Blockchain and AI are promising techniques for next-generation wireless networks. can establish a secure decentralized resource sharing environment. be explored to solve problems with uncertain, time-variant, complex features. Both of these have recently seen surge in interest. The integration two further enhance the performance In this article, we first propose intelligent architecture networks by integrating blockchain into enable flexible sharing. Then empowered content caching problem...

Douyin, internationally known as TikTok, has become one of the most successful short-video platforms. To maintain its popularity, Douyin to provide better Quality Experience (QoE) growing user base. Understanding characteristics videos is thus critical service improvement and system design. In this paper, we present an initial study on fundamental based a dataset over 260 thousand short collected across three months. The are found be significantly different from traditional online videos,...

Generating a new font library is very labor-intensive and time-consuming job for glyph-rich scripts. Few-shot generation thus required, as it requires only few glyph references without fine-tuning during test. Existing methods follow the style-content disentanglement paradigm expect novel fonts to be produced by combining style codes of reference glyphs content representations source. However, these few-shot either fail capture content-independent representations, or employ localized...

Aiming at the problem of privacy leakage during data sharing in Internet Things, a blockchain based secure platform with fine-grained access control(BSDS-FA) is proposed. First, this paper proposes new hierarchical attribute-based encryption algorithm, which uses attribute structure and multi-level authorization center. The algorithm implements flexible control by distributing different user attributes to centers. Then, it combined Fabric technology solve huge decryption cost for users...

This paper presents a novel Region-Aware Face Swapping (RAFSwap) network to achieve identity-consistent harmonious high-resolution face generation in local-global manner: 1) Local Facial (FRA) branch augments local identity-relevant features by introducing the Transformer effectively model misaligned crossscale semantic interaction. 2) Global Source Feature-Adaptive (SFA) further complements global cues for generating swapped faces. Besides, we propose Mask Predictor (FMP) module...

Training a text-to-image generator in the general domain (e.g., Dall.e, CogView) requires huge amounts of paired text-image data, which is too expensive to collect. In this paper, we propose self-supervised scheme named as CLIP-GEN for generation with language-image priors extracted pre-trained CLIP model. our approach, only require set unlabeled images train generator. Specifically, given an image without text labels, first extract embedding united language-vision space encoder CLIP. Next,...

Conventional yaw stability strategy of distributed drive electric vehicle (DDEV) is usually realised by torque distribution strategy. However, the instantaneous variations four independent tyres slip ratio and effect disturbance have not been considered sufficiently. Therefore, it difficult to realise robustness for DDEV under various operating conditions. To solve this problem, a novel model predictive controller-based compensation control system (MPC-CCS) proposed in paper. The MPC-CCS...

The interest of artificial intelligence (AI) and mobile edge computing (MEC) has increased in recent years. As a fundamental building block for the development computing, new techniques are needed network. Driven by Internet Things (IoT) AI, MEC, as an emerging paradigm, enables cloud services AI frontier close proximity to users, pushing cache resources from network edge. With continuous in-depth research on intelligent networks, green would be paradigm MEC. is playing more important role...

Set synchronization is an essential job for distributed applications. In many cases, given two sets A and B, applications need to identify those elements that appear in set but not vice versa. Bloom filter, a spaceefficient data structure representing supporting membership queries, has been employed as lightweight method realize with low false positive probability. Unfortunately, bloom filters their variants can only be applied simple rather than more general multisets, which allow multiple...

Due to the increasing growth of technologies and diversity user needs in field information technology, position cloud computing is becoming more apparent. The development infrastructure any organization requires spending a lot money, time, manpower, which sometimes does not fit into operational capacity an organization. Therefore, organizations tend use such advance their goals. A fully open distributed structure its services makes it attractive target for attackers. This includes multiple...

Orbital angular momentum (OAM) beams have demonstrated significant potential in the fields of communication, imaging and quantum information science. In particular, active OAM metasurfaces enable flexible on-demand control without altering their structures, significantly enriching applications beams. However, due to limitations electrode wiring unit structure efficiency, most existing metasurface designs are reflective, hindering further development this field. Here, we propose an induced...

The dramatically growing trend of vehicles equipped with driving camera recorders has allowed realizing real-time crowdsourced video sharing in vehicular edge computing (VEC). Such cameras can assist monitoring objects directly front and behind the vehicles, enabling them to provide important visual information through streaming case possible accidents. Exploiting on-board units (OBUs) for VEC allow drivers passengers share access on-road surveillance services. However, data security privacy...

Considering the inevitable trends for heterogeneous network convergence, Cognitive Radio Network (CRN) concept has been proposed with some essential characteristics to achieve adaptation and global end-to-end goals. This motivates a more flexible effective dynamic channel assignment scheme which can utilize licensed spectrum effectively through reusing idle opportunistically. paper focuses on offers optimal resource allocation mechanism satisfy requirement of users networks in transmission....

Software-defined networking (SDN) enables the separation of control and data planes, allowing for centralized management network. Without adequate access methods, risk unauthorized to network its resources increases significantly. This can result in various security breaches. In addition, if authorized devices are attacked or controlled by hackers, they may turn into malicious devices, which cause severe damage their abnormal behaviour goes undetected privileges not promptly restricted. To...

Recently 3D-aware GAN methods with neural radiance field have developed rapidly. However, current model the whole image as an overall field, which limits partial semantic editability of synthetic results. Since NeRF renders pixel by pixel, it is possible to split in spatial dimension. We propose a Compositional Neural Radiance Field (CNeRF) for portrait synthesis and manipulation. CNeRF divides regions learns independent each region, finally fuses them complete image. Thus we can manipulate...