Crowd counting is generally a pixel-wise regression task and owns the broad application prospects. In order to generate high-quality density maps, recent remarkable works mostly adopt the long skip connection or encod...
详细信息
Autonomous driving applications rely on processing large amounts of data in order to ensure sufficient perception performance. In this context, Convolutional Neuronal networks (CNNs), which are used for object detecti...
Autonomous driving applications rely on processing large amounts of data in order to ensure sufficient perception performance. In this context, Convolutional Neuronal networks (CNNs), which are used for object detection in camera images, are an integral part of sensor data processing. However, safety-related hard deadlines are counteracted by the required high data rates that stress the platform w.r.t. memory interference. Providing high camera frame rates under hard latency requirements is still an open issue. In a case study focusing on high-performance multi-core architectures, we evaluate in detail the advantages of a private L2 and shared L3 cache architecture for CNN processing and show how to remove malicious data synchronization effects. In this context we deploy MobileNet as well as YOLO on an Intel i5 processor and demonstrate their applicability to a worst-case design under high CNN frame rates. Last, we provide a heuristic optimization scheme that is able to efficiently find feasible high frame rate configurations. Contrary to common assumptions, our results show that high-performance multi-core COTS platforms are suitable for the application of CNNs even under hard deadline constraints and, hence, offer substantial gains in cost and productivity due to their greater ease of programming.
The shift towards renewable energies is increasing communication demands, particularly in novel energy grid architectures. One such approach is the concept of cellular energy systems, which divide the grid into region...
详细信息
In cities, mobile communication has the ground for smart collaborative traffic applications. 5G architecturesopen, as known, the digital engagement among devices. Thus, this research shows the current state of the ce...
详细信息
ISBN:
(数字)9781665467469
ISBN:
(纸本)9781665467469
In cities, mobile communication has the ground for smart collaborative traffic applications. 5G architecturesopen, as known, the digital engagement among devices. Thus, this research shows the current state of the cellular network implemented in Iasi city considering a 5G module into a public transport vehicle. The proposed methods for data acquisition, processing, and visualization, illustrate results that highlight zonal/road characteristics favorable or not for V2X communications. Essentially, a GPS data correction solution stands and allows behavior isolation of the vehicle in the traffic. This approach can be used to measure over time the LTE and 5G systems.
This paper tackles the optimization of placement and quantity of open RAN Data Centers (ODCs) for hosting virtualized open Distributed Units (O-DUs) in 5G NR networks, promoting flexible and scalable SDN architectures...
详细信息
ISBN:
(数字)9798350380538
ISBN:
(纸本)9798350380545
This paper tackles the optimization of placement and quantity of open RAN Data Centers (ODCs) for hosting virtualized open Distributed Units (O-DUs) in 5G NR networks, promoting flexible and scalable SDN architectures. Utilizing a Multi-Objective Facility Location Problem (MO-FLP) framework, we aim to minimize infrastructure costs related to network function virtualization (NFV) while meeting strict fronthaul latency and processing capacity requirements. Our method employs Non-dominated Sorting Genetic Algorithms (NSGA-II) to manage these trade-offs, achieving significant cost-efficiency and network performance gains through extensive city-wide 5G NR deployment simulations. We validated our approach in Natal and Manaus, evaluating ODC locations, capacities, and fiber usage. This study offers new insights and practical strategies for efficient 5G NR deployments, addressing key challenges in modern telecommunications infrastructure.
The paper presents an original research of fake accounts on social media using an artificial neural network for their identification. Specifically designed and implemented application was used to identify specific fea...
详细信息
ISBN:
(纸本)9789526924465
The paper presents an original research of fake accounts on social media using an artificial neural network for their identification. Specifically designed and implemented application was used to identify specific features of fake accounts and study the principles and reasons of their generation. Considering the possible use cases and scenarios of its practical use including everyday social media surfing there was examined a possibility to implement a mobile application based on Java in the Android Studio programming environment. Based on the study of 500 real and 500 fake VKontakte accounts there was determined a number of conclusions on the original features of fake accounts. The provided research allowed extending the list of criteria for identifying fake accounts by an original set of patterns. The developed conclusions allow formulating the statements on what criteria can be used for further identification of fake accounts in practical applications.
The described analytical tools were already fully functioning in 2018, and they are continually being enhanced. The article discusses one tool's potential for use in public support institutions. The case study of ...
详细信息
Social networks crawling is in the focus of active research the last years. One of the challenging task is to collect target nodes in an initially unknown graph given a budget of crawling steps. Predicting a node prop...
详细信息
Quantum technologies are emerging. Dedicated languages for programming Quantum machines are emerging as well and already used in different settings. Orthogonal to this development, Model-Driven Engineering (MDE) is ex...
详细信息
ISBN:
(纸本)9781665444620
Quantum technologies are emerging. Dedicated languages for programming Quantum machines are emerging as well and already used in different settings. Orthogonal to this development, Model-Driven Engineering (MDE) is explored to ease the development of software systems by applying modeling techniques such as Domain-Specific Modeling Languages and generative techniques such as code generation. In this position paper, we argue for a dedicated research line which deals with the exploration of how MDE may be applied for Quantum technologies. This combination would allow to speed-up the development of Quantum software, open the door for domain experts to utilize Quantum technologies, and may provide an additional abstraction layer over existing Quantum hardware architectures and programming languages. We outline several research challenges which we derived from a demonstration case of how to exploit domain-specific modeling for social network analysis on Quantum technologies.
This paper develops an adaptive dynamic programming-based event-triggered robust control method for fully cooperative games of nonlinear multi-player systems with uncertainties. By designing a modified value function,...
详细信息
ISBN:
(数字)9781665408530
ISBN:
(纸本)9781665408530;9781665408523
This paper develops an adaptive dynamic programming-based event-triggered robust control method for fully cooperative games of nonlinear multi-player systems with uncertainties. By designing a modified value function, the robust control problem is transformed into a fully cooperative (FC) game of the nominal system. Then, a critic neural network is constructed to solve the Hamilton-Jacobi-Bellman equation of the FC game, and the optimal control laws of all players are obtained. Under the event-triggered mechanism, the designed event-triggering condition is employed to determine whether the control laws should be updated or not. The developed method saves computation and communication resources, since both the updating frequency of control laws and the communication between controllers and actuators are reduced. Furthermore, the closed-loop system is proven to be uniformly ultimately bounded thanks to the Lyapunov's direct method. A numerical example is employed to verify the effectiveness of the developed method.
暂无评论