Internet of Vehicles (IoVs) is highly characterized by collaborative environment data sensing, computing and processing. Emerging Big Data Artificial Intelligence (AI) technologies show significant advantages efficiency for knowledge sharing among intelligent vehicles. However, it challenging to guarantee the security privacy during process. Moreover, conventional AI-based algorithms cannot work properly in distributed vehicular networks. In this paper, a hierarchical blockchain framework...

In this paper, a kind of morphing skin embedded with pneumatic muscle fibers is proposed from the bionics perspective. The elastic modulus designed experimentally determined and their output force tested internal air pressure varying 0 to 0.4 MPa. experimental results show that contraction ratio using given material could reach up 26.8%. Isothermal tensile tests are conducted on fabricated skin, compared theoretical predictions based rule mixture. When strain lower than 3% in its...

As a typical smart material, shape memory polymers (SMPs) have the capability of variable stiffness in response to external stimuli, such as heat, electricity, magnetism and solvents. In this research, polymer composite (SMPC) tube composed multi-layered filament wound structures is investigated. The SMPC possesses considerable flexibility under high temperature rigidity low temperature. Significant changes effective engineering modulus can be achieved through regulating environment Based on...

Patients with relapsed or refractory multiple myeloma (R/R-MM) are more susceptible to develop severe coronavirus disease 2019 (COVID-19) for their immunocompromised states. Despite good responses B-cell maturation antigen (BCMA)-targeted chimeric receptor (CAR)-T cell therapy, deficiencies in humoral immunity following CAR-T infusions can still cause life-threatening complications these patients. We conducted a comparative study delineate the clinical characteristics and outcomes between...

With the rapid advancements in remote sensing technology, spectral information from hyperspectral images has become increasingly rich, facilitating detailed analysis of Earth's surface objects. However, this abundance data poses significant challenges processing, such as curse dimensionality leading to “Hughes” phenomenon, "strong correlation" due high resolution, and "non-linear characteristics" caused by varied reflectance rates. Therefore,...

With a large number of devices and wealth user data sets, the Internet Things (IoT) has become great host for federated learning (FL). At same time, massive amount in IoT results desperate demand privacy preserving. The onion router (Tor) is promising method to solve issue IoT-based FL by anonymity. However, devices' resource too limited execute cryptographic operations Tor. Moreover, network traffics Tor can be easily controlled malicious routers with fake high self-reported bandwidth. In...

Internet of Things (IoT) is an essential information infrastructure for smart cities in data sensing, collecting, merging, transmitting, or even reverse controlling. Whereas, the devices IoT are fundamentally different processing abilities. Combined with fog computing, some stronger nodes who have better computing capabilities can be taken as to help other weak task processing, such analysis, etc. Consequently, a new problem raised that how schedule tasks among nodes, so achieve minimum...

In this study, a kind of morphing skin composed silicon rubber and shape memory polymer composite tube is designed, manufactured, investigated. Significant stiffness variation for the can be obtained through changing its environment temperature. First, in order to investigate basic elastic properties skin, Rule Mixture used predict effective engineering modulus ratio with different matrix volume fractions. The tensile test conducted on two states (cold state heated state) validate accuracy...

Purpose This study considers online searching by health information consumers as a learning process. We focus on search sequences, query reformulation, and conceptual changes. Design/methodology/approach A qualitative user (30 participants; three seeking tasks) investigated mobile behavior. Recorded screen activity, questionnaires, in-depth personal interview data were collected analyzed. Findings (1) Search platform sequences of in process exacted their features highlighted. (2) Query...

Distributed machine learning (DML) can realize massive dataset training when no single node work out the accurate results within an acceptable time. However, this will inevitably expose more potential targets to attackers compared with non-distributed environment. In paper, we classify DML into basic-DML and semi-DML. basic-DML, center server dispatches tasks distributed machines aggregates their results. While in semi-DML, further devotes resources addition its duty basic-DML. We firstly...

In this research, a fluidic flexible matrix composites (F 2 MC) tube composed of composite (FMC) and inner liner is investigated. Significant changes in effective axial elastic modulus could be achieved through controlling the interior fluid. Based on classical laminated-plate theory anisotropic elasticity, three-dimensional analytical method proposed to characterize mechanical behavior F MC tube. comparison with experiment result, analysis deemed possess satisfying accuracy prediction...

Software Defined Wireless Rechargeable Sensor Networks (SDWRSNs) are an inexorable trend for (WSNs), including Network (WRSNs). However, the traditional network devices cannot be completely substituted in short term. Hybrid SDWRSNs, where software defined and coexist, will last a long time. SDWRSNs bring new challenges as well opportunities energy saving issues, which is still key problem considering that wireless chargers also exhaustible, especially some rigid environment out of main...

A new structure damage identification method is proposed based on wavelet packet analysis of the time domain response. It known that distribution node energy response varies if damaged. The presented i

The Service Function Chain (SFC) is a mechanism which transmit packets through different services in order. With Software Defined Networking (SDN) and Network Virtualization (NFV), network can be flexible deployed on the middleboxes. However, end-to-end transmission delay of SFC will influence time network, key indicator to performance network. One way decrease optimize sequence service In this paper, we put forward Middlebox Placement Optimization (MPO) algorithm. Different from previous...

Domain adaptation helps generalizing object detection models to target domain data with distribution shift. It is often achieved by adapting access the whole data. In a more realistic scenario, unpredictable until inference stage. This motivates us explore an model at test-time, a.k.a. test-time (TTA). this work, we approach adaptive (TTAOD) from two perspective. First, adopt self-training paradigm generate pseudo labeled objects exponential moving average model. The labels are further used...

The powerful information diffusion ability of online social networks (OSNs) attracts many attackers to spread malware, because malware can fiercely among massive active users with close relationship in OSNs. A message-recallable function user profile modern OSNs delete messages containing malicious link, and stop spreading. However, existing works on propagation models have not concerned the new mechanism (MRM). This paper proposes a novel model analyze performance MRM. For analyzing model,...

Malware, which spreads among hosts, brings potential damage to computer systems. Its propagation will even be facilitated when search engine exists. In this paper, we try dig deeper into the infected time of communities and malware behaviour in a context. order address these problems, two layer model is proposed for via engine. Derived from model, have found that exists, reciprocal different follows power law distribution number increases exponentially at early stage. Both hypothesis are...

In this presented paper, spandex fibers with high elasticity and recovery ratio were added into shape memory epoxy resin, the mechanical properties improved obviously compared pure resin. Compared material, elastic modulus of sample 20vol% was increased by 28.2%, tensile stress 49.7%, fracture strain 16.4%. Then graphitized multi-walled carbon nanotubes (GMWNTs) mixed reinforced SMPCs to make it conductive. It found that surface-modified (by acid treatment) GMWNTs incorporated very well...

Recent work has showcased the powerful capability of large language models (LLMs) in recalling knowledge and reasoning. However, reliability LLMs combining these two capabilities into reasoning through multi-hop facts not been widely explored. This paper systematically investigates possibilities for to utilize shortcuts based on direct connections between initial terminal entities knowledge. We first explore existence factual Knowledge Neurons, revealing that: (i) strength is highly...

Abstract Background Cancer‐targeted T‐cell receptor T (TCR‐T) cells hold promise in treating cancers such as hematological malignancies and breast cancers. However, approaches to obtain cancer‐reactive TCR‐T have been unsuccessful. Methods Here, we developed a novel strategy screen for cancer‐targeted using special humanized mouse model with person‐specific immune fingerprints. Rare steady‐state circulating hematopoietic stem progenitor were expanded via three‐dimensional culture of...