摘要：Journal of Communications (JCM) is a top venue for high quality research that advances state-of-the-art contributions in the area of the new technologies. The focus is on Dependable Computing for Ubiquitous Services (DCUS). The latest developments in theories, systems, methods, algorithms and applications in communications have enabled new dimensions in dependable computing for ubiquitous services, such as dependable hardware and dependable software. As applications of dependable computing have permeated in every aspects of daily life, the dependability of computing has become increasingly critical.The aim of the special issue is to provide fast publication outlet for refereed, high quality original research papers in the various aspects of advances in Reliability and its applications. The special issue would focus on the research challenges and issues in the design and implementation (theories, technologies, architecture and applications) on DCUS. DCUS shall always welcome all research results on the traditional and on-going developing reliability technologies and next generation reliability technologies. Novel techniques, algorithms, architectures, and experiences regarding DCUS are invited.In this special issue, we present several papers for Dependable Computing for Ubiquitous Services.The first paper “A Sudoku-based Secret Image Sharing Scheme with Reversibility” by C.-C. Chang, P. Lin, Z. Wang, and M. Li, derives the secret shadows and generates the meaningful shadow images by adopting the sudoku. In this scheme, the sudoku grid is setting to 16×16 and divided into sixteen 4×4 blocks. Thus, authors can embed 4×(t-1) secret bits into each pixel pair of the host image. Besides, the embeddable secret capacity can be improved according to the threshold t in the (t, n)-threshold sharing system. The experiments show that the shadows can be successfully camouflaged in the host image with satisfactory quality. The distortion of the embedded host pixels is limited within range [0, 3]. Moreover, the proposed scheme provides a large capacity for embedded secret data.The second paper “On Evaluating BGP Routing Stress Attack” by Wenping Deng, Peidong Zhu, Xicheng Lu, and Bernhard Plattner, investigates a new attack on BGP routing system inspired from synchronization and resonance in complex system. The attack applies routing stress by periodically injecting and propagating excessive BGP routing advertisements, which are beyond the processing ability and the storage capacity of the BGP routers in the routing system. Authors first describe a BGP routing stress attack method inspired from synchronization and resonance of complex network. Then they devise a cascading failure model to evaluate the robustness under BGP routing stress attack on the real Internet AS-level topology. The experimental results show that BGP routing stress attack can eventually lead to a high proportion of failures and bring about serious impacts on the connectivity of the Internet routing system.The third paper “Power-Aware and Reliable Sensor Selection Based on Trust for Wireless Sensor Networks” by G. Han, L. Shu, J. Ma, J. H. Park, and J. Ni, constructs a new trust model for WSNs based on the trust mechanism in human society. Based on this trust model, authors propose a novel power-aware and reliable scheme for sensor selection (PRS). The remaining energy of a node can be considered as an important restrictive factor for the proposed scheme. Simulation results show that PRS scheme can improve the system stability, defend against the attacks of malicious nodes, and prolong the lifetime of WSNs, effectively.The forth paper “A Survey on Decentralized Flocking Schemes for a Set of Autonomous Mobile Robots” by N. Xiong, Y. Yang, J. He, J. H. Park, T. Kim, and Y. He, summarizes the existed flocking algorithms and discuss their characteristics. Then authors briefly described their fault tolerant flocking algorithms in different models. Finally they proposed the potential future research directions for dynamic flocking of a group of mobile robots. In all, this work can provide a good reference for the researchers working on dynamic cooperation of robots in distributed system. The fifth paper “Web Services Resilience Evaluation using LDS Load dependent Server Models” by M. Rak, R. Aversa, B. D. Martino, and A. Sgueglia, propose to model Web Services (WS) as Load Dependent Servers (LDS), i.e. systems whose response time to a given request depends on the load at the time the request is received. Adoption of LDS-based models enables ones to have a simple way to represent the system state. This paper is proposed to use this simple representation to quantify the system resilience comparing models built using off-line measurement and models built using on-line measurement.The sixth paper “M3S: Markovian Mobility Management Simulator for Mobility Conscious Routing in Global Networks” by P. K. Suri, and K. Taneja, proposes M3S followed by MCR (Mobility Conscious Routing) algorithm that estimates the future location of a MD according to the user’s movement history, i.e., prior movement patterns. Each MD computes the new proposed field CS (Connection Status) and estimates about the stability of each adjacent link through the transient probability matrix that overrides the default path selection mechanisms. The transient matrices reflecting the movement pattern of MDs are stored in the routing table and for each link, connection status to each immediate neighbor is kept in the form of CS (0,1) over a time interval. With this criterion, MCR can selectively send packets only through the more reliable slice of the topology. Finally the simulation results confirms that MCR algorithm based on M3S results in relative to DSR, more stable routes thereby preventing performance taxation to a reasonable extent due to frequent link breakage in MANETs.The seventh paper “A Dependable Cluster Based Topology in P2P Networks” by G. Wei, Y. Gu, Y. Ge and Y. Ling, proposes a cluster based model to solve the heavy-tail problem. By combining topology-control and priority-order, the proposed model increase the replications of rare chunks in the network and make file chunks distribute evenly. Chunk tables are designed to analyze the chunk distribution outside the cluster and let nodes in a cluster download the chunks lacked in other cluster. This paper also estimates a network size roughly and estimate the cluster numbers quantitatively. The simulation experiments show the cluster based model can make file chunks distribute more evenly and increase the download completion rate. The results of this paper are useful for the unstructured and decentralized P2P network systems to increase the completion rate, especially helpful to some un-popular file sharing swarms.The eighth paper “Channel Assignment with Topology Preservation for Multi-radio Wireless Mesh Networks” by H. Cheng, N. Xiong, G. Chen and X. Zhuang, focuses on the channel assignment problem with the original topology preservation for the multi-radio mesh networks, and aims at minimizing the overall network interference. The authors have first illustrated that the conversion from the channel assignment problem to the Max k-cut or Min k-partition problem cannot correctly evaluate the network interference because of the broadcasting characteristic of the wireless communication. Then they have formulated the problem to an Integer Line Programming (ILP) and developed both a centralized heuristic algorithm and a distributed algorithm for the channel assignment. Extensive simulations have demonstrated that our algorithms have good performance in both dense and sparse networks and thus serve as a practical solution in the multi-radio mesh networks.The ninth paper “Optimal Security Patch Management Policies Maximizing System Availability” by T. Uemura and T. Dohi, quantitatively evaluates dependability/security of a computer-based system subject to Denial of Service (DoS) attacks. More specifically, the authors develop two semi-Markov models for describing the stochastic behavior of systems with different security patch release strategies. The optimal security patch management policies are then formulated and analytically derived to maximize the steady state system availability. They further perform the sensitivity analysis of model parameters through numerical experiments and refer to the effectiveness of our preventive patch management policies.The last paper “Adaptive Checkpointing” by Z. Chen demonstrates that, good fault tolerance efficiency can be achieved by adaptively choosing where to store the checkpoint data at run time according to the specific characteristics of the platform. The authors analyze the performance of different checkpointing schemes and propose an efficient adaptive checkpointing scheme to incorporate fault tolerance into high performance computing applications.
