Thesis (Ph.D.)--Lehigh University, 2008. / Adviser: Mooi-Choo Chuah.

Based on a dynamic model that simulates the behavior of TCP flows and active queue management (AQM) scheme two specific congestion avoidance modes, feedback control law and real time control law, have been presented respectively in the thesis. The control laws proposed monitor the status of buffers and multiplexor of the router and detect incipient congestion by sending warning signals to the sources. In our work, the AQM scheme can be modeled as a feedback control system governed by stochastic differential equations driven by a doubly stochastic point process with intensities being the controls. Besides, the AQM scheme can also be modeled as a real time control system in which the state of the router is monitored online and an optimal control must be executed in real time for each of sampled time intervals. Key Words: Feedback Control, Real time Control, RED, Dynamic Model, TCP, AQM.

Trust plays a crucial role in online commerce, collaboration and interactions in general. Reputation systems have emerged in recent years as a technology and tools for fostering trust in online interactions. These community tools help people make decisions on whom they can trust, encourage trustworthy behavior, and preclude dishonest and unskilled participants. Despite their successful use, today's reputation systems face many problems, one of which is the lack of portability. Since every existing reputation system is bound to a specific website, a user's reputation is locked in one community only, making the user unable to use that reputation outside the community. Portable reputation systems will solve this problem by making a user's reputation information accessible to any online entities s/he wishes to interact with. This thesis introduces the concepts of trust, online trust and reputation, introduces reputation systems, analyzes the importance of trust and reputation systems in online interactions, conducts a study on portability of online reputation, and proposes a model for reputation system which supports the feature of portability.

Abstract not available.

Online reputation systems are currently receiving increased attention while online interactions are flourishing. However, they lack one important feature: globality. Users are allowed to build a reputation within one online community, and sometimes several reputations within several independent online communities, but each reputation is only valid within the corresponding community. Moreover, such reputation is usually aggregated by the provider of the online reputation system, giving the querying agent no say in the process. This thesis presents a novel solution to this problem. We conduct a literature review on existing trust and reputation models and classify these models using proper criteria. We introduce an online reputation system that collects reputation information about a ratee from several online communities and allows for this information to be aggregated according to the inquiring agent's own requirements. We propose a configurable aggregation method for local and global reputation based on a discrete statistical model, taking into account several factors and parameters that qualify the reputation. We also implement a prototype of the proposed reputation computation model.

Variable-step variable-order 3-stage Hermite-Birkhoff (HB) methods HB( p)3 of order p = 5 to 15 are constructed for solving nonstiff differential equations. Forcing a Taylor expansion of the numerical solution to agree with an expansion of the true solution leads to multistep and Runge-Kutta type order conditions which are reorganized into linear confluent Vandermonde-type systems of HB type. Fast algorithms are developed for solving these systems in O(p2) operations to obtain HB interpolation polynomials in terms of generalized Lagrange basis functions. The order and stepsize of these methods are controlled by four local error estimators. These methods, when programmed in Matlab, are superior to Matlab's ode113 in solving several problems often used to test higher order ODE solvers on the basis of the number of steps, CPU time, and maximum global error. On the other hand, HB(5-15)3 are programmed in object-oriented C++ and the Dormand-Prince 13-stage nested Runge-Kutta pair DP(8,7)13M are programmed in C. DP(8,7) is found to use less CPU time, have smaller maximum global error but require a larger number of function evaluations than HB(5-15)3. However, for expensive equations, such as the Cubicwave, HB(5-15)3 is superior. In the C++ program, array and matrix are considered to be new objects. Algorithms, testing programs and new objects are structured separately as header files.

Session Initiation Protocol (SIP), an easy and simple Internet application layer protocol used to establish a session, and passive Internet Protocol (IP) devices like SIP-based surveillance cameras, form a perspective combination that is on demand in the network market. The only necessity of the predefined network is to acquire a secure mechanism due to the private information that the devices transmit on the Internet network. This thesis proposes an architecture for securing information that passive IP devices deliver into public IP networks. The architecture provides a security mechanism for authentication, authorization, and audit (AAA). The mechanism combines the security features of the authentication server with SIP architecture to provide AAA service to registered users requesting access to passive IP devices. Also, the Authentication-Authorization Database (AADB) that allows device certificate management is introduced. The dynamic password and dynamic session key utilized by the Public Key Infrastructure (PKI) scheme are also introduced to enhance the security features of devices by authenticating and administering user accesses and device accesses. The AAA with the authentication database and dynamic authentication mechanism ensures a secure IP network based on SIP protocol.

Free Software offers people the freedom to share or change the software. By accessing the source code, people can adapt or modify the software to meet their special needs. This facilitates research, education and business. However, adding special purpose functions to Free Software is still a big challenge for developers. Though some software packages provide functions that have potential utility for certain needs, these functions might not be directly used or are very difficult to use, especially by students. This thesis aimed to provide developers with some feasible approaches to assisting novice users. It attempts to add forecasting functionality and easy-to-use user interface to different types of software packages, such as R and Scilab. Several example interfaces are implemented in multiple programming languages (such as Perl, JavaScript and HTML) on the Windows and Linux platforms. We conducted analyses on the attempts made and the lessons learned in these implementations. This provides prototypes and caveats for further efforts, which may have relevance to effectively exploiting Free Software, especially for online teaching.

This research investigated the use of web-centric collaborative techniques and methods for the software industry new product development (NPD) process. After carefully reviewing previous literature, a questionnaire was designed targeting individuals who currently work on or previously participated in the software new product development (NPD) process for the past six months. The survey was launched through the third-party web researching company, InsightExpress, where 200 completed surveys were collected. As the observation from the results of this research indicates, the web-based level of collaborative tools does increase the level of integration between different departments in the NPD process. However, the impact of web-based collaborative tools on either overall performance or performance elements (technical success, financial success, budgetary, time-to market or schedule) is not evident. Moreover, the level of adapting web-based collaborative tools in the software industry is still not as popular as it was anticipated to be.

There are many software visualization tools available today to help software engineers to explore software systems. However, when a system is huge, some of these tools do not satisfy the exploration requirements. The big problem is that the techniques the tools use do not provide an effective display and access mechanism to handle huge information spaces within the limitations imposed by available screen space. To alleviate the problem, this thesis describes methods that help users to explore huge software systems. In particular, we apply dynamic browsing incorporating such details as an extra result box mechanism, plus pattern based searching to help users to handle large query results. Then the thesis introduces the algorithms we apply to generate the layouts. We propose the radial angle model to visualize the internal structures of rooted trees. Also we apply the spring model to visualize the external structures among rooted trees. Next, the thesis describes various animation methods we use to smooth the transitions, track the focus of exploration, clarify unexpected results, and illustrate complex operations. In addition, we modify traditional camera animation, and propose an animation timing scheme 'slow-in fast-out' to exaggerate the reality. Next, the thesis describes a series of experiments we conducted to assess the effectiveness of the browsing, layout algorithm and animation techniques we implemented. Finally the thesis describes how we use the analysis of the experiment results to guide our future research.