Internet and Distributed Computing Systems

Frontmatter

Implementation of Self-adaptive Middleware for Mobile Vehicle Tracking Applications on Edge Computing

Abstract

Unstructured data gathered from various IoT sensors is rapidly increasing due to inexpensive electronic devices and high-speed networks. On the other hand, mobile edge computing (MEC) is an attractive data processing method that can shorten the communication distance and reduce the latency of computation-intensive tasks by distributing data to the edge servers close to the users, unlike processing data on clouds that are located far from users. In the present paper, we propose a specialized self-adaptive middleware for reconfiguration of image/video contents for adaptation to changes with the movement of a vehicle. The key concept behind this approach is to introduce the rule-based relocation of objects among sensor devices, edge servers, and existing clouds as a basic adaptation mechanism to recognize and track mobile vehicles. Experimental results show that tracking precision with a state-of-the-art tracker is up to 89% for MEC.

Jingtao Sun, Cheng Yang, Tomoya Tanjo, Kazushige Sage, Kento Aida

Towards the Succinct Representation of m Out of n

Abstract

Combinations of n objects taken m at the time are ubiquitous in a wide range of combinatorial problems. In this paper, we introduce a novel approach to generate combinations from given integer numbers by using a gradient-based algorithm through plural number of CPU and GPU processors. The time complexity is bounded by \(O(m^2)\) when using a single processor, and bounded by \(O(m \log m)\) when using at most \(O(m/\log m)\) processors. Relevant computational experiments confirmed the practical efficiency within computationally allowable limits. Our approach offers the building blocks to represent combinations with succinct encoding and complexity being independent of n, which is meritorious when n is very large, or when n is time-varying.

Victor Parque, Tomoyuki Miyashita

Extending the Advisor Concept to Deal with Known-Ahead Transportation Tasks

Abstract

The efficiency improvement advisor can improve the quality of the emergent solutions created by self-organizing emergent multi-agent systems by identifying recurring tasks. In particular, those recurring tasks that the agents in the self-organizing system do not solve well become valuable knowledge because this knowledge is used to create exception rules for the appropriate agents that improve their task-fulfilling behavior. In this paper, we present an extension to the advisor that allows it to use certain knowledge about future tasks in addition to the (somewhat uncertain) knowledge gained from the system history. By now creating groups of exception rules for each expected task, the self-organizing emergent system can achieve near optimal solutions for static problem instances and good solutions for a range of expected tasks, while still being able to deal with dynamic (and unpredicted) tasks, as shown by experiments in a pickup and delivery transportation scenario.

Nick Nygren, Jörg Denzinger

A Framework for Task-Guided Virtual Machine Live Migration

Abstract

Virtual machine is an emulation environment developed for dependable computing. Live migration mechanism provides the functionality that moves an ongoing virtual machine across hosts seamlessly so as to provide non-stop services. However, service quality cannot be satisfied if excessive pages are synchronized at the stop-and-copy phase. In this paper, a task-guided framework is proposed for live migration to start at an opportune time such that a short service downtime can be guaranteed. In our framework, a code entity which updates pages within a small range or in a low frequency is tagged. Once a tag has been detected, the coordinator either approves a pending live migration request or withdraws the permission to an ongoing live migration depending on the tag types. The prototype which is capable of task-guided live migration has been implemented. Our experiments show that both the service downtime and the task execution time have been improved by our task-guided approach regardless of the possibly existing suspension overhead.

Cho-Chin Lin, Yuan-Han Kuo

Verifiable Privacy-Preserving Payment Mechanism for Smart Grids

Abstract

Smart grids have become a future trend due to the development of technology and increased energy demand and consumption. In smart grids, a user’s electricity consumption is recorded by their smart meters, and the smart meters submit the data to the operation center in each time unit for monitoring. The operation center analyzes the data it receives to estimate user’s electricity usage in the next time unit and to ensure dynamic energy distribution. Compared to traditional grids, the electricity can be flexibly controlled, and waste is decreased in smart grids. However, details of user’s daily lives may be leaked out through the frequent monitoring of user’s electricity usage, which causes the problem of privacy preserving. To solve the problem, data aggregation mechanisms are adopted in this environment. The power usage data in the same units are aggregated before being sent to the operation center. This aggregation prevents personal electricity usage data from being shared with the operation center. Thus, a user’s privacy is protected. Along with the increase in the number of research studies on smart grids, many studies on the privacy-preserving issues of power usage have been published. However, both power usage data and electricity payment data may jeopardize user’s privacy. The operation center is able to obtain user’s private information by analyzing a user’s electricity payments. Therefore, we propose a verifiable privacy-preserving payment mechanism for smart grids. In our scheme, users can submit electricity payments without revealing any private information and the operation center can verify the correctness of the payment.

Chun-I Fan, Yi-Fan Tseng, Jheng-Jia Huang, Yen-Hao Chen, Hsin-Nan Kuo

Increasing Interoperability Between Heterogeneous Smart City Applications

Abstract

Due to the increasing need for networked systems we can observe a rapid advance of IT-solutions in various sectors. However, most of the developed systems are custom-tailored solutions for specific problems and application areas, leaving us with a set of diverse frameworks. The resulting jungle of heterogeneous systems makes it difficult to find common interfaces for interconnecting the underlying businesses with each other, especially in regard to Smart City concepts. We envision a new paradigm shift towards “Smart City as a service” fueled by increased interoperability between different services with an additional emphasis on privacy-preserving data processing. This would contribute to a new level of connectivity between the environment, service providers, and people, facilitating our daily activities and enhancing the level of trust of the users. In order to achieve interoperability in the context of smart, connected cities, we propose the design of a generic, platform-independent novel architecture for interconnecting heterogeneous systems, their services, and user pools.

Alexander Rech, Markus Pistauer, Christian Steger

Reduced Transmission in Multi-server Coded Caching

Abstract

Coded caching has been widely used in computer networks for shifting some transmissions from the peak traffic time to the off-peak traffic time. Multi-server coded caching, which can share responsibility for the total amount of transmission by means of the collaboration among these servers, can be seen everywhere in our life. In this paper we consider the centralized caching system with three servers setting (two data servers and one parity check server) and propose a modified caching scheme which has performance better than the previously known schemes.

Minquan Cheng, Qiaoling Zhang, Jing Jiang, Ruizhong Wei

Distributed Sensor Fusion for Activity Detection in Smart Buildings

Abstract

Modern building systems often utilize multiple, physically separated sensors (sometimes of different type) to better detect occupancy. Such systems typically rely on a central processor where global data fusion takes place. However, such centralized architectures give rise to problems with communication and computational bottlenecks and are susceptible to total system failure should the central facility fail. There are significant advantages in distributing operations over multiple processing nodes. In the wake of this need, this paper addresses the problem of presence detection in a building by employing a decentralized sensing architecture. We focus on a network of radar sensor nodes, each with its own processing facility. Each node runs a hidden Markov model algorithm to provide a local estimate of the occupancy state and share it via wireless links. We introduce a distributed fusion algorithm that optimally combines the information generated by local nodes, having access to their private information, and recovers exactly the global estimation. System performance is evaluated in real world conditions, where sensor errors and communication may not exactly follow idealized model assumptions.

C. Papatsimpa, J. P. M. G. Linnartz

Climbing Ranking Position via Long-Distance Backlinks

Abstract

The best attachment consists in finding a good strategy that allows a node inside a network to achieve a high rank. This is an open issue due to its intrinsic computational complexity and to the giant dimension of the involved networks. The ranking of a node has an important impact both in economics and structural term e.g., a higher rank could leverage the number of contacts or the trusting of the node. This paper presents a heuristics aiming at finding a good solution whose complexity is \(N\log {N}\). The results show that better rank improvement comes by acquiring long distance in-links whilst human intuition would suggest to select neighbours. The paper discusses the algorithm and simulation on random and scale-free networks.

V. Carchiolo, M. Grassia, A. Longheu, M. Malgeri, G. Mangioni

Financial Application on an Openstack Based Private Cloud

Abstract

We build a private Cloud using off-the-shelf servers and do extensive experiments for application and performance testing. We studied a real-world application of financial option pricing by implementing two algorithms (Monte-Carlo simulation and binomial lattice for option pricing) on this private Cloud and used this application for the purpose of accuracy testing in comparison to a well established closed-form solution available as Black-Scholes-Merton formula. Also, using these algorithms we analyze performance of Cloud VMs. We compare the performance with standalone servers and found that the performance of Cloud VM is better to standalone servers when (a) the number of vCPUs are limited to a single node and (b) load balancing issues are not considered.

Deepak Bajpai, Muskan Vinayak, Ruppa K. Thulasiram, Parimala Thulasiraman

Towards Island Networks: SDN-Enabled Virtual Private Networks with Peer-to-Peer Overlay Links for Edge Computing

Abstract

While solutions to many challenges posed by IoT lie at the network’s edge, they cannot forego services available in the cloud which has over a decade of research and engineering to be leveraged. To bridge this gap, hybrid approaches in networking that account for characteristics of both edge and cloud systems are necessary. On cloud data centers, significant progress has been made on applying Software Defined Networking (SDN) to address networking challenges such as scalability, addressing, virtualization, and traffic engineering; administrators are now well-versed at managing data center SDN deployments in enterprise systems. However, the applicability of SDN in edge networks has not yet been thoroughly investigated. We propose a hybrid system that incorporates SDN software switches and overlay networks to build dynamic layer 2 virtual networks connecting hosts across the edge (and in the cloud) with links that are peer-to-peer Internet tunnels. These tunnels are terminated as subordinate devices to SDN switches and seamlessly enable the traditional SDN functionalities such that cloud and edge resources can be aggregated.

Kensworth Subratie, Renato Figueiredo

Almost-Fully Secured Fully Dynamic Group Signatures with Efficient Verifier-Local Revocation and Time-Bound Keys

Abstract

One of the prominent requirements in group signature schemes is revoking group members who are misbehaved or resigned. Among the revocation approaches Verifier-local revocation (VLR) is more convenient than others because VLR requires updating only the verifiers with revocation messages. Accordingly, at the signature verification, the verifiers check whether the signer is not in the given revocation detail list. However, the cost of the revocation check increases linearly with the size of the revocation details. Moreover, original VLR group signature schemes rely on a weaker security notion. Achieving both efficient member revocation and reliably strong security for a group signature scheme is technically a challenge. This paper suggests a fully dynamic group signature scheme that performs an efficient member revocation with VLR and which is much more secure than the original VLR schemes.

Maharage Nisansala Sevwandi Perera, Takeshi Koshiba

Path Planning for Multi-robot Systems in Intelligent Warehouse

Abstract

In this work, the path planning problem of robotics group is surveyed in the context of multi-robot transportation tasks on an intelligent warehouse. The mainly researches focused on the a shortest path theory and algorithm for single robot system. To solve the path planning problem on multi-robot systems, a novel approach is presented for multi-robot systems in an intelligent warehouse by using the method of artificial potential function (APF) in this paper. The proposed improving method of APF that motioned the strategy of wall-following with priority and how to solve the unavoidable troubles in obstacle avoidance for multi-robot systems such as local minima, non-reachable target, collision and traffic jams. Finally, several numerical simulations were provided to show the effectiveness, and the performance of the proposed method with the theoretical results.

Hailong Chen, Qiang Wang, Meng Yu, Jingjing Cao, Jingtao Sun

Dynamic Framework for Reconfiguring Computing Resources in the Inter-cloud and Its Application to Genome Analysis Workflows

Abstract

This paper proposes a framework that dynamically reconfigures an application environment by adding and removing computing resources during runtime. The main idea is that the conditions for the resources used for reconfiguration can be translated into constraints on specifications, such as the number of cores, memory size, and resource location. Our framework consists of two subsystems: an application scheduler, which determines the constraints on specifications for each application, and a resource allocator, which finds resources that satisfy the constraints established by the application scheduler. This structure enables us to apply various reconfiguration strategies by replacing the application scheduler, and also enables us to investigate new allocation strategies for the resource allocator.

As an example of the proposed framework, we developed a reconfiguration module for Galaxy, a workflow manager used in the bioinformatics field. Galaxy can act as an application scheduler by interacting with the reconfiguration module and Galaxy users can take advantage of our reconfiguration framework while using their own interface. The application scheduler applies an embedded strategy to decide when reconfiguration is invoked, whereas it can apply different reconfiguration algorithms to determine constraints on specifications by replacing algorithm modules for reconfiguration. We also describe a scheme for collecting resource metrics, such as CPU usage and memory usage, for use by the reconfiguration algorithms. Finally we conducted preliminary experiments to show the reconfiguration during runtime is necessary because the prediction of resource requirements may fail even if the algorithm uses previous execution records.

Tomoya Tanjo, Jingtao Sun, Kazushige Saga, Atsuko Takefusa, Kento Aida

Game-Theoretic Approach to Self-stabilizing Minimal Independent Dominating Sets

Abstract

An independent dominating set (IDS) is a set of vertices in a graph that ensures both independence and domination. The former property asserts that no vertices in the set are adjacent to each other while the latter demands that every vertex not in the set is adjacent to at least one vertex in the IDS. We extended two prior game designs, one for independent set and the other for dominating set, to three IDS game designs where players independently determine whether they should be in or out of the set based on their own interests. Regardless of the game play sequence, the result is a minimal IDS (i.e., no proper subset of the result is also an IDS). We turned the designs into three self-stabilizing distributed algorithms that always end up with an IDS regardless of the initial configurations. Simulation results show that all the game designs produce relatively small IDSs with reasonable convergence rate in representative network topologies.

Li-Hsing Yen, Guang-Hong Sun

Towards Social Signal Separation Based on Reconstruction Independent Component Analysis

Abstract

We all know that the ratio of social data noise is pretty significant. Therefore, tackling with noise problem is always obtained attention from data scientists. In this paper, we present a research of using reconstruction independent component analysis algorithm for blind separation social event signals from their mixtures (i.e., mixture is the combination of source signal and noise). This issue can be categorized as cocktail party problem. Despite cocktail party problem is a classical topic, however, dealing with social media data can be considered as a new research trend. From the case study with two events on Twitter, we demonstrate that our approach is quite promising. Further, our work can be applied for recommendation systems, or is used as a pre-processing step for other studies (e.g., focus search and event detection).

Hoang Long Nguyen, Khac-Hoai Nam Bui, Nayoung Jo, Jason J. Jung, David Camacho

Performance, Resilience, and Security in Moving Data from the Fog to the Cloud: The DYNAMO Transfer Framework Approach

Abstract

The data crowdsourcing paradigm applied in coastal and marine monitoring and management has been developed only recently due to the challenges of the marine environment. The pervasive internet of things technology is contributing to increase the number of connected instrumented devices available for data crowd-sourcing. A main issue in the fog/edge/cloud paradigm is that collected data need to be moved from tiny low power devices to cloud resources in order to be processed. This paper is about the DYNAMO data transfer framework enabling the data transfer feature in a internet of floating things scenario. The proposed framework is our solution to mitigate the effects of extreme and delay tolerant environments.

Raffaele Montella, Diana Di Luccio, Sokol Kosta, Giulio Giunta, Ian Foster

Development of a Support System to Resolve Network Troubles by Mobile Robots

Abstract

Universities have campus networks and different network administration policies. In some universities, network users can install network devices, which are printers, routers or Ethernet switches, in the networks of the rooms of users for themselves. All the users are not familiar with IT and cannot resolve the network troubles whose causes are the installed network devices. The users ask campus network operators of the universities to resolve the troubles instantaneously, but the operators do not know the network environment of the users or resolve the troubles. The network devices do not have management functions such SNMP because the users do not think that the functions are necessary. These network troubles are burdens of campus network operators and increase the cost of network administration. This paper proposes the support system which resolves the network troubles in the user network environments. This paper also implements and evaluates the support system.

Kohichi Ogawa, Noriaki Yoshiura

A Benchmark Model for the Creation of Compute Instance Performance Footprints

Abstract

Cloud benchmarking has become a hot topic in cloud computing research. The idea to attach performance footprints to compute resources in order to select an appropriate setup for any application is very appealing. Especially in the scientific cloud, a lot of resources can be preserved by using just the right setup instead of needlessly over-provisioned instances. In this paper, we briefly list existing efforts that have been made in this area and explain the need for a generic benchmark model to combine the results found in previous work to reduce the benchmarking effort for new resources and applications. We propose such a model which is build on our previously presented resource and application model and highlight its advantages. We show how the model can be used to store benchmarking data and how the data is linked to the application and the resources. Also, we explain how the data, in combination with an infrastructure as code tool, can be utilized to automatically create and execute any application and any micro benchmark in the cloud with low manual effort. Finally, we present some of the observations we made while benchmarking compute instances at two major cloud providers.

Markus Ullrich, Jörg Lässig, Jingtao Sun, Martin Gaedke, Kento Aida

Developing Agent-Based Smart Objects for IoT Edge Computing: Mobile Crowdsensing Use Case

Abstract

Software agents have been exploited to handle the inherent dynamicity in the Internet of Things (IoT) systems, as agents are capable of autonomous, reactive and proactive operation in response to changes in their local environment. Agents, operating at the network edge, enable leveraging cloud resources into the proximity of the user devices. However, poor interoperability with the existing IoT systems and the lack of a systematic methodology for IoT system development with the agent paradigm have hindered the utilization of software agent technologies in IoT. In this paper, we describe the development process and the system architecture of a mobile crowdsensing service, provided by an agent-based smart object that comprises agents in both edge and user devices. Mobile crowdsensing is an example of such an application that relies on large-scale participatory sensor networks, where participants have active roles in producing information about their environment with their smartphones. This scheme introduces challenges in handling dynamic opportunistic resource availability, due to mobility and unpredicted actions of the participants. We present how ACOSO-Meth (Agent-oriented Cooperative Smart Object-Methodology) guidelines the development process systematically from the analysis to the actual agent-based implementation of a crowdsensing service. The implementation is done with the ROAgent framework that utilizes resource-oriented architecture and REST principles to integrate agent-based smart objects seamlessly with the programmable Web.

Teemu Leppänen, Claudio Savaglio, Lauri Lovén, Wilma Russo, Giuseppe Di Fatta, Jukka Riekki, Giancarlo Fortino

Path Planning of Robotic Fish in Unknown Environment with Improved Reinforcement Learning Algorithm

Abstract

Path planning is the primary task for robotic fish, especially when the environment under water of robotic fish is unknown. The conventional reinforcement learning algorithms usually exhibit a poor convergence property in unknown environment. In order to find the optimal path and increase the convergence speed in the unknown environment, an improved reinforcement learning method utilizing a simulated annealing approach is proposed in robotic fish navigation. The simulated annealing policy with a novel cooling method rather than a general ɛ-greedy policy is taken for action choice. The algorithm convergence speed is improved by a novel reward function with goal-oriented strategy. Then the stopping condition of the proposed reinforcement learning algorithm is rectified as well. In this work, the robotic fish is designed and the prototype is produced by 3D printing technology. Then the proposed algorithm is examined in the 2D unpredictable environment to obtain greedy actions. Experimental results show that the proposed algorithms can generate an optimal path in unknown environment for robotic fish and increase the convergence speed as well as balance the exploration and exploitation.

Jingbo Hu, Jie Mei, Dingfang Chen, Lijie Li, Zhengshu Cheng

Review of Swarm Intelligence Algorithms for Multi-objective Flowshop Scheduling

Abstract

Swarm intelligence algorithm (SIA) is an important artificial intelligence technology, which has been widely applied in various research fields. Recently, adopting various multi-objective SIAs (MOSIAs) to solve multi-objective flow shop scheduling problem (MOFSP) has attracted wide research attention. However, there are fewer review papers on the MOFSP. Many new MOSIAs have been proposed to solve MOFSP in the last decade. Therefore, in this study, MOSIAs of MOFSP over the past decade are briefly reviewed and analyzed. Based on the existing problems and new trend of Industry 4.0, several new promising future research directions are pointed out. These research directions are: (1) new hybrid MOSIA; (2) MOSIA with high computational efficiency; (3) MOSIA based on machine learning and big data; (4) multi-objective approach; (5) many-objective flowshop scheduling.

Lijun He, Wenfeng Li, Yu Zhang, Jingjing Cao

Exploiting Long Distance Connections to Strengthen Network Robustness

Abstract

Network fault tolerance (also known as resilience or robustness) is becoming a highly relevant topic, expecially in real networks, where it is essential to know to what extent it is still working notwithstanding its failures. Different questions need attention to guarantee robustness, as how it can be effectively and efficiently (i.e. rapidly) assessed, and which factors it depends on, as network structure, network dynamics and failure mechanisms. All studies aim at finding a way to hold (or increase) resilience; in this work we propose a strategy to improve robustness for Scale-free networks by adding links between highly distant nodes in the network; results show that even adding few long-distance links leads to a significant improvement of resilience, therefore this can be assumed as an effective (and possibly with low cost) approach for increasing robustness in networks.

V. Carchiolo, M. Grassia, A. Longheu, M. Malgeri, G. Mangioni

An Online Adaptive Sampling Rate Learning Framework for Sensor-Based Human Activity Recognition

Abstract

In the field of sensor based human activity recognition, fixed sampling rate scheme is difficult to accommodate the dynamic characteristics of streaming data. It may directly leads to high energy consumption or activities detail missing problems. In this paper, an efficiency online activity recognition framework is proposed by integrating sampling rate optimization with novel class detection and recurring class detection algorithms. Based on the proposed framework, we believe that this system can effectively save battery life and computation capacity without decreasing the overall recognition performance.

Zeyi Jin, Jingjing Cao, Jingtao Sun, Wenfeng Li, Qiang Wang

A Secure Video-Based Robust and Aesthetic 2D Barcode

Abstract

The conventional barcode is currently employed in some mobile payment applications. The barcode serves as a token which represents the identity of a user. However, attacks on the payment process can be initiated by intercepting the barcode images in various ways. In this work, we proposed a ideo-based Robust and Aesthetic 2D barCode (vRA Code) so that the security of the token is protected while the efficiency in the decoding process is guaranteed. Experiments with different embedded video contents and capturing angles have been conducted to show the practicality of the proposed system. Experimental results have demonstrated that a vRA Code can be decoded in about 1.5 s.

Changsheng Chen, Fengbo Lan, Wai Ho Mow

A Migratable Container-Based Replication Management for Inter-cloud

Abstract

This paper proposes a novel approach to rapidly deploy a migratable container-based replication management for inter-cloud. The key idea behind the proposed approach is to introduce the coordination of multiple autonomic managers in inter-cloud environments. Compared with existing researches, we can not only absorb features from heterogeneous clouds through the Base Container, but also we can periodically update the deployed container-based replication mechanism through asynchronous processes, in order to improve data consistency problem. In addition, this paper analyzes the feasibility through comparing between file reading and writing in same and different clouds in the proposed approach.

Mingkang Chen, Jingtao Sun

Dilated Deep Residual Network for Post-processing in TPG Based Image Coding

Abstract

Lossy image compression algorithms like JPEG usually introduce visually annoying artifacts on decoded images, such as blocking artifacts, blurring and ringing effects. The tiny portable graphics (TPG) based image/video compression technique is proposed to improve JPEG compression performance. However, the lossy compression artifacts cannot be fully removed, especially at low coding bit-rates. Recently, some shallow convolutional neural network (CNN) models have been proposed as post-processing techniques to reduce compression artifacts. Learning from the fact that deep CNNs have shown extraordinary ability in high-level vision problems, we propose to investigate how a deeper CNN can further enhance the quality of decoded images. Specifically, we adopt a network with 16 residual blocks. In order to increase the receptive field, we change the first convolution layer in the first five residual blocks to dilated convolution with size 2. The primary experimental results show that the proposed model can outperform existing CNN based post-processing methods.

Yuan Yuan, Jingtao Sun, Miaohui Wang

Underground Intelligent Logistic System Integrated with Internet of Things

Abstract

In this paper, the urban underground intelligent logistic system is investigated to supply the demand of smart cities. The architecture, network layout, management and distribution are discussed in brief. Combined with Internet of Things (IoT) technology, smart monitoring, information integration, and intelligent decision-making will be built in the system to optimize the service efficiency and operational costs.

Qiang Yang, Guohao Li, Ting Cai, Qiang Wang

Backmatter