Driven by numerous emerging services and applications of mobile devices, multi-access edge computing (MEC) is regarded as a promising technique for massive Internet Things (IoT) with 6G networks to alleviate core network congestion reduce service latency. However, the conventional MEC suffers from infrastructure without cloud server (CS) cooperation multiple servers (ESs), which cannot deal large-scale computation tasks in ultra-dense smart environments. This paper investigates issue...

Model partitioning is a promising technique for improving the efficiency of distributed inference by executing partial deep neural network (DNN) models on edge servers (ESs) or Internet-of-Things (IoT) devices. However, due to heterogeneous resources ESs and IoT devices in mobile computing (MEC) networks, it non-trivial guarantee DNN speed satisfy specific delay constraints. Meanwhile, many existing have complex architecture with numerous blocks, which leads huge search space fine-grained...

Geographic routing has been considered as an attractive approach for wireless mobile ad hoc and sensor networks due to its effectiveness scalability. Over the past few decades, numerous geographic protocols have proposed extensively in 2D space. However, these are no longer valid if or distributed 3D environments, such space, atmosphere, ocean. Because of surprisingly difficulty designing compared networks, only a prior studies focused on specifically designed networks. In this article, we...

As ubiquitous and personalized services are growing boomingly, an increasingly large amount of traffic is generated over the network by massive mobile devices. a result, content caching gradually extending to edges provide low-latency services, improve quality service, reduce redundant data traffic. Compared conventional delivery networks, caches in edge networks with smaller sizes usually have accommodate more bursty requests. In this article, we propose evolving learning-based policy,...

Network function virtualization (NFV) is a promising paradigm that network functions can be deployed on commodity servers instead of dedicated to enhance the resource utilization and reduce management difficulty. Based NFV technology, complex service composed series ordered virtual functions, known as chain (SFC). In this context, how efficiently place SFCs in acceptable running time improve quality while meeting constraints physical critical issue for infrastructure providers. paper, we...

With the rapid growth of networked multimedia services in Internet, wireless network traffic has increased dramatically. However, current mainstream content caching schemes do not take into account cooperation different edge servers, resulting deteriorated system performance. In this paper, we propose a learning-based scheme to enable mutual among servers with limited resources, thus effectively reducing delivery latency. Specifically, formulate cooperative problem as an optimization...

Due to its high mobility and low cost, unmanned aerial vehicle mounted base station (UAV-BS) can be deployed in a fast cost-efficient manner for providing wireless services areas where traditional terrestrial infrastructures cannot laid technical economic reasons. In this letter, we investigate the problem of joint three-dimensional (3D) deployment power allocation maximizing system throughput UAV-BS system. To solve non-convex problem, propose deep deterministic policy gradient (DDPG) based...

To address the challenge of efficiently integrating multi-source heterogeneous data to improve accuracy public safety event prediction, this study proposes and validates a novel prediction model, GATPNet, based on data. The model integrates Graph Attention Networks (GAT), Spatiotemporal Transformers, Proximal Policy Optimization (PPO) achieve effective fusion, spatiotemporal feature extraction, real-time decision support. Through experiments conducted Los Angeles Crime Data CrisisLexT26...

To support rapidly increasing multimedia services of smart cities, mobile edge computing (MEC) networks can significantly reduce content acquisition latency. However, due to user mobility and the possibility re-association, it is challenging obtain a proper caching association policy. In this article, we investigate issue joint with high in MEC by minimizing latency handover address problem, optimize original mixed time-scale problem two stages: long-time scale short-time association....

In this article, we propose an energy-efficiency-aware integrated architecture of cooperative computing (CC) to support the demands amount in blockchain-based Internet Things (IoT). Specifically, assume that multiple servers are placed at each data access point (DAP). The across DAPs can be virtualized constitute a CC pool flexibly allocate resource. When received from DAP is accumulated certain length one block, blockchain will implemented generate correct Nonce value meeting threshold hash...

Network function virtualization (NFV) is a promising paradigm where network functions are migrated from dedicated hardware appliances onto software middleboxes to promote service agility and reduce management costs. Benefiting the NFV, chain (SFC) has emerged as popular form. It allows traffic pass through series of virtual in specific order required by business logic arrange complex service. However, SFC deployment facing new challenges seeking trade-off between pursuing objective high...

With the explosive growth of mobile multimedia services and artificial intelligence applications involving automatic speech analysis (ASA), devices are increasingly unable to handle these computation-intensive tasks generated by users due limited computing resource. Besides, existing cloud paradigm is not capable processing such real-time delay-sensitive ASA tasks. In this paper, leveraging edge deep learning (DL), we investigate task offloading for DL-empowered in edge-cloud networks...

The decision of request mapping-which server to handle user and response routing-which transit route carry back has great impact on the performance cost Content Delivery Networks (CDNs). Request mapping routing are traditionally treated independently. information invisibility inconsistent objectives may lead worse high cost. However, rapid globalization Internet eXchange Points (IXPs) facilitated cooperation between CDN ISP. In this paper, we consider jointly. We formulate joint problem...

In this letter, we propose a novel cooperative caching scheme for the next-generation Internet - content oriented networking (CON), trying to minimize access delay mobile users. We formulate problem as mixed integer programming model and heuristic solution based on Lagrangian relaxation. Simulation results show that can greatly reduce delay.

With the rapid increase of data in mobile edge computing (MEC) networks, devices (MDs) have been generating many computation-latency-sensitive tasks. As MDs are limited by resources terms storage, computation, and bandwidth, part tasks to be offloaded networks or remote cloud for more efficient processing. Hence, task offloading plays a vital role this scene. Existing works about mainly aim at one-shot rarely consider dependencies In paper, we focus on minimizing maximum delay processing...