From Ubiquitous Sensing to Cloud Computing: Technologies and Applications

With the advances of sensing technologies (e.g., smart devices, wireless sensor networks, and Internet of Things) and cloud computing [1, 2], there have been a large number of applications [3–5] not only exploiting the sensor network to collect data but also using the cloud computing to discover desired information. Ubiquitous sensing is a hot but complex topic in a large number of research areas in the last years. It can collect a great volume of data from various applications to detect certain events. In addition, cloud computing is usually used to handle such “big data” from sensors and to discover feasible information and knowledge. Therefore, integration between ubiquitous sensing and cloud computing is a significant research issue. Many research projects have been investigating algorithms, technologies, and platforms to make various ubiquitous applications more innovative.

In this SI, we received numerous outstanding article submissions. We then sent these submissions to qualified experts for review. Finally, based on the review results and the suggestions of reviewers, nine articles were accepted to be included into the special issue. The articles are simply introduced as follows.

The work from R.-K. Sheu et al. entitled “Design and implementation of file deduplication framework on HDFS” proposed a simple and feasible application framework of data deduplication for developers based on the most popular Hadoop system. They designed and implemented RFD-HDFS and FD-HDFS framework and evaluated them. The experiments results show that the duplicated data space can be saved and the upload performance is not affected by the integrated schemes significantly.

In the paper entitled “Implementation and evaluation of large-scale video surveillance system based on P2P architecture and cloud computing”, authors proposed an architecture for video surveillance service by integrating P2P and Hadoop-like file system technology. The core concept of the proposed architecture is exploiting the concept of Hadoop file system to replicate streaming video to other front end devices for improving the reliability of large-scale video surveillance system. In addition, authors implemented a large-scale VSS (LVSS) and proposed an evaluation framework based on the proposed architecture.

The authors of the paper entitled “Innovative IaaS management system for sensor devices and IT resources,” would like to propose an innovative IaaS management system, named Sensor Cloud (SC), for sensor devices in which the sensor devices are managed as one of the IT resources in the same way as CPUs, storage, and networks in the cloud computing. The SC allows multiple users to share the sensor devices instead of the sensor data. In addition, they implemented SC and deployed it on actual server environments hosting the existing cloud software. The evaluation results of SC show that the transmission delay for sensor data between the sensor devices and the users is efficient and effective.

The paper entitled “A data gathering method based on a mobile sink for minimizing the data loss in wireless sensor networks” from J. Yoo et al. proposed a novel data gathering method based on a mobile sink considering the data loss in wireless sensor networks. The scheme actively sends the stored data to the mobile sink by considering the amount of collected data in the cluster header and the mobile patterns of the sink node. Authors also conducted some experiments and result showed that the proposed scheme reduced the data loss by sending data toward the moving path of a mobile sink.

The paper entitled “Dynamic deduplication decision in a Hadoop distributed file system,” from C.-S. Liao et al. proposed a dynamic deduplication decision algorithm to improve the storage utilization of a data center which uses HDFS as its file system. The proposed system aimed at small enterprises and organizations to effectively utilize storage space by removing duplicates without investing any additional funding in infrastructure. The paper also conducted many experiments to evaluate system performance. The result showed that the proposed approach can significantly improve the utilization of storage space.

The work from P. Sthapit and J.-Y. Pyun entitled “Handover strategies in beacon-enabled mobile sensor network,” would like to propose three algorithms in order to achieve an efficient handover in IEEE 802.15.4 network environment. The first one is proactive reassociation decision scheme (PRD) which can anticipate if the node is going to lose connectivity by analyzing the LQI history before it really happens. The second one is called greedy channel scan (GCS) that can decrease both time and energy required for association. The third is to predict the node's direction of movement while LQI of beacon is received and to select the coordinator who has maximum available bandwidth. The analytical and simulation results demonstrated that the schemes are highly efficient in terms of both energy and time.

The paper entitled “A Voronoi-based sensor handover protocol for target tracking in distributed visual sensor networks,” from Dr. T.-W. Sung and C.-S. Yang used the concept of Voronoi diagram for target tracking in visual sensor networks. The work proposed a Voronoi-based distributed sensor handover protocol for visual sensor networks. The simulation result shows the benefits of the proposed approach in terms of target-detected latency, target-tracked ratio, and average target distance.

The paper entitled “Ontology-based, process-oriented, and society-independent agent system for cloud computing,” introduces an ontology-based, process-oriented, and society-independent agent system which allows all the organizations to utilize it by defining and importing their societies’ ontology and some process patterns, which can be instantiated from the process’ ontology, into the system. The suggested system was evaluated for the quality and the applicability perspective. The experimental results showed that the proposed system is applicable in all societies.

The work from L. Tang et al. entitled “Methodology and tools for pervasive application development,” presented a development environment, called PerDE, which supports a novel design method. The approach combined the notion of the situation with an application model and provided a domain-specific design language and a set of graphical toolkits covering the development life cycle of a pervasive application.