi-path streaming, network based routing, flow control are jointly optimized to maximize a utility function defined heterogeneous receivers. Content priority of layers is considered during the routing determine optimal paths associated data rates each layer. Our scheme attempts find content distribution meshes with minimum path costs layer while satisfying inter-layer dependency coding....

Target grouping, which is essentially a data clustering problem, research hotspot in the field of battlefield situation assessment. To address an unknown environment, affinity propagation based on improved whale optimization algorithm (APBWOA) proposed from perspective clustering. First, we propose chaotic map and nonlinear inertia weight improvement called CPIW-WOA, uses circle to generate initial populations introduces weights improve its convergence efficiency. The test results nine...

Common CT Imaging Signs of Lung Diseases (CISL) are defined as the imaging signs that frequently appear in lung images from patients and play important roles diagnosis diseases. This paper proposes a new method multiple classifier fusion to recognize CISLs, which is based on confusion matrices classifiers classification confidence values outputted by classifiers. The matrix reflects historical reliability decision-making classifier, while difference between for competing classes current its...

Abstract Under the premise of many types and large numbers targets, how to solve problem rational allocation dynamic multi-objectives has always been a hot issue for experts scholars. In this paper, distributed target method based on auction algorithm is used kind problem. This paper establishes mathematical model air defense combat, uses Finally, effectiveness real-time performance are proved by simulation.

Recently, multilingual artificial intelligence assistants, exemplified by ChatGPT, have gained immense popularity. As a crucial gateway to human-computer interaction, automatic speech recognition (ASR) has also garnered significant attention, as evidenced systems like Whisper. However, the proprietary nature of training data impeded researchers' efforts study ASR. This paper introduces MSR-86K, an evolving, large-scale corpus for research. The is derived from publicly accessible videos on...

With the development of convolutional neural networks (CNNs), researchers have acquired satisfactory performances on computer vision tasks such as image classification and semantic segmentation. However, CNNs contain millons weight parameters demand huge storage computation resources, making it hard to deploy constrained hardwares. To compress model size accelerate deployment speed, many compression methods been proposed, including quantization that aim reduce network redundancy by...

Under the background of great changes in military science and technology theory, order to solve problem massive high-dimensional situation knowledge processing process battlefield assessment.The current dimensionality reduction methods often ignore influence algorithm complexity model representation ability on when solving knowledge. In balance this problem, paper proposes a dimension method based Sparse Autoencoder, which has good performance achieving information obtaining its abstract...

Aiming at the paradox of D-S evidence theory and computations exponential growth in dealing with large scale conflict combination, a new weighted combination method was proposed, which used coefficient distance order to measure conflict. Through analysis single representations weaknesses, compound has been put forward, meanwhile, center current were defined, weight determined coefficient. The experiment results show that algorithm settles effectively, same time, computing speed greatly enhanced.