arXiv subjects

Vasilis Friderikos

Publications and source records attributed to Vasilis Friderikos.

30 records · Page 2Linked to original sources

arXiv2021

Trajectory Design in Rechargeable UAV Aided Wireless Networks: A Unified Framework for Energy Efficiency and Flight Time Minimization

This paper studies a rechargeable unmanned aerial vehicle (UAV) assisted wireless network, where a UAV is dispatched to disseminate information to a group of ground terminals (GTs) and returns to a recharging station (RS) before the on-board battery is depleted. The central aim is to design a UAV trajectory with the minimum total time duration, including the flight and recharging time, by optimizing the flying velocity, transmit power and hovering positions jointly. A flow-based mathematical programming formulation is proposed to provide optimal for joint optimization of flying and recharging time. Furthermore, to attack the curse of dimensionality for optimal decision making, a two-step method is proposed. In the first step, the UAV hovering positions is fixed and initialize a feasible trajectory design by solving a travelling salesman problem with energy constraints (TSPE) problem. In the second step, for the given initial trajectory, the time consumption for each sub-tour is minimized by optimizing the flying velocity, transmit power and hovering positions jointly. Numerical results show that the proposed method outperforms state of art techniques and reduces the aggregate time duration in an efficient way.

cs.NI↗

arXiv2021

Enhanced LSTM-based Service Decomposition for Mobile Augmented Reality

Undoubtedly, Mobile Augmented Reality (MAR) applications for 5G and Beyond wireless networks are witnessing a notable attention recently. However, they require significant computational and storage resources at the end device and/or the network via Edge Cloud (EC) support. In this work, a MAR service is considered under the lenses of microservices where MAR service components can be decomposed and anchored at different locations ranging from the end device to different ECs in order to optimize the overall service and network efficiency. To this end, we propose a mobility aware MAR service decomposition using a Long Short Term Memory (LSTM) deep neural network to provide efficient pro-active decision making in real-time. More specifically, the LSTM deep neural network is trained with optimal solutions derived from a mathematical programming formulation in an offline manner. Then, decision making at the inference stage is used to optimize service decomposition of MAR services. A wide set of numerical investigations reveal that the mobility aware LSTM deep neural network manage to outperform recently proposed schemes in terms of both decision making quality as well as computational time.

cs.NI↗

arXiv2020

A Survey of Deep Learning for Data Caching in Edge Network

The concept of edge caching provision in emerging 5G and beyond mobile networks is a promising method to deal both with the traffic congestion problem in the core network as well as reducing latency to access popular content. In that respect end user demand for popular content can be satisfied by proactively caching it at the network edge, i.e, at close proximity to the users. In addition to model based caching schemes learning-based edge caching optimizations has recently attracted significant attention and the aim hereafter is to capture these recent advances for both model based and data driven techniques in the area of proactive caching. This paper summarizes the utilization of deep learning for data caching in edge network. We first outline the typical research topics in content caching and formulate a taxonomy based on network hierarchical structure. Then, a number of key types of deep learning algorithms are presented, ranging from supervised learning to unsupervised learning as well as reinforcement learning. Furthermore, a comparison of state-of-the-art literature is provided from the aspects of caching topics and deep learning methods. Finally, we discuss research challenges and future directions of applying deep learning for caching

cs.NI↗

arXiv2020

Network Orchestration in Mobile Networks via a Synergy of Model-driven and AI-based Techniques

As data traffic volume continues to increase, caching of popular content at strategic network locations closer to the end user can enhance not only user experience but ease the utilization of highly congested links in the network. A key challenge in the area of proactive caching is finding the optimal locations to host the popular content items under various optimization criteria. These problems are combinatorial in nature and therefore finding optimal and/or near optimal decisions is computationally expensive. In this paper a framework is proposed to reduce the computational complexity of the underlying integer mathematical program by first predicting decision variables related to optimal locations using a deep convolutional neural network (CNN). The CNN is trained in an offline manner with optimal solutions and is then used to feed a much smaller optimization problems which is amenable for real-time decision making. Numerical investigations reveal that the proposed approach can provide in an online manner high quality decision making; a feature which is crucially important for real-world implementations.

cs.NI↗

arXiv2020

Caching as an Image Characterization Problem using Deep Convolutional Neural Networks

Caching of popular content closer to the mobile user can significantly increase overall user experience as well as network efficiency by decongesting backbone network segments in the case of congestion episodes. In order to find the optimal caching locations, many conventional approaches rely on solving a complex optimization problem that suffers from the curse of dimensionality, which may fail to support online decision making. In this paper we propose a framework to amalgamate model based optimization with data driven techniques by transforming an optimization problem to a grayscale image and train a convolutional neural network (CNN) to predict optimal caching location policies. The rationale for the proposed modelling comes from CNN's superiority to capture features in grayscale images reaching human level performance in image recognition problems. The CNN is trained with optimal solutions and numerical investigations reveal that the performance can increase by more than 400% compared to powerful randomized greedy algorithms. To this end, the proposed technique seems as a promising way forward to the holy grail aspect in resource orchestration which is providing high quality decision making in real time.

cs.NI↗

arXiv2018

Dynamic Placement of VNF Chains for Proactive Caching in Mobile Edge Networks

Notwithstanding the significant research effort Network Function Virtualization (NFV) architectures received over the last few years little attention has been placed on optimizing proactive caching when considering it as a service chain. Since caching of popular content is envisioned to be one of the key technologies in emerging 5G networks to increase network efficiency and overall end user perceived quality of service we explicitly consider in this paper the interplay and subsequent optimization of caching based VNF service chains. To this end, we detail a novel mathematical programming framework tailored to VNF caching chains and detail also a scale-free heuristic to provide competitive solutions for large network instances since the problem itself can be seen as a variant of the classical NP-hard Uncapacitated Facility Location (UFL) problem. A wide set of numerical investigations are presented for characterizing the attainable system performance of the proposed schemes.

cs.NI↗

arXiv2018

Will SDN be part of 5G?

For many, this is no longer a valid question and the case is considered settled with SDN/NFV (Software Defined Networking/Network Function Virtualization) providing the inevitable innovation enablers solving many outstanding management issues regarding 5G. However, given the monumental task of softwarization of radio access network (RAN) while 5G is just around the corner and some companies have started unveiling their 5G equipment already, the concern is very realistic that we may only see some point solutions involving SDN technology instead of a fully SDN-enabled RAN. This survey paper identifies all important obstacles in the way and looks at the state of the art of the relevant solutions. This survey is different from the previous surveys on SDN-based RAN as it focuses on the salient problems and discusses solutions proposed within and outside SDN literature. Our main focus is on fronthaul, backward compatibility, supposedly disruptive nature of SDN deployment, business cases and monetization of SDN related upgrades, latency of general purpose processors (GPP), and additional security vulnerabilities, softwarization brings along to the RAN. We have also provided a summary of the architectural developments in SDN-based RAN landscape as not all work can be covered under the focused issues. This paper provides a comprehensive survey on the state of the art of SDN-based RAN and clearly points out the gaps in the technology.

cs.NI↗

arXiv2016

Bio-Inspired Resource Allocation for Relay-Aided Device-to-Device Communications

The Device-to-Device (D2D) communication principle is a key enabler of direct localized communication between mobile nodes and is expected to propel a plethora of novel multimedia services. However, even though it offers a wide set of capabilities mainly due to the proximity and resource reuse gains, interference must be carefully controlled to maximize the achievable rate for coexisting cellular and D2D users. The scope of this work is to provide an interference-aware real-time resource allocation (RA) framework for relay-aided D2D communications that underlay cellular networks. The main objective is to maximize the overall network throughput by guaranteeing a minimum rate threshold for cellular and D2D links. To this direction, genetic algorithms (GAs) are proven to be powerful and versatile methodologies that account for not only enhanced performance but also reduced computational complexity in emerging wireless networks. Numerical investigations highlight the performance gains compared to baseline RA methods and especially in highly dense scenarios which will be the case in future 5G networks.

cs.NI↗

arXiv2016

Optimal Virtualized Inter-Tenant Resource Sharing for Device-to-Device Communications in 5G Networks

Device-to-Device (D2D) communication is expected to enable a number of new services and applications in future mobile networks and has attracted significant research interest over the last few years. Remarkably, little attention has been placed on the issue of D2D communication for users belonging to different operators. In this paper, we focus on this aspect for D2D users that belong to different tenants (virtual network operators), assuming virtualized and programmable future 5G wireless networks. Under the assumption of a cross-tenant orchestrator, we show that significant gains can be achieved in terms of network performance by optimizing resource sharing from the different tenants, i.e., slices of the substrate physical network topology. To this end, a sum-rate optimization framework is proposed for optimal sharing of the virtualized resources. Via a wide site of numerical investigations, we prove the efficacy of the proposed solution and the achievable gains compared to legacy approaches.

cs.NI↗

arXiv2016

Interference-Aware Decoupled Cell Association in Device-to-Device based 5G Networks

Cell association in cellular networks is an important aspect that impacts network capacity and eventually quality of experience. The scope of this work is to investigate the different and generalized cell association (CAS) strategies for Device-to-Device (D2D) communications in a cellular network infrastructure. To realize this, we optimize D2D-based cell association by using the notion of uplink and downlink decoupling that was proven to offer significant performance gains. We propose an integer linear programming (ILP) optimization framework to achieve efficient D2D cell association that minimizes the interference caused by D2D devices onto cellular communications in the uplink as well as improve the D2D resource utilization efficiency. Simulation results based on Vodafone's LTE field trial network in a dense urban scenario highlight the performance gains and render this proposal a candidate design approach for future 5G networks.

cs.NI↗

arXiv2015

Realizing the Tactile Internet: Haptic Communications over Next Generation 5G Cellular Networks

Prior Internet designs encompassed the fixed, mobile and lately the things Internet. In a natural evolution to these, the notion of the Tactile Internet is emerging which allows one to transmit touch and actuation in real-time. With voice and data communications driving the designs of the current Internets, the Tactile Internet will enable haptic communications, which in turn will be a paradigm shift in how skills and labor are digitally delivered globally. Design efforts for both the Tactile Internet and the underlying haptic communications are in its infancy. The aim of this article is thus to review some of the most stringent design challenges, as well as proposing first avenues for specific solutions to enable the Tactile Internet revolution.

cs.NI↗

arXiv2015

QoE-Centric Localized Mobility Management for Future Mobile Networks

Mobility support in future networks will be predominately based on micro mobility protocols. Current proposed schemes such as Hierarchical Mobile IPv6 (HMIPv6) and more importantly Proxy Mobile IPv6 (PMIPv6) provide localized mobility support by electing a node within the network (topologically close to the Access Routers (AR)) to act as mobility anchor. Such schemes can significantly improve handover latency, as well as the end-to-end signalling overhead, but might entail scalability issues, which in some instances, do not fit adequately with the current explosion of mobile Internet traffic, and the evolutionary trend towards flat network architectures. The notion of using Distributed Mobility Management (DMM) allows for decentralization by anchoring nodes at their AR. The idea is that for sessions with duration less than the cell residence time efficient mobility can be supported. However, DMM might be highly suboptimal in instances where nodes perform multiple handovers during the session lifetime. Hence, one approach cannot be effective for all types of applications. In this paper a hybrid mobility management solution, integrated with a new routing scheme, is proposed. The scheme selects the most suitable mobility approach, centralized or distributed, by taking into account the flows' requirement, in terms of Quality of Experience (QoE), and new routing constraints. The results of optimization problem show that the proposed approach can achieve efficient resource utilisation, ease network congestion, and lead to a significant improvement in the QoE.

cs.NI↗