Showing papers presented at "IEEE International Conference on Advanced Infocomm Technology in 2012"

Mantle: A Novel DOSN Leveraging Free Storage and Local Software

Abstract: We present Mantle, a novel framework for decentralized Online Social Networks (OSN). Mantle defines a privacy-preserving OSN kernel and can leverage any storage facilities for storing user data. Mantle is designed to execute all OSN specific logic on user devices. It supports the development of third-party applications and mediates all actions of the latter to protect user data, metadata and interactions. In this work, we focus on Mantle’s interaction and notification mechanisms, which use publish/subscribe model.

Evaluation of SIP Call Setup Delay for VoIP in IMS

Abstract: IP Mulimedia Subsystem (IMS) is an architecture that can provide innovative solutions for multimedia services deployment regardless the type and network topology. New delivered value added services require significant changes to the multi-services IP network design, including the activation of multiple functions such as quality of service (QoS) capabilities, security mechanisms, multicast routing, etc. To ensure that services delivery meets the high expectations of end users, factors affecting QoS or user’s quality of experience (QoE) must be properly considered. This paper will give a study of a the ’SIP Call Setup Delay’ metric which will serve us to accomplish the QoE measurement in the case of VoIP sessions carried over IPv4 and IPv6 IMS networks.

Improve Prefetch Performance by Splitting the Cache Replacement Queue

Abstract: The performance of a prefetch cache is dependent on both the prefetch technique and the cache replacement policy. Both these algorithms execute independently of each other, but they share a data structure - the cache replacement queue. This paper shows that even with a simple prefetch technique, there is an increase in hit rate when the LRU replacement queue is split into two equal sized queues. A more significant performance improvement is possible with a sophisticated prefetch technique and by splitting the queue unequally.

CO2 Laser Writing of Long-Period Fiber Gratings in Polarization-Maintaining Fiber under Tension

Abstract: We demonstrate CO2 laser writing of long-period fiber gratings (LPFGs) in a panda polarization-maintaining (PM) fiber under external tension. The transmission spectrum of the LPFGs depends on the irradiation direction of the writing beam with respect to the principal axes of the fiber. When the irradiation direction is along the fast axis of the PM fiber, the transmission spectra of the LPFG has two more rejection bands, compared to the transmission spectra of the LPFG written in the PM fiber without tension. Our results provide a better understanding of the formation of LPFGs written by CO2 laser.

Energy Efficient Aggregation in Wireless Sensor Networks

Abstract: Wireless sensor networks are getting more and more attention from researchers and industrial communities. They consist of distributed event-based sensors and a base station for data processing. These sensor networks have severe energy constraints. Data aggregation mechanisms are used to decrease the power consumption in these networks by combining several messages. In this paper, we present an energy efficient data aggregation mechanism. In this mechanism, the node which is aggregating and transmitting the message to the base station is chosen based on its energy and proximity to the base station. The nodes decide to participate in the aggregation process based on a participation relevance function. The efficiency of this aggregation mechanism is shown through simulation results in terms of network lifetime and reception rate.

Collective Intelligence Based Place Recommendation System

Abstract: Recently, various place recommendation systems are being used by users under social network environment. As the place service gets composed of only the limited categories in case of being made up of the expert knowledge, it has a factor of lacking the information desired by users. In this study, a collective intelligence based QnA information in which many of users participate in was used in order to supplement the details lacking in the place information that was being built up through the existing expert knowledge. Based on these, this paper proposes a place recommendation service using the collective intelligence and heterogeneous information source convergence. Also, the fact that the method we have proposed is able to provide the place recommendation service in a useful way under the location based social network environment was shown through the experiment.

Descriptional Entropy: Application to Security Software Analysis

Abstract: We propose here a tool we call descriptional entropy, that generalizes the classical Shannon Entropy. The tool is interesting per se and truly generic, i.e. it can be used with any files, it allows to define the ”complexity” of a sequence, whatever is this sequence, it could be anything, from pictures to numerical data files or genetic sequences or softwares. It can be used for example to index, to sort or compare a set of files (or fragments of files).

Overcoming the Deficiencies of Collaborative Detection of Spatially-Correlated Events in WSN

Abstract: Collaborative WSN are known to efficiently correlate sensors’ events to detect spatially-correlated events with a low latency. However, because of the drastic reduction of messages sent to the network administrator, it is difficult for him/her to understand in what state the network is. In this paper, we propose a modality-agnostic collaborative detection of spatio-temporally correlated events that drastically lowers power consumption by both reducing and localising communication. This contribution can then be used to expose better auditing capabilities that overcome the problems usually found in collaborative networks. These capabilities can in turn be used by both the administrator and the network itself to, for instance, automatically react to faulty sensors. Experiments validate the interest of our proposal in an intrusion detection scenario.

GeoSUZ: A Geocast Routing Protocol in Sub-ZORs for VANETs

Abstract: Vehicular Ad Hoc Networks (VANETs) are characterized by highly speed nodes, highly dynamic topology and frequent links disconnection. This raises a number of challenges especially in the field of data dissemination. In our work, we focus on Geocast routing in VANETs which consists of routing a message from a unique source vehicle to all vehicles located in a well geographically defined destination area called ZOR (Zone Of Relevance). In literature, ZORs are often assumed to be of any form and still chosen according to the scenarios and motivation needs of the authors. In this paper, we consider a ZOR as a set of sub-ZORs and we choose simple geometrical forms of sub-ZORs so that they will be easy to implement and to represent mathematically. We provide a geometrical vision angles based technique that allows to know if two sub-ZORs are in the same direction in order to send them a single message and hence, reduce the messages overhead. We then introduce a new routing protocol in Sub-ZORs (GeoSUZ) for VANETs based on our geometrical vision angles and greedy forwarding techniques. We compare GeoSUZ to GPSR routing protocol (Greedy Perimeter Stateless Routing for Wireless Networks) [11] and some numerical results show a significant gain in terms of number of messages sent over the network.

Adaptative Delay Aware Data Gathering Strategy for Wireless Sensor Networks

Abstract: Extending the network lifetime is a challenging problem in wireless sensor networks. Using additional mobile agents called data collectors in order to gather data from sensor nodes and carry them to the sink, reveals a convenient solution. Intuitively, a great sensor nodes energy conservation can be achieved by single hop data gathering strategies. However, these strategies may lead to significantly increase the data latency in the network. Previous studies propose to introduce local relay transmissions in order to reduce the data gathering time. Most of these strategies rely on a static parameter in order to choose the data collector trajectory. In this paper, we focus on a novel data gathering strategy that aims at dynamically selecting the data collector trajectory according to the required data delivery delay deadline. In other words, the data collector may change its trajectory according to the urgency of the gathered data. For this purpose, we propose an algorithm for coloring the network area into red and green disks, according to the delay deadline required by the sensor nodes. Only the red disks must be visited during a tour of the data collector. Simulation results demonstrate that the proposed scheme significantly outperforms static cluster based gathering strategies by efficiently adapting the tolerated data latency to the consumed energy rate.

On the Security of Lv et al.’s Three-Party Authenticated Key Exchange Protocol Using One-Time Key

Abstract: In 2012, Lv et al. proposed a new three-party authenticated key exchange (3PAKE) protocol using one-time key which aims to achieve more efficiency with the same security level of other existing 3PAKE protocols. Lv et al. claimed that their proposed 3PAKE protocol is secure against various known attacks. However, this paper points out Lv et al.’s 3PAKE protocol is still vulnerable to off-line dictionary attacks unlike their claim.

UNIVERSALLY: A Context-Aware Architecture for Multimedia Access in Digital Homes

Abstract: The objective of this work is to improve the user’s experience when using media resources in a digital home. We consider the use and access of media in any context and particularly in a resource-constrained context such as using mobile terminals through a wireless access. We focus our work on the Digital Living Network Alliance architecture since it knows a high penetration in homes but with only a small amount of users due to the complexity of the architecture. We propose UNIVERSALLY that aims to simplify the complexity of digital home architectures and optimize them. We consider limited terminals and users that are not always aware about setting up their devices. Unnecessary network resources and periodic traffic are avoided when no media resource is in use. The context awareness improves the user’s experience by adapting media formats and avoiding the traffic of streaming resources that are not understood by the terminal.

On Selective Placement for Uniform Cache Objects

Abstract: For several decades, numerous studies for the cache replacement algorithms have been performed in order to enhance cache performance. But, as the performance of a new algorithm has improved, the complexity of the algorithm has gradually increased. In the end, in most of practices, the LRU algorithm has been still successfully used, because of its good hit ratio, low complexity of O(1), and so on.

A user-centric network architecture for sustainable rural areas

Abstract: Heterogeneous networks offer interesting solutions to problems encountered in user-centric network architecture. Encompassing various communication technologies, they offer great potentials for addressing some of the challenges that ICT based remote services face. In this work, we focus on their deployment in rural areas and developing countries. More specifically, we examine how heterogeneous networks can be used in a user-centric architecture to improve application interactivity, interoperability, and network utilization. Restrictions of each constituent technology cause the architecture to have an upper limit in supporting simultaneous interactive applications. To investigate these limits, and to identify potential enhancements, we study an interactive education model. The considered interactivity, facilitated by heterogeneous networks, is between clients in rural areas and servers in an urban area. The underlying model architecture involves several communication technologies such as WiFi, Ethernet, WiMAX, and UMTS. Several scenarios relevant to this architecture are simulated and analyzed. For each scenario, videoconferencing sessions are initiated with variant number of users. The performance of the architecture in terms of capacity and key QoS parameters such as delay variation, end-to-end delay, and packet loss is evaluated. The results show that for most typical situations, WiFi-WiMAX combinations outperform other integrations.

Signal Detection Based on Maximum-Minimum Eigenvalue in Rician Fading Channel

Abstract: Signal detection is an essential section of Cognitive Radio. In this paper, a method is adopted based on the eigenvalue of the autocorrelation matrix of the received signal. This method is based on the random matrix theory, using the ratio from the maximum eigenvalue to the minimum eigenvalue as the test statistic to detect the existence of primary users, so as to improve the spectrum efficiency. Simulation based on the signal detection in the Rician fading channel, the results show that the performance of the adopted method is better than energy detection in the same fading environment.

SAVI - A Model for Video Workload Generation Based on Scene Length

Abstract: The development of models for synthetic network traffic generation is essential for performance evaluation of transmission systems The workload generation models are employed in network simulations, testing and performance predictions This article presents a new model for workload generation of video encoded with MPEG, called SAVI (Scene Aware Video Workload Generation Model) This model provides an explanation for the short and long range dependence of video traffic, and was designed based on movie scene length The resulting model presents good possibilities to support generalizations according to the movie style Besides, SAVI is easier to parametrize if compared with the available models and presents low computational complexity The model was developed by analyzing the traces of movies publicly available and has been tested through computer simulations The synthetic traffic generated mimics properly the characteristics of real traffic, including the short and long range dependence

Traffic Engineering Approaches in P2P Environments

Abstract: Nowadays, P2P applications proliferate in the Internet with distinct utilization contexts, being also an attractive model for the deployment of advanced Internet services. However, there are several undesirable effects that are caused by such applications, raising coexistence problems with Internet Service Providers (ISPs). In this context, using as case study BitTorrent like applications, this work explores a collaborative framework allowing that advanced efforts could be carried on between P2P applications and network level entities. In order to illustrate such framework, several Traffic Engineering (TE) mechanisms are devised in order to align some P2P dynamics with particular objectives pursued by network administrators. The simulation results show that both the proposed framework and corresponding illustrative mechanisms are viable and can effectively foster future research efforts within this field.

Extension of Path Computation Element (PCE) Framework for Resource Provisioning Based on User Profile in Dynamic Circuit Network

Abstract: Dynamic circuit network (DCN) refers to an on-demand virtual circuit (VC) service in which a user makes a request for a circuit in advance via a Web page. On-demand Secure Circuits and Advance Reservation System (OSCARS) is well-known DCN software. All requests in DCN/OSCARS are guaranteed bandwidth. Therefore, the quality of service (QoS) can be determined by the request blocking probability (RBP). Several QoS provisioning mechanisms, such as the bandwidth allocation policy (BAP) and preemption, have been proposed in the literature in order to ensure low RBPs for high-priority users. This paper proposes the extension of path computation elements (PCEs) in OSCARS version (v.) 6, called the UP-PCE, which considers the user attributes in its computation. Our extended OSCARS with the UP-PCE can determine two mechanisms: the control of the available topology and the BAP.

A low-complexity and high-performance beamforming scheme for mmWave WPAN systems

Abstract: This work presents an efficient technique which significantly enhances the performance of current codebook-based beamforming scheme for mmWave WPAN communication systems. The proposed technique, named CB-AoDE, is firstly applied to the transmit side, and estimates the angle-of-departure (AoD) based on the feedback SNRs from the receive side during the beam-search process. The estimated AoD is then used to generate a more accurate transmit beam vector. The same procedure can be as well applied to estimate the angle-of-arrival (AoA) which helps generate the corresponding receive beam vector. Simulation results show that the proposed technique performs averagely 1.8dB, 1.4dB and 1dB better than the codebook-based scheme in AWGN, light-of-sight (LOS), and Non-LOS (NLOS) channels, respectively. Moreover, the proposed one only needs one DAC and ADC at transmit and receive sides, respectively. The hardware cost and computational complexity are thus tremendously reduced.

Loss Probability and Delay of Finite-Buffer Queues with Discrete ON-OFF Markovian Arrivals and Geometric Service Times

Abstract: Discrete-time queues are commonly used in the performance analysis of many computer and communications systems. As the independent and identically distributed (i.i.d.) arrival traffic model frequently adopted (due to its analytical tractability) in the literature does not capture some important properties of real-world traffics, such as self-similarity, long-range dependence, and burstiness, it is more realistic to model the arrival traffic as a time-correlated process.

Towards Urban Vehicular Networking

Abstract: Inter-Vehicle Communication (IVC) is attracting considerable attention from the research community and the automotive industry, where it is beneficial in providing Intelligent Transportation System (ITS) as well as assistance services for drivers and passengers In this context, Vehicular Networks are emerging as a novel category of wireless networks, spontaneously formed between moving vehicles equipped with wireless interfaces that could have similar or different radio interface technologies, employing short-range to medium-range communication systems The distinguished characteristics of vehicular networks such as high mobility, potentially large scale, and network partitioning introduce several challenges, which can greatly impact the future deployment of these networks In this Keynote, we focus on IVC in urban environments Our main goal is to discuss the feasibility of Vehicular Ad hoc Networking for road traffic estimation, vehicular data routing as well as information dissemination in urban environments This will be discussed while considering the characteristics and constraints imposed by vehicular networks and their applications

Backhaul Topology Design and Weighted Max-Min Fair Capacity Allocation in Wireless Mesh Networks

Abstract: In this paper we focus on the Multi-Channel, Multi-Radio Wireless Mesh Network design issue. We propose an effective planning approach which considers two folds. First, to determine the backhaul topology by interconnecting wirelessly the mesh routers (MRs). Second, to share based on the weighted max-min fair scheme the available capacity between connected mesh routers. We formulate both aspects as a global problem. Due to the high computational complexity, we deal with the global problem by decoupling it into two separated problems: bakhaul topology formation and the weighted max-min fair capacity allocation problems. These latter aspects are treated sequentially where we adopt local search heuristics in order to address the scalability issue. Numerical results are presented to demonstrate the effectiveness of our proposition. Thus, we evaluate the proposed heuristics according to the exact solution of the global problem, and we discuss their opportunities.

Performance Evaluation of Fairness between High-Speed TCPs in Wireless Environment

Abstract: Among various TCP congestion controls proposed for high delay-bandwidth networks thus far, CUBIC and Compound TCP have been implemented as default congestion control in Linux and Microsoft Windows OS, respectively. CUBIC is loss-based high-speed congestion control. Compound TCP is constructed with NewReno TCP and delay-based high-speed congestion control. When these two versions of TCP shares bottleneck link, their aggressiveness and different behaviors might fall into unfair condition. In current network environment, bandwidth of wired networks is sufficiently large and a wireless access network can be one of the most likely bottleneck points. This wireless access network has different feature than wired networks, i.e. wireless caused loss which occurs independently of congestion. In this paper, we evaluate fairness among CUBIC and Compound TCP sessions sharing high-speed wireless channel. Our simulation results show that Compound TCP is seriously suffered by CUBIC with low loss rate and this unfair condition is relieved with increase of packet loss rate because loss based CUBIC is gradually degraded by wireless caused packet loss.

Providing Reliability for Transactional Mobile Agents

Abstract: A transactional agent is a mobile agent that migrates from a site to another one in order to execute a distributed transaction assigned by a user. Works on transactional mobile agents have identified two problems that can not be solved by the agent alone. The first one is related to the reliability. The lack of a fault tolerant infrastructure and methodologies that address fault tolerant execution of mobile agents highlights a major drawback of this technology. The second problem for which the agent needs assistance is related to the atomic validation of the transaction. In this paper we address the mobile agent fault tolerance and the transactional support.

Highway Preventive Congestion Control through Input Regulation in Vehicular Networks

Abstract: In this paper, we developed some insights into the design of a preventive congestion control applied in vehicular network. The proposed mechanism, implemented at the highway entrances, regulates input traffic and performs vehicular traffic shaping. Our congestion control mechanism deals with different classes of vehicles and is based on a specific priority ticket pool scheme with queue-length threshold scheduling policy, tailored to vehicular networks. Mathematical model based on the embedded Markov chain method shows the benefits of the proposed scheme and highlights the impact of the system parameters on the overall performance. Our technique meets drivers expectations as they will experience bounded performance parameters and limited burst vehicular traffic size.

Mechanisms to Locate Non-cooperative Transmitters in Wireless Networks

Abstract: This paper proposes new mechanisms for locating a non-mobile transmitter in wireless communication or sensor networks. They rely on a set of trusted cooperative receivers that are able to measure the residual strengths of the received signals from the transmitter. These mechanisms cannot rely on the information provided by the transmitter since the latter can be malicious. Laurendeau and Barbeau [1] proposed such a localisation mechanism. Unfortunately, it uses an approximation algorithm which is too restrictive. The difficulty comes for the fact that no good approximation of the difference of two log-normal variables exists. Thus, we propose new algorithms that rely either on better approximations or a geometric interpretation of the problem.

Predictive Sink Mobility for Target Tracking in Sensor Networks

Abstract: In sensor networks, the information is generated by sensors deployed in a geographic area, and sent to a node called ”sink” (or gateway). Since the node’s energy is battery limited, an efficient management of this resource affects the network’s lifetime. Our work presents a new approach called RPL (Repositioning, Prediction, Localization), that aims to extend the lifetime of the sensor network for a target tracking application. This is realized by switching sensors between active/sleep states and moving the sink close to the target’s future position. The movement of the sink reduces the energy needed for a packet transmission, and minimizes the number of hops between the sink and the emitting sensors. The proposed scheme is validated through simulations.

Energy Optimization of Mesh Access Networks

Abstract: This paper deals with the minimization of energy consumption in mesh access networks. As proposed in 3GPP, future wireless access networks will be based in part on mesh technology. Indeed, as all access points cannot be connected by optic fibers to the core networks, we propose to realize a mesh network to connect altogether hotspots and femtocells. To optimize the energy consumption we have developed a Start-and-Stop mechanism permitting to turn-off all unused physical machines. Moreover, we developed a virtual environment to work with virtual machines instead physical machines. This permits to move virtual resources from one physical point to another one. Finally, we describe the piloting system based on the autonomic paradigm, and associated with a knowledge plane. This piloting system is able to minimize the energy consumption in future mesh access networks.

Adaptive Contention Window for Zone-Based Dissemination of Vehicular Traffic

Abstract: Inter-Vehicular Communication (IVC) is very promising for advanced applications in traffic systems. However, current efforts and implementations still require more attention in order to turn such applications into reality. The data dissemination and the vehicular communication are two key concerns in order to make more feasible alternative traffic information systems. In this paper we focus on the traffic data dissemination and the inter-vehicle communication applied to our traffic guidance architecture presented in [5]. We present a twofold proposal: a mechanism for V2I, V2V traffic data dissemination and a simple but efficient mechanism to enhance broadcast communication. We show through extensive evaluations that our composite solution tends to be efficient under the design of our specific application guidelines.

Supporting Data Center Management through Clustering of System Data Streams

Abstract: Aggregating large data sets related to hardware and software resources into clusters is at the basis of several operations and strategies for management and control. High variability and noise characterizing data collected from system resources monitoring prevent the application of existing solutions that are affected by low accuracy and scarce robustness.