Hongzhou Shi

PCSM: A Context Sharing Model in Peer-to-Peer Ubiquitous Computing Environment

This paper proposes the peer-to-peer context sharing model (PCSM) which is a ubiquitous computing oriented peer-to-peer context sharing model. Owing to the distribution and limit of resources of the mobile ubiquitous network, the PCSM model constructs a context management framework based on the mechanism of registration-query. Through designing the broadcast messages for available terminals discovery and unicast messages for remote registration, the model implements remote registration of Context Database Agents, and builds up an architecture for sharing and fusing of contexts in a peer-to-peer ubiquitous computing environment to satisfy the requirements of sentient computing applications. In contrast with some current context management systems, the PCSM model partly solves the inherent problem of high cost for communication and storage redundancy through the mechanism of optimization. Besides, the PCSM model enables computing nodes to share contexts, on which the decisions depend. Based on the PCSM model, a prototype of instant messenger with ubiquitous computing ability is provided in the end.

A Service Discovery Framework for Ubiquitous Computing

This paper presents a novel service discovery framework for ubiquitous computing called Ubiquitous Computing Service Discovery Framework (UCSDF). UCSDF offers a more flexible and scalable architecture which can combine the local services with remote services. That characteristic makes UCSDF different from many other service discovery frameworks and more adaptive to ubiquitous computing environments.

A Fuzzy Petri Net Model towards Context-Awareness Based Personalized Recommendation

In a personalized recommender system of a ubiquitous computing environment, the decision on recommendation depends on some uncertain factors. Fuzzy system has an ability of solving the reasoning uncertainty, and gets widely used in the context awareness based personalized recommender system. In this paper, we present a fuzzy reasoning model based on the Fuzzy Petri Net. The model considers the requirements of task fulfillment, users preferences and QoS of the system. With difference from those existing recommender systems, we focus on the formalized modeling of fuzzy reasoning of the personalized recommendation. Through formalization of fuzzy reasoning, we propose an approach to construct a rule-based reasoning system. The feasibility of the proposed approach is validated by an example of the personalized recommender engine.

Design of Virtual Personal Computing Environment for Ubiquitous Computing

This paper presents a ubiquitous computing technology called Virtual Personal Computing Environment (VPCE), which enables users to access their personalized computing environments anywhere using any available terminals. The key feature of the VPCE is that it can be constructed dynamically by reassembling application services and resources distributed over networks, which is different from the traditional desktop personal computing environment that is tied to an individual computer to execute. In VPCE, users work with a heterogeneous collection of application services, and data access by application as well as the application accessed by users are distributed over networks, while the applications and data distribution is transparent to the users.

Ubiquitous Computing-Oriented Distributed Fuzzy Reasoning Petri Net Modeling and Simulation

In a ubiquitous computing environment, contexts are initially got and stored on those nodes scattered over the environment. However, the traditional reasoning about contexts applied a centralized approach which aggravated the load of reasoning server and cost for communication of context. As an important approach for context reasoning, the rule-based reasoning can be easily decomposed of independent propositions. Therefore, the rule-based context reasoning can be decomposed and distributed over those nodes of the ubiquitous environment to decrease the computing load of reasoning server. In this paper, we propose a distributed fuzzy reasoning Petri net model (dFRPN) towards the decomposition of distributed fuzzy reasoning. dFRPN model is able to formalize the distribution of fuzzy reasoning over every node of the whole environment. dFRPN model is characterized as a hierarchical structure to make a description of reasoning on nodes more clear. Considering the limited capabilities of some mobile nodes such as PDA and smart phone, we add the special migrating transition to define the load detection and corresponding actions. At the end of the paper, the feasibility of dFRPN model is validated through a case of context-awareness based personalized recommendation system.

A Hierarchical Service Discovery Framework for Ubiquitous Computing

This paper presents a novel service discovery framework for ubiquitous computing called hierarchical ubiquitous computing service discovery framework (HUCSDF). HUCSDF offers a more flexible and scalable architecture which can combine the local services with remote services. Based on the novel architecture, HUCSDF possesses some useful characteristics such as supporting migration of users personal computing environment, group collaboration, interoperation with other kind of services and integration of context-aware information. These characteristics make HUCSDF different from many other service discovery frameworks and more adaptive to ubiquitous computing environments.

A Policy-based Resource Monitoring System in Distributed Computing Environmen

The next generation resource monitoring system in distributed computing environments should not only accurately monitor the environment but also automatically reconfigure the environment. In this paper, we present INAPmon, a robust and scalable policy-based resource monitoring system in a multi- server network computer environment. It monitors the resource usage with sensors and aggregates the data in servers for reconfiguring the environment, which aims to guarantee load balance between servers and impartiality between users. This policy-based approach indeed provides a novel perspective for the design of future resource monitoring system.

DFre: A Distributed Fuzzy Reasoning Engine for Personalization Recommendation

In a pervasive computing environment, the personalized recommender system incorporates contexts into recommendation and becomes a multiple dimensional decision expert system. In this paper, we present DFre, a distributed fuzzy reasoning engine for personalization recommendation. With difference from those existing rule-based systems, the DFre puts an emphasis on the distribution of the recommendation in the pervasive computing environment. With the distributed data management framework, the DFre has an ability of getting those distributed primitive data and fuzzy rules, which makes the distributed reasoning become true. Finally, the feasibility of DFre in the user-centric recommender system is validated through the Flowing Desktop which is an example of the personalized recommender engine.

Context-Awareness Based Multidimensional Personalized Recommendation

This paper proposes the CaPeR which is the context-awareness based architecture for personalized recommendation. The CaPeR provides the personalized recommender engine and the peer-to-peer context management framework. With a hybrid approach, the personalized recommender engine combines those contexts into the decisions on recommendations to get more comprehensive recommendation effectiveness. Owing to the distribution and limit of resources of the mobile terminals, the CaPeR constructs a context management framework based on the mechanism of Registration query. Through designing an efficient mechanism of message intercommunications, the CaPeR builds up the architecture for sharing and fusing of contexts to satisfy the requirements of recommendations. At the end of this paper, the Flowing Desktop is given a description as the prototype of the CaPeR.