Browse the ITC publications of previous conferences below or visit our grouppage at bibsonomy.org
2020
Zhao, Xinlong and Song, Xianxin and Liu, Xin and Qin, Xiaoqi and Li, Hang
A Distributed Data Sampling and Relay Scheme for Obtaining Fresh Updates in Multihop Networks
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{xin20ITC32,
title = { A Distributed Data Sampling and Relay Scheme for Obtaining Fresh Updates in Multihop Networks },
year = { 2020 },
address = { Osaka, Japan },
author = { Zhao, Xinlong and Song, Xianxin and Liu, Xin and Qin, Xiaoqi and Li, Hang },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: New uses of wireless communication are envisioned to realize real-time control applications through networked monitoring. In such scenarios, the freshness of collected status updates is of critical importance for system performance, which can be quantified by the performance metric called Age of Information (AoI). In this paper, we apply this performance metric to study an age minimization problem in a wireless monitoring system where sensor nodes are employed to collect status information about their surroundings. The sampled data are forwarded to central controller via multiple hops. The problem formulation involvesjoint consideration of data sampling at source, flow routing, and link scheduling along the transmission path. The formulated problem falls in the form of an integer nonlinear program (INLP). To efficiently solve the problem, we construct AoI reward function to decouple the time coupling among sampling and relay decisions. Considering the system stability, we employ Lyapunov optimization technique to develop a distributed sampling and relay scheme. Simulation results show that the performance of our proposed scheme is competitive in terms of delivering status updates in a timely manner.
Gajić, Marija and Furdek, Marija and Heegaard, Poul
A Framework for Spatial and Temporal Evaluation of Network Disaster Recovery
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{marija20ITC32,
title = { A Framework for Spatial and Temporal Evaluation of Network Disaster Recovery },
year = { 2020 },
address = { Osaka, Japan },
author = { Gajić, Marija and Furdek, Marija and Heegaard, Poul },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: The support of vital societal functions requires a reliable communication network, especially in the presence of crises and disastrous events. Disasters caused by natural factors including earthquakes, fires, floods or hurricanes can disable network elements such as links and nodes and cause widespread disruption in end users connectivity to network services. Effects of disasters can vary over space and time due to disaster escalation and propagation. Network recovery from disasters requires understanding of both the spatial properties of the hazard at hand, and their temporal evolution. While the former has already been addressed in the literature, existing models and measures are unable to capture the temporal aspects of disaster recovery. This paper proposes a framework for spatial and temporal evaluation of network disaster recovery. It allows for modelling random spatial patterns of disasters in a geographical grid. The temporal aspects captured in our framework include changes due to the progression of a potentially shape-changing disaster across the affected area, as well as to the recovery actions of adaptive network reconfiguration and topology reconstruction undertaken by the network operator. The framework applicability is demonstrated on a content delivery network use case example, where we capture the evolving network performance in terms of the average shortest path length between the peers and the content replicas hosted by servers. By providing insights into the spatial and temporal effects of both disaster escalation and remediation measures, our proposed framework lays down the groundwork for flexible disaster modelling and recovery sequence optimization.
Hyytiä, Esa and Righter, Rhonda and Virtamo, Jorma
Admission Control to M/G/1 Subject to General Class-Specific Admission and Rejection Costs
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{esa20ITC32-2,
title = { Admission Control to M/G/1 Subject to General Class-Specific Admission and Rejection Costs },
year = { 2020 },
address = { Osaka, Japan },
author = { Hyytiä, Esa and Righter, Rhonda and Virtamo, Jorma },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: We consider the M/G/1 queue where job sizes become known upon arrival subject to a general cost structure. More specifically, we are interested in determining the optimal admission policy to the (size-aware) system with multiple jobclasses each having its own admission and rejection costs. The cost for admitting a job is a class-specific function of the waiting time. As a special case, we consider a deadline cost structure where admitting a job that will be late has a smaller cost than rejecting it. We analyse the system within the framework of Markov decision processes, and derive expressions that enable us to determine the size-aware value function, and the optimal class-specific admission control, as well as the resulting mean cost. The availability of the value function allows one to develop efficient dispatching policies for a system with heterogeneous parallel servers.
Lindner, Sebastian and Fisser, Leonard and Timm-Giel, Andreas
Coexistence of Shared-Spectrum Radio Systems through Medium Access Pattern Learning using Artificial Neural Networks
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{seb20ITC32,
title = { Coexistence of Shared-Spectrum Radio Systems through Medium Access Pattern Learning using Artificial Neural Networks },
year = { 2020 },
address = { Osaka, Japan },
author = { Lindner, Sebastian and Fisser, Leonard and Timm-Giel, Andreas },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: Abstract—Spectrum scarcity requires novel approaches for sharing frequency resources between different radio systems. Where coordination is not possible, intelligent approaches are needed, allowing a novel ”secondary” system to access unused resources of a legacy (primary) system without requiring modifications of this primary system. Machine Learning is a promising approach to recognize patterns of the primarysystem and adapt the channel access accordingly. In this contribution we investigate the capability of Feed-Forward Deep Learning and Long Short Term Memory (LSTM) Recurrent Neural Networks (RNNs) to detect communication patterns of the primary user. Therefore we take the example of a new aeronautical system (LDACS) coexisting with three different systems. Firstly the coexistence with the Distance Measurement Equipment (DME) providing a deterministic interference to the secondary user and secondly with two synthetic channel access patterns, realized by a 2-state Markov model, modeling a bursty channel access behavior, as well as through a sequential channel access model. It can be shown that the Markov property of a Gilbert-Elliot channel model limits the predictability; nonetheless, we show that the model characteristics can be fully learned, which could leverage the design of interference avoidance systems that make use of this knowledge. The determinism of DME allows an error-free prediction, and it is shown that the reliability of sequential access model prediction depends on the model’s parameter. The limits of Feed-Forward Deep Neural Networks are highlighted, and why LSTM RNNs are state-of-the-art models in this problem domain. We show that these models are capable of online learning, as well as of learning correlations over long periods of time. In the spirit of open science, the implementation files are made available in the conclusion.
Inoue, Koki and Suzuki, Dai and Kurita, Toshihiko and Imai, Satoshi
Cooperative Task Scheduling for Personal Identity Verification in Networked Systems
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{kok20ITC32,
title = { Cooperative Task Scheduling for Personal Identity Verification in Networked Systems },
year = { 2020 },
address = { Osaka, Japan },
author = { Inoue, Koki and Suzuki, Dai and Kurita, Toshihiko and Imai, Satoshi },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: Several working groups are coping with an ecosystem in which a user manages his/her own digital identity (ID) information among different organizations or companies in a decentralized manner. Accordingly, we developed a platform for trusted ID exchange called IDentitY eXchange. In the platform, the personal identity verification process will be realized by verifying credentials about users’ information issued by other organizations. Through this kind of ID cooperation, users can prove their ID online using the credentials and will no longer need to take procedure for every organization when updating their ID information registered there. To update their ID information among multiple organizations, users have to plan a schedule that represents an order of ID cooperation requests from each organization to other organizations. However, the organizations' policies to identify a user and relationships among the organizations make the scheduling problem a complicated one. In this study, we formulate a scheduling problem in the cooperative network as an integer linear program and propose a heuristic method based on the graph structure. Numerical experiments show that the heuristic approach has feasible scalability for practical use. Finally, we discuss other use cases of the proposed formulation, especially regarding logistics management, such as vehicle scheduling for transporting products from suppliers to consumers.
Hyytiä, Esa and Righter, Rhonda
Discrete-Time Analysis of the Blockchain Distributed Ledger Technology
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{esa20ITC32-1,
title = { Discrete-Time Analysis of the Blockchain Distributed Ledger Technology },
year = { 2020 },
address = { Osaka, Japan },
author = { Hyytiä, Esa and Righter, Rhonda },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: A dispatching system is a parallel server system where new jobs must be assigned to a server immediately upon arrival. We consider how to improve dispatching decisions by combining basic assignment policies that do not require state information into two levels: the first level dispatcher assigns jobs to a set of second level dispatchers, each with their own pool of servers. In each level or stage the decision is made by a static policy or Round-Robin principle. Such policies are fast and scale well as only local information is needed. The order of policies, whether RR should be first or second, gives rise to two dispatching policy classes, RATS and STAR. We show that the two-level STAR policy always outperforms RATS, and often outperforms any single-level policy. Moreover, STAR policies are robust across a range of parameter values and distributions for inter-arrival times and job sizes.
Arfaoui, Afaf and El-Azouzi, Rachid and Haddad, Majed and Sabir, Essaid
Flexible Network Slicing Assisted 5G for Video Streaming with Effective and Efficient Isolation
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{afa20ITC32,
title = { Flexible Network Slicing Assisted 5G for Video Streaming with Effective and Efficient Isolation },
year = { 2020 },
address = { Osaka, Japan },
author = { Arfaoui, Afaf and El-Azouzi, Rachid and Haddad, Majed and Sabir, Essaid },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: Network slicing assisted 5G is driven by providing a wide range of services that aims to satisfy various pre-service requirements. There is consensus that network slicing is a key enabler for the service-oriented 5G, that aims to cope with the increasing complexity of these networks. One of the major objectives of network slicing is to provide a different level of resource isolation, through resource abstraction and virtualization and the ability to efficiently share network resources. In this paper, we focus on video streaming services in the presence of other services with different QoS requirements. We propose a novel approach for resource sharing that provides interslice protection, flexibility, load-driven elasticity, and network efficiency. In particular, we design two-level multi-scale allocation schedulers for an efficient and low complexity RAN slicing by exploiting the characteristic of adaptive traffic such as video streaming service. Our mathematical analysis and simulation results confirm the benefits of resource abstraction and exhibit the added value of our solution
M. Navarro, Jose and Rossi, Dario
HURRA! Human readable router anomaly detection
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{jos20ITC32,
title = { HURRA! Human readable router anomaly detection },
year = { 2020 },
address = { Osaka, Japan },
author = { M. Navarro, Jose and Rossi, Dario },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: This paper presents HURRA, a system that aims to reduce the time spent by human operators in the process of network troubleshooting. To do so, it comprises two modules that are plugged after any anomaly detection algorithm: (i) a first attention mechanism, that ranks the present features in terms of their relation with the anomaly and (ii) a second module able to incorporates previous expert knowledge seamlessly, without any need of human interaction nor decisions. We show the efficacy of these simple processes on a collection of real router datasets obtained from tens of ISPs which exhibit a rich variety of anomalies and very heterogeneous set of KPIs, on which we gather manually annotated ground truth by the operator solving the troubleshooting ticket. Our experimental evaluation shows that (i) the proposed system is effective in achieving high levels of agreement with the expert, that (ii) even a simple statistical approach is able to extracting useful information from expert knowledge gained in past cases to further improve performance and finally that (iii) the main difficulty in live deployment concerns the automated selection of the anoma
Mohammadhossein Tabatabaee, Seyed and Le Boudec, Jean-Yves and Boyer, Marc
Interleaved Weighted Round-Robin: A Network Calculus Analysis
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{sey20ITC32,
title = { Interleaved Weighted Round-Robin: A Network Calculus Analysis },
year = { 2020 },
address = { Osaka, Japan },
author = { Mohammadhossein Tabatabaee, Seyed and Le Boudec, Jean-Yves and Boyer, Marc },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: Weighted Round-Robin (WRR) is often used, due to its simplicity, for scheduling packets or tasks. With WRR, a number of packets equal to the weight allocated to a flow can be served consecutively, which leads to a bursty service. Interleaved Weighted Round-Robin (IWRR) is a variant that mitigates this effect.We are interested in finding bounds on worstcase delay obtained with IWRR. To this end, we use a network calculus approach and find a strict service curve for IWRR. The result is obtained using the pseudo-inverse of a function. We show that the strict service curve is the best obtainable one, and that delay bounds derived from it are tight (i.e., worst-case) for flows of packets of constant size. Furthermore, the IWRR strict service curve dominates the strict service curve for WRR that was previously published. We provide some numerical examples to illustrate the reduction in worst-case delays caused by IWRR compared to WRR.
Modina, Naresh and El-Azouzi, Rachid and De Pellegrini, Francesco and Sadoc Menasche, Daniel
Joint Traffic Offloading and Aging Control in 5G IoT Networks
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{nar20ITC32,
title = { Joint Traffic Offloading and Aging Control in 5G IoT Networks },
year = { 2020 },
address = { Osaka, Japan },
author = { Modina, Naresh and El-Azouzi, Rachid and De Pellegrini, Francesco and Sadoc Menasche, Daniel },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: The widespread adoption of 5G cellular technology is becoming the major driver for the growth of IoT-based applications. In this paper we consider a Mobile Service Provider (MSP) that launches a smart city service based on IoT data readings. In order to serve IoT data collected across different locations, the MSP dynamically negotiates and rescales bandwidth and service functions. 5G network slicing functions are key to lease appropriate amount of resources over heterogeneous access technologies and different site types. Also, different infrastructure providers will charge slicing service depending on specific access technology supported across regional sites and IoT data collection patterns. We introduce a pricing mechanism based on Age of Information (AoI) to reduce the cost of MSPs. It provides incentives for devices to smooth traffic by shifting part of the traffic load from highly congested and more expensive locations to lesser charged ones, while meeting QoS requirements of the IoT service. The proposed optimal pricing scheme comprises a two-stage decision process, where the MSP determines the pricing of each location and devices schedule uploads of collected data based on the optimal uploading policy. Simulations show that the MSP attains consistent cost reductions tuning the trade-off between slicing costs and the AoI of uploaded IoT data.
Scholz, Dominik and Stubbe, Henning and Gallenmüller, Sebastian and Carle, Georg
Key Properties of Programmable Data Plane Targets
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{dom20ITC32,
title = { Key Properties of Programmable Data Plane Targets },
year = { 2020 },
address = { Osaka, Japan },
author = { Scholz, Dominik and Stubbe, Henning and Gallenmüller, Sebastian and Carle, Georg },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: We currently see a shift from fixed-function network devices with limited configurability towards network devices with a fully programmable processing pipeline. A prominent example of this development is P4 that provides a language and reference architecture model to design and program network devices. The core element of this reference model is the programmable matchaction table that defines the processing steps for the network packets. In this paper, we demonstrate that these tables, which we use to create our own modeling framework, are the key driver of device performance. P4-programmable devices come in a wide variety regarding their underlying hardware architecture, such as CPU-based systems or ASICs, as representatives of both ends of the spectrum. CPU-based P4 target platforms offer limited performance but are easily extensible. ASIC P4 targets have dedicated P4 processing pipelines with limited programmability but offer highly optimized performance. To reflect these fundamental differences, our modeling framework incorporates different approaches to accurately model and predict the performance of P4-enabled devices.
Diamanti, Alessio and Manuel Sanchez Vilchez, José and Secci, Stefano
LSTM-based radiography for anomaly detection in softwarized infrastructures
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{ale20ITC32,
title = { LSTM-based radiography for anomaly detection in softwarized infrastructures },
year = { 2020 },
address = { Osaka, Japan },
author = { Diamanti, Alessio and Manuel Sanchez Vilchez, José and Secci, Stefano },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: Legacy and novel network services are expected to be migrated and designed to be deployed in fully virtualized environments. Starting with 5G, NFV becomes a formally required brick in the specifications, for services integrated within the infrastructure provider networks. This evolution leads to deployment of virtual resources Virtual-Machine (VM)-based, container-based and/or server-less platforms, all calling for a deep virtualization of infrastructure components. Such a network softwarization also unleashes further logical network virtualization, easing multi-layered, multi-actor and multi-access services, so as to be able to fulfill high availability, security, privacy and resilience requirements. However, the derived increased components heterogeneity makes the detection and the characterization of anomalies difficult, hence the relationship between anomaly detection and corresponding reconfiguration of the NFV stack to mitigate anomalies. In this article we propose an unsupervised machine-learning data-driven approach based on Long-Short-Term-Memory (LSTM) autoencoders to detect and characterize anomalies in virtualized networking services. With a radiography visualization, this approach can spot and describe deviations from nominal parameter values of any virtualized network service by means of a lightweight and iterative mean-squared reconstruction error analysis of LSTM-based autoencoders. We implement and validate the proposed methodology through experimental tests on a vIMS proof-of-concept deployed using Kubernetes.
Kawasaki, Rio and Takano, Chisa and Aida, Masaki
Laplace Equation-Based High-Speed Autonomous Clustering for MANET
In 32th International Teletraffic Congress (ITC 32), Ph.D. Workshop. Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{rio20ITC32-WS,
title = { Laplace Equation-Based High-Speed Autonomous Clustering for MANET },
year = { 2020 },
address = { Osaka, Japan },
author = { Kawasaki, Rio and Takano, Chisa and Aida, Masaki },
booktitle = { 32th International Teletraffic Congress (ITC 32), Ph.D. Workshop }
}
Abstract: Autonomous decentralized clustering technologies to realize hierarchical route control for mobile ad hoc network terminals have been studied. The autonomous decentralized clustering must be able to realize clustering that can adapt to various movement patterns of terminals. Particularly, it must be effective even when user density is very high, such as around a train station during rush hour. However, existing methods have found that the cluster structure cannot be maintained when user density becomes very high. This paper proposes an autonomous decentralized clustering method that can maintain the cluster structure even if user density is high. We introduce an autonomous decentralized algorithm based on the Laplace equation that enables the method to quickly respond to user movement.
Yamauchi, Masaaki and Tanaka, Masahiro and Ohsita, Yuichi and Murata, Masayuki and Ueda, Kensuke and Kato, Yoshiaki
Modeling Home IoT Traffic using Users’ in-Home Activities for Detection of Anomalous Operations
In 32th International Teletraffic Congress (ITC 32), Ph.D. Workshop. Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{mas20ITC32-WS,
title = { Modeling Home IoT Traffic using Users' in-Home Activities for Detection of Anomalous Operations },
year = { 2020 },
address = { Osaka, Japan },
author = { Yamauchi, Masaaki and Tanaka, Masahiro and Ohsita, Yuichi and Murata, Masayuki and Ueda, Kensuke and Kato, Yoshiaki },
booktitle = { 32th International Teletraffic Congress (ITC 32), Ph.D. Workshop }
}
Abstract: In this paper, we modeled home IoT traffic based on users' in-home activities by the sensors and operations of home IoT devices. Then, we applied the model to the detection of anomalous operations. We evaluated our model by using a dataset obtained in an actual home environment. The results demonstrated that the detection using our method achieved $72.3%$ detections with less than $20.1%$ misdetections.
Hoshiyama, Takako
Performance Evaluation Algorithm for GI/G/m Queues
In 32th International Teletraffic Congress (ITC 32), Ph.D. Workshop. Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{tak20ITC32-WS,
title = { Performance Evaluation Algorithm for GI/G/m Queues },
year = { 2020 },
address = { Osaka, Japan },
author = { Hoshiyama, Takako },
booktitle = { 32th International Teletraffic Congress (ITC 32), Ph.D. Workshop }
}
Abstract: In this study, we propose an algorithm to evaluate the performance of a GI/G/m queuing network model using a general continuous probability distribution. Queueing models have traditionally been used for quantitative evaluation of packet and phone line congestion is reasonable to use predictions and evaluations using simulation models with stochastic processes. We propose a hybrid algorithm that uses Monte Carlo simulation to generate statistical data with continuous probability distribution. The data is analyzed based on the selected evaluation metric and the length of the queue using simulation and approximate solution methods. The algorithm is suited for environments that require fast verification.
Chagdali, Abdellatif and Elayoubi, Salah Eddine and Antonia, Maria Masucci and Simonian, Alain
Performance of URLLC Traffic Scheduling Policies with Redundancy
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{abd20ITC32,
title = { Performance of URLLC Traffic Scheduling Policies with Redundancy },
year = { 2020 },
address = { Osaka, Japan },
author = { Chagdali, Abdellatif and Elayoubi, Salah Eddine and Antonia, Maria Masucci and Simonian, Alain },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: In this paper, we study the performance of packet scheduling schemes for Ultra-Reliable Low-Latency Communications (URLLC) services. In order to ensure a low queuing time, we exploit the redundant coverage in many 5G Radio Access Networks (RAN) scenarios, where two frequency layers or Radio Access Technologies (RAT) are integrated. We consider
three packet scheduling and redundancy schemes, namely Jointhe-Shortest-Queue (JSQ), systematic Redundancy (RED), and redundancy with Cancellation upon completion (CAN). On the basis of queuing theory results, we develop expressions for the reliability, defined as the probability that the packet is transmitted before some given target delay. We show that RED has a good performance at low load, while JSQ is better when the load increases; CAN outperforms all other schemes. We then show how the obtained results can be used for the dimensioning of needed 5G radio resources and discuss the trade-off between performance and implementation complexity.
Zhang, Shan and Wang, Liudi and Zhang, Nu and Luo, Hongbing and Zhou, Sheng
Placement of Dynamic Content Items in Mobile Edge Caching
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{sha20ITC32,
title = { Placement of Dynamic Content Items in Mobile Edge Caching },
year = { 2020 },
address = { Osaka, Japan },
author = { Zhang, Shan and Wang, Liudi and Zhang, Nu and Luo, Hongbing and Zhou, Sheng },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: The emerging intelligent mobile applications such as augment/virtual reality and automatons driving require extensive context information, generating dynamic items whose content may change with time and environment. In this regard, the dynamic items should be refreshed to the newest versions timely, when cached at the network edge. The state-of-the art content placement methods mostly target on static items, and cache the popular items with priority to maximize the local hit rate. However, popular content items may have higher dynamics and require frequent cache refresh, which consumes more transmissions and degrades the system efficiency. In this work, we revisit the content placement problem at mobile edge, where the content dynamics and popularity are jointly considered when choosing items to cache. To this end, the effective hit rate of an cached content item is derived with respect to the corresponding request rate (i.e., popularity) and version update rate (i.e., dynamics). Specifically, the effective hit rate is proved to increase with the request arrival rate in an convex manner and decrease with the version update rate. Then, the popularityand-lifetime-aware placement (PALAP) scheme is proposed to maximize the overall effective hit rate of the local cache when providing dynamic content services. Extensive simulations are conducted on OMNeT++ platform to validate the obtained analytical results. In addition, simulation results show that the proposed scheme can double the cache hit rate compared with the conventional most-popular-first placement (MPFP) scheme, if the content requests are less concentrated. Accordingly, the average service delay can be reduced effectively, especially when the system is heavily loaded.
Aida, Masaki and Hashizume, Ayako and Takano, Chisa and Murata, Masayuki
Polarization Model of Online Social Networks Based on the Concept of Spontaneous Symmetry Breaking
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{mas20ITC32,
title = { Polarization Model of Online Social Networks Based on the Concept of Spontaneous Symmetry Breaking },
year = { 2020 },
address = { Osaka, Japan },
author = { Aida, Masaki and Hashizume, Ayako and Takano, Chisa and Murata, Masayuki },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: The spread of information networks has not only made it easier for people to access a variety of information sources but also greatly enhanced the ability of individuals to disseminate information. Unfortunately, however, the problem of slander in online social networks shows that the evolving information network environment does not necessarily support mutual understanding in society. Since information with particular bias is distributed only to those communities that prefer it, the division of society into various opposing groups is strengthened. This phenomenon is called polarization. It is necessary to understand the mechanism of polarization to establish technologies that can counter polarization. This paper introduces a fundamental model for understanding polarization that is based on the concept of spontaneous symmetry breaking; our starting point is the oscillation model that describes user dynamics in online social networks.
Suer, Marie-Theres and Thein, Christoph and Tchouankem, Hugues and Wolf, Lars
Reliability and Latency Performance of Multi-Connectivity Scheduling Schemes in Multi-User Scenarios
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{marie20ITC32,
title = { Reliability and Latency Performance of Multi-Connectivity Scheduling Schemes in Multi-User Scenarios },
year = { 2020 },
address = { Osaka, Japan },
author = { Suer, Marie-Theres and Thein, Christoph and Tchouankem, Hugues and Wolf, Lars },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: Industrial applications such as closed-loop control require communication systems which provide low latency and high reliability. Current wireless communication systems are not able to fulfill all strict requirements posed by industrial applications. Multi-Connectivity (MC), i.e. using multiple communication paths at the same time, is a promising measure to enhance latency and reliability of wireless communication systems. Different scheduling schemes, i.e. Load Balancing (LB), Packet Duplication (PD) and Packet Splitting (PS), can be utilized to distribute packets over multiple paths. After having identified the impact of link homo- and heterogeneity, SNR regime and link correlation in previous work, the impact of network load and capacity on the different MC scheduling schemes is evaluated in this paper. Measurements in private LTE and WiFi 802.11ac networks are performed for different number of users. Our evaluations support that network load and capacity have a strong impact on latency and reliability performance of MC scheduling schemes. In low load scenarios, PD or PS achieve the best performance, for scenarios with low link correlation and high link correlation, respectively. In scenarios where the network load is close to the capacity limit on the other hand, LB provides the lowest latency and highest reliability. The capacity limit in the multiple user scenario strongly depends on the channel access scheme and its ability to deal with increasing number of users. The insights gained in this work can be used to develop dynamic MC schedulers that adapt the MC scheduling scheme based on SNR regime, link correlation and network load in order to achieve low latency and high reliability in scenarios with varying radio and network conditions.
Jiang, Zhiyuan
Round-Robin is Provably Near-Optimal for Minimizing Age with HARQ over Heterogeneous Unreliable Multiaccess Channels
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{zhi20ITC32,
title = { Round-Robin is Provably Near-Optimal for Minimizing Age with HARQ over Heterogeneous Unreliable Multiaccess Channels },
year = { 2020 },
address = { Osaka, Japan },
author = { Jiang, Zhiyuan },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: In a heterogeneous unreliable multiaccess network, wherein terminals share a common wireless channel with distinctive error probabilities, existing works have showed that a persistent round-robin (RR-P) scheduling policy (i.e., greedy policy) can be arbitrarily worse than the optimum in terms of Age of Information (AoI) under standard Automatic Repeat reQuest (ARQ), and one must resort to Whittle’s index approach for optimal AoI. In this paper, practical Hybrid ARQ (HARQ) schemes which are widely-used in today’s wireless networks are considered. We show that RR-P is very close to optimum with asymptotically many terminals in this case, by explicitly deriving tight, closed-form AoI gaps between optimum and achievable AoI by RR-P. In particular, it is rigorously proved that for RR-P, under HARQ models concerning fading channels (resp. finiteblocklength regime), the relative AoI gap compared with the optimum is within a constant of ($\sqrte−1)^2/4\sqrte≈ 6.4%$ (resp. $6.2%$ with error exponential decay rate of $0.5$). In addition, RR-P enjoys the distinct advantage of implementation simplicity with channel-unaware and easy-to-decentralize operations, making it favorable in practice.
Bouganim, Theo and Araldo, Andrea and Lavignotte, Antoine and Oussedik, Nessim and Guez, Gabriel
The Cost of Learning Fast with Reinforcement Learning for Edge Cache Allocation
In 32th International Teletraffic Congress (ITC 32), Ph.D. Workshop. Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{the20ITC32-WS,
title = { The Cost of Learning Fast with Reinforcement Learning for Edge Cache Allocation },
year = { 2020 },
address = { Osaka, Japan },
author = { Bouganim, Theo and Araldo, Andrea and Lavignotte, Antoine and Oussedik, Nessim and Guez, Gabriel },
booktitle = { 32th International Teletraffic Congress (ITC 32), Ph.D. Workshop }
}
Abstract: We study data-driven cache allocation in Multi-Tenant Edge Computing: a Network Operator (NO) owns storage at the Edge and dynamically allocates it to third party application Content Providers (CPs). CPs can cache a part of their catalog and satisfy locally users' requests, thus reducing inter-domain traffic. The objective of the NO is to find the optimal cache allocation, which minimizes the total inter-domain traffic bandwidth, which constitutes an operational cost. Since CPs' traffic is encrypted, NO' s allocation strategy is based solely on the amount of traffic measured.
In this exploratory work, we solve this problem via Reinforcement Learning (RL). RL has mainly been intended to be trained in simulation, before applying it in real scenarios. We instead employ RL online, training it directly while the system is hot and running. An important factor emerges in this case: in order to learn the optimal cache allocation, the NO needs to perturb the allocation several times and measure how the inter-domain traffic changes; when perturbing the allocation, the NO has to pay a perturbation cost. While it has no physical meaning in simulation, it cannot be ignored in a hot and running system. We explore in this work the trade-off between perturbing a lot the system in order to learn a good allocation faster, or learning slower to reduce the perturbation cost. We show results from simulation and make the entire code available as open-source.
Wang, Lehan and Sun, Jingzhou and Zhou, Sheng and Niu, Zhisheng
Timely Status Update Based on Urgency of Information with Statistical Context
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{leh20ITC32,
title = { Timely Status Update Based on Urgency of Information with Statistical Context },
year = { 2020 },
address = { Osaka, Japan },
author = { Wang, Lehan and Sun, Jingzhou and Zhou, Sheng and Niu, Zhisheng },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: Real-time status update has recently brought the widespread attentions, especially in the field of remote control. Accordingly, a metric named Urgency of Information (UoI) has been proposed to capture the timeliness of information and the evolution of the context at the same time. It is defined as the product of the context-aware weight and the cost function of the estimation error. However, the metric itself does not indicate how to determine the contextual weight. Based on UoI, this paper proposes a dynamic threshold-based scheduling policy which only requires the conditional expectation of the weight in the next time slot by Lyapunov method. When the environment is unknown, We also use the online algorithm SARSA to obtain a policy. The simulation results show that both policies are near-optimal when the update resources are relatively rich, and the SARSA-based policy works better under the tight resources constraint over the Lyapunov method.
Kline, Jeffery and Aelony, Avram and Carpenter, Brian and Barford, Paul
Triangulated Rank-ordering of Web domains
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{jef20ITC32,
title = { Triangulated Rank-ordering of Web domains },
year = { 2020 },
address = { Osaka, Japan },
author = { Kline, Jeffery and Aelony, Avram and Carpenter, Brian and Barford, Paul },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: The relative popularity of web sites, as expressed in published rankings, is of fundamental value in many contexts including search, advertising and research. In this paper, we consider the surprisingly challenging problem of generating consistent and reliable web site rankings based on unique visitors per day. We illustrate the challenge this represents using data from three large and independently-sourced Internet user panels. We begin by showing that generating a website ranking based simply on the observed unique daily visitors produces highly inconsistent rankings–even among the most popular sites. To mitigate the problems of bias and measurement error, we introduce a general methodology that identifies ``canonical panelists'': an abstract class of user that exhibits consistent behavior across panels. Our definition is based on the epistemological technique of triangulation, which refers to observing the same object from multiple perspectives at the same moment in time. We show that panelists in the canonical class exhibit desirable characteristics including improved persistence. Most significantly, we show that defining a domain’s rank as a function of the aggregate behavior of canonical panelists improves overall alignment of rankings across all three of our panels.
Favale, Thomas and Trevisan, Martino and Drago, Idilio and Mellia, Marco
α-MON: Anonymized Passive Traffic Monitoring
In 32th International Teletraffic Congress (ITC 32). Osaka, Japan 2020
[BibTeX]
[Abstract]
[Download]
[BibSonomy]
@inproceedings{tho20ITC32,
title = { α-MON: Anonymized Passive Traffic Monitoring },
year = { 2020 },
address = { Osaka, Japan },
author = { Favale, Thomas and Trevisan, Martino and Drago, Idilio and Mellia, Marco },
booktitle = { 32th International Teletraffic Congress (ITC 32) }
}
Abstract: Packet measurements are essential for several applications, such as cyber-security, accounting and troubleshooting. They, however, threaten privacy by exposing sensitive information. Anonymization has been the answer to this challenge, i.e., replacing sensitive information by obfuscated copies. Anonymization of packet traces, however, comes with some drawbacks. First, it reduces the value of data. Second, it requires to consider diverse protocols because information may leak from many nonencrypted fields. Third, it must be performed at high speeds directly at the monitor, to prevent private data from leaking, calling for real-time solutions. We present α-MON, a flexible tool for privacy-preserving packet monitoring. It replicates input packet streams to different consumers while anonymizing values according to flexible policies that cover all protocol layers. Beside classic anonymization mechanisms such as IP address obfuscation, α-MON supports α-anonymization, a novel solution to obfuscate values that can be uniquely traced back to limited sets of users. Differently from classic anonymization approaches, α-anonymity works on a streaming fashion, with zero delay, operating at high-speed links on a packet-by-packet basis. We evaluate α-MON performance using packet traces collected from an ISP network. Results show that it enables α-anonymity in real-time. α-MON is available to the community as an open-source project.
