418 research outputs found
Offloading Decision Making Strategies and Applications in Mobile Edge Computing
No full textMobile edge computing (MEC), as a primary characteristic of fifth-generation (5G) networks, is a critical computing measure for providing a highly distributed computing environment in applications such as virtual reality enhancement, smart cities, connected vehicles and healthcare, etc. Intelligent computational offloading decision-making strategies in MEC can effectively schedule computational tasks for remote applications and decrease latency and energy consumption in dealing with tasks. As a notable application in MEC, this study concentrates on intelligent offloading decision-making strategies and their applications in MEC, where theoretical methods in game-theoretical modelling are investigated and formulated by considering various metrics such as latency constraints, energy consumption, the revenue of network operators, etc. Based on the foundation in modelling, efficient and effective solutions are explored to solve game-theoretical models in different application scenarios. More specifically, the main work and contributions in this thesis are listed as follows:
• We investigate game-theoretical modelling and consider it as the primary method to formulate offloading decision-making strategies. The related metrics such as latency and energy consumption are integrated into a two-stage game theory framework for UAV scheduling. The optimal decision searching is developed with the objectives of reputation increase and energy conservation of UAVs. The profit of a network operator is maximized in the game-playing process and the network economy research is further explored.
• To extend the exploration of offloading decision-making strategies in the network economy of MEC, we bridge the gap between revenues of network operators with offloading decision-making strategies. Revenues are realistic and ultimate goals of network operators in competitive markets. Therefore, a pricing scheme combined with the aims of reduction and restriction of energy consumption and latency is studied for revenue maximization of service operators.
• Besides, we develop the offloading decision-making strategy for a complicated application scenario, where revenue maximization of network operators in maritime communications assisted by hybrid satellite-UAV-terrestrial networks is elucidated. In this work, we first formulate a two-stage game model considering the cost of offloading and the revenue of network operators. Then we conduct an equilibrium analysis to verify the existence of the game model and clarify the process of bridging machine learning to solve the game theoretical model.
• In addition, we further explore a more effective solution to solve the formulated game-theoretical model by considering scalable offloading decision-making strategies in vehicular edge computing (VEC) along with the number increase in edge vehicles. A neural network is developed for solving the game-theoretical model which is built for the scenario with a single vehicle, and then an approach integrated with transfer learning is formulated to extend the offloading decision strategy to a large-scale vehicle scenario, by which the scalable optimization of game-theoretical offloading for VEC can be reached
Performance Modelling and Resource Allocation of the Emerging Network Architectures for Future Internet
No full textWith the rapid development of information and communications technologies, the traditional network architecture has approached to its performance limit, and thus is unable to meet the requirements of various resource-hungry applications. Significant infrastructure improvements to the network domain are urgently needed to guarantee the continuous network evolution and innovation. To address this important challenge, tremendous research efforts have been made to foster the evolution to Future Internet. Long-term Evolution Advanced (LTE-A), Software Defined Networking (SDN) and Network Function Virtualisation (NFV) have been proposed as the key promising network architectures for Future Internet and attract significant attentions in the network and telecom community. This research mainly focuses on the performance modelling and resource allocations of these three architectures. The major contributions are three-fold:
1) LTE-A has been proposed by the 3rd Generation Partnership Project (3GPP) as a promising candidate for the evolution of LTE wireless communication. One of the major features of LTE-A is the concept of Carrier Aggregation (CA). CA enables the network operators to exploit the fragmented spectrum and increase the peak transmission data rate, however, this technical innovation introduces serious unbalanced loads among in the radio resource allocation of LTE-A. To alleviate this problem, a novel QoS-aware resource allocation scheme, termed as Cross-CC User Migration (CUM) scheme, is proposed in this research to support real-time services, taking into consideration the system throughput, user fairness and QoS constraints.
2) SDN is an emerging technology towards next-generation Internet. In order to improve the performance of the SDN network, a preemption-based packet-scheduling scheme is firstly proposed in this research to improve the global fairness and reduce the packet loss rate in SDN data plane. Furthermore, in order to achieve a comprehensive and deeper understanding of the performance behaviour of SDN network, this work develops two analytical models to investigate the performance of SDN in the presence of Poisson Process and Markov Modulated Poisson Process (MMPP) respectively.
3) NFV is regarded as a disruptive technology for telecommunication service providers to reduce the Capital Expenditure (CAPEX) and Operational Expenditure (OPEX) through decoupling individual network functions from the underlying hardware devices. While NFV faces a significant challenging problem of Service-Level-Agreement (SLA) guarantee during service provisioning. In order to bridge this gap, a novel comprehensive analytical model based on stochastic network calculus is proposed in this research to investigate end-to-end performance of NFV network.
The resource allocation strategies proposed in this study significantly improve the network performance in terms of packet loss probability, global allocation fairness and throughput per user in LTE-A and SDN networks; the analytical models designed in this study can accurately predict the network performances of SDN and NFV networks. Both theoretical analysis and simulation experiments are conducted to demonstrate the effectiveness of the proposed algorithms and the accuracy of the designed models. In addition, the models are used as practical and cost-effective tools to pinpoint the performance bottlenecks of SDN and NFV networks under various network conditions
Solar photovoltaics can help China fulfill a net-zero electricity system by 2050 even facing climate change risks
No full textAs China has pledged to become carbon neutral by 2060, electrifying its energy sector is no doubt one of the priority measures to support the transition towards a more sustainable and decarbonized energy system. Solar photovoltaics (PV) has been known as one of the most promising renewable technologies to facilitate the electrification of energy systems. The feasibility of utilizing PV to implement a nationwide decarbonized electricity system now becomes an urgent unanswered question, especially in the context of global climate change and rapid economic growth in China. Here, by using a GIS-based multiple-criteria decision-making approach we address this question by conducting a comprehensive feasibility analysis with consideration of various economic, technological, logistical, and climate change factors. We show that it is feasible for China to fulfill a net-zero electricity system by 2050, through the installation of 7.46 TW solar PV panels on about 1.8% of the national land area (mostly in western China) with a total capital investment of 4.55 trillion USD in the next 30 years. Besides, we show that future climate change may lead to a slight decrease (less than 5%) in solar energy potential, but this would not affect the capability of the nationwide PV system to meet the need for a fully-electrified energy system.National Natural Science Foundation of Chin
Algorithms and performance analysis for narrowband internet of things (NB-IoT) and broadband LTE coexisting system
No full textThis chapter describes a comprehensive uplink coexistence system model for narrowband Internet of Things IoT (NB-IoT) and long-term evolution for the purpose of investigating the potential influence caused by the mismatched sampling rate. Most of the NB-IoT–related research reported in the literature has focused on frame structure design, physical layer analysis, random access network and scheduling, for example. In order to achieve the possibility of quick deployment, the NB-IoT is designed to operate on existing cellular networks, for example, the Evolved Universal Mobile Telecommunications System Terrestrial Radio Access and Global System for Mobile Communication. Three different modes could be assigned for NB-IoT operation—stand-alone mode, in-band mode, and guard-band mode—so that the spectrum resource could be utilized efficiently and flexibly. In addition, an arbitrary sample duration for the NB-IoT device is considered to construct the foundation of sample duration optimization
On the Effectiveness of Invisible Backdoor Attacks in Federated Learning
No full textFederated Learning (FL) enables collaborative training of machine learning models across multiple devices, while preserving data privacy. However, it also introduces vulnerabilities to backdoor attacks, where malicious updates can corrupt the global model. This work focuses on the largely unexplored domain of Federated invisible Backdoor Attacks (FiBA s), which use visually indistinguishable triggers to manipulate model behavior without being detected. We investigate the feasibility and effectiveness of these attacks, considering the unique challenges posed by FL, such as limited local training time and the need for model update aggregation. Our study presents a comprehensive evaluation of the attack success rate (ASR) of invisible BAs under various settings and defenses. Based on our observations, we propose a backdoor trigger hiding technique based on low model attention regions to improve attack resilience in federated settings. Our findings provide critical insights into the optimization of invisible BAs in FL and highlights the need for robust defense mechanisms to safeguard FL systems
Wrinkle direction detection and its application on robotic cloth wrinkle removal
No full textDeformable Object Manipulation (DOM) is an important field of research as it contributes to practical tasks such as cloth handling, cable routing, surgical operation etc. The sensing in DOM is now considered as one of the major challenges in robotics due to the complex dynamics and high degree of freedom of deformable objects. One challenge is to find a suitable representation with low dimensionality and reliable accuracy. The aim of this thesis to develop an algorithm to represent the state of the deformable objects like cloth in low-dimensional vectors, together with a framework based on visual servoing to flatten cloth-like objects. We present a novel pipeline for cloth flattening, which determines a stretching direction (in 2D vector) and an operation point for the robot to removes the wrinkles. The performance of the perception algorithm is validated in simulation and real-world experiment. The whole framework is evaluated in the real-world experiment, which is compared with a human operator. The results show that our framework efficiently determines the direction of wrinkles on the cloth in the simulation as well as the real robot experiment. Besides, the proposed framework has a good performance close to that of a human operator in terms of cloth flattening tasks.Mechanical Engineering | Vehicle Engineering | Cognitive Robotic
CdS-phenanthroline derivative hybrid cathode interlayers for high performance inverted organic solar cells
No full textPhenanthroline based organic semiconductors (BCP, Bphen, Mphen, and Phen) are used to hybrid with CdS as cathode interlayers in inverted organic solar cells (OSCs). We observed that selecting the polar solvent and hydrophobic interlayers with a diphenyl group could improve the performance of the organic photovoltaic devices. The modification to CdS can effectively improve its electron mobility, film morphology, interfacial contact, and energy level alignment, which finally leads to a significant enhancement of device performance. Through incorporating the CdS-P hybrids (CdS-BCP, CdS-Bphen, CdS-Mphen, and CdS-Phen) as cathode buffer layers, the device PCE (PTB7 : PC71BM as the active layer) is greatly improved from 3.09% to 8.36, 7.84, 6.69, and 6.57%, respectively, compared with devices fabricated on the pristine CdS interlayer. These results indicate that the common inorganic semiconductor like CdS can be modified using some organic semiconductors to produce general applicable electron transport layers applied in OSCs. Our work puts forward new insights for the development of new interface modification materials and fabrication of high efficiency devices
- …
