This paper proposed a scalable chip multiprocessor architecture, which is called Hyper-scalar combined with the concept of superscalar and multithreaded architecture; hence, this architecture can enhance single-threaded performance by using core group and also supports multithreaded applications. System programmer can dynamically allocate the core groups to accelerate a single thread by extended system instructions. In order to solve the data dependence between all issued instructions the virtual shared register file is proposed. This mechanism allows the data in local register files to be accessed from other cores through the data switching path hardware and the instructions are executed only when the operands are available. The instructions within a single-threaded application can be dispatched to variable cores without re-compilation. This execution paradigm accelerates the single-threaded performance more flexibly. In the case of simulation and experimental framework, the ESL Model written in SystemC, a modeling language based on C++ is to provide hardware-oriented simulation platform and the MediaBench suite is selected for the experiments. On average, the Hyper-scalar architecture can accelerate single-threaded performance by 30% to 110% using 2 ~ 8 cores.

It has been proofed that Reconfigurable Computing System possesses the potential to promote system efficiency. Fine-Grain Reconfigurable Computing System, which integrates the co-design of software and hardware, is the prevailing current in system designing with low clock rate and high efficiency. The thesis focuses on computing-oriented Fine-Grain Reconfigurable Computing System of software-hardware interwork, and setting up invoking working model for software program and hardware module as well. The designing of software and hardware can be integrated into one program, which deals with the hardware as a function. By quick computing of hardware, it can promote system efficiency. This interwork mechanism can be combined into traditional instruction execute pipeline. It composes manageable Fine-Grain for Reconfigurable Computing System as a new processor architecture, which brings up new command to support execution of software-hardware interwork mechanism. The model architecture is verified by algorithms of multimedia application, such as Motion Estimation and DCT. In addition, it presents the optimizing model by analysis and comparison of software and hardware efficiency.

Many things can be communicated through dialogue, including information, thoughts, feelings, attitudes, beliefs and personal experiences. More recently, dialogues have been used in focus group research and in program evaluations. Despite the increasing prevalence of dialogue in research and evaluation, much is still unknown about dialogue, especially how dialogue emerges and occurs within a group setting. The aim of the current study was to describe and identify the various factors involved in a dialogue, examine the relationships among these factors, and conceptualize the process of dialogue within a multi-stakeholder participatory evaluation. A qualitative analysis of three focus groups, each comprised of eight to ten participants, yielded several findings. First, several factors were found to help facilitate the interactions between multiple stakeholders in dialogue, including the development of common ground and specific contributions made by participants. Secondly, communication within these multiple stakeholder groups was found to alternate between two individuals, a dyadic exchange, or between multiple participants, a complex exchange. Thirdly, the moderator and participants were found to take on each other roles. Finally, from these conversations, a model was developed to illustrate the progression of a dialogue in these groups. These results have many implications for program evaluators, focus group leaders, and other practitioners in the field.

This thesis explains and tests a wavelet based implicit numerical method for the solving of partial differential equations. Intended for problems with localized small-scale interactions, the method exploits the form of the wavelet decomposition to divide the implicit system created by the time discretization into multiple, smaller, systems that can be solved sequentially. Included are tests of this method on linear and non-linear problems, with both its results and the time required to calculate them compared to basic models. It was found that the method requires less computational effort than the high resolution control results. Furthermore, the method showed convergence towards high resolution control results.

There is a growing demand for faster, improved data and voice services in rural areas without modern telecom infrastructure. A wireless network is often the only feasible solution for providing network access in this environment, due to the sparse populations and difficult natural conditions. A system solution that incorporates the Multipoint Communication System (MCS) algorithm created by TRLabs into the available IEEE 802.11b Wireless Local Area Network (WLAN) devices was proposed and studied in this thesis. It combines the advantages of both systems, that is, the MCS capability of integrating Voice over Internet Protocol (VoIP) and data services and the IEEE 802.11b standard, currently the most widely used in WLAN products. A system test bed was set up inside Network Simulator-2 (NS-2). The data and VoIP performance was tested. Modifications to the original MCS algorithm to improve system performance were made throughout this thesis. In a constant rate radio channel, data performance (throughput and transmission efficiency) was measured using the original MCS algorithm, which was comparable to the standard Distribution Coordination Function (DCF) operation of IEEE 802.11b when both were simulated at similar conditions. On an 802.11b platform, the Automatic Rate Fallback (ARF) feature was incorporated into the original MCS algorithm. However, when clients with different data rates were present in the same channel, all the clients involved received unacceptably low and equal data throughput, dragged down by the low rate clients. A modified MCS data polling algorithm was proposed with the capability of repeated polling, which eliminated the negative effect of low rate clients in a multi-rate channel. In addition, the original MCS algorithm was modified to be more efficient in the voice polling process. The voice performance and data throughput were tested at various conditions. However, the one-by-one polling still resulted in very low voice transmission efficiency. The time wasted became more severe with increasing relay distance and channel rate (only 8.5% in an 11 Mbps channel at 30 km). A new voice handling process similar to Time Division Multiple Access (TDMA) mode was implemented and simulated. Its voice efficiency can be kept at 25% at any setting of relay distance and channel rate. Data transmission in the same channel can also benefit from using the new voice scheme. The normalized saturation throughput could be improved by 13.5% if there were 40 voice clients involved in an 11 Mbps channel at the relay distance of 15 km, compared to the original MCS algorithm. More improvement in voice efficiency, voice capacity, and data throughput can be achieved at longer relay distance, or with more voice calls set up.

There is a growing demand for faster, improved data and voice services in rural areas without modern telecom infrastructure. A wireless network is often the only feasible solution for providing network access in this environment, due to the sparse populations and difficult natural conditions. A system solution that incorporates the Multipoint Communication System (MCS) algorithm created by TRLabs into the available IEEE 802.11b Wireless Local Area Network (WLAN) devices was proposed and studied in this thesis. It combines the advantages of both systems, that is, the MCS capability of integrating Voice over Internet Protocol (VoIP) and data services and the IEEE 802.11b standard, currently the most widely used in WLAN products. A system test bed was set up inside Network Simulator-2 (NS-2). The data and VoIP performance was tested. Modifications to the original MCS algorithm to improve system performance were made throughout this thesis. In a constant rate radio channel, data performance (throughput and transmission efficiency) was measured using the original MCS algorithm, which was comparable to the standard Distribution Coordination Function (DCF) operation of IEEE 802.11b when both were simulated at similar conditions. On an 802.11b platform, the Automatic Rate Fallback (ARF) feature was incorporated into the original MCS algorithm. However, when clients with different data rates were present in the same channel, all the clients involved received unacceptably low and equal data throughput, dragged down by the low rate clients. A modified MCS data polling algorithm was proposed with the capability of repeated polling, which eliminated the negative effect of low rate clients in a multi-rate channel. In addition, the original MCS algorithm was modified to be more efficient in the voice polling process. The voice performance and data throughput were tested at various conditions. However, the one-by-one polling still resulted in very low voice transmission efficiency. The time wasted became more severe with increasing relay distance and channel rate (only 8.5% in an 11 Mbps channel at 30 km). A new voice handling process similar to Time Division Multiple Access (TDMA) mode was implemented and simulated. Its voice efficiency can be kept at 25% at any setting of relay distance and channel rate. Data transmission in the same channel can also benefit from using the new voice scheme. The normalized saturation throughput could be improved by 13.5% if there were 40 voice clients involved in an 11 Mbps channel at the relay distance of 15 km, compared to the original MCS algorithm. More improvement in voice efficiency, voice capacity, and data throughput can be achieved at longer relay distance, or with more voice calls set up.

Many things can be communicated through dialogue, including information, thoughts, feelings, attitudes, beliefs and personal experiences. More recently, dialogues have been used in focus group research and in program evaluations. Despite the increasing prevalence of dialogue in research and evaluation, much is still unknown about dialogue, especially how dialogue emerges and occurs within a group setting. The aim of the current study was to describe and identify the various factors involved in a dialogue, examine the relationships among these factors, and conceptualize the process of dialogue within a multi-stakeholder participatory evaluation. A qualitative analysis of three focus groups, each comprised of eight to ten participants, yielded several findings. First, several factors were found to help facilitate the interactions between multiple stakeholders in dialogue, including the development of common ground and specific contributions made by participants. Secondly, communication within these multiple stakeholder groups was found to alternate between two individuals, a dyadic exchange, or between multiple participants, a complex exchange. Thirdly, the moderator and participants were found to take on each other roles. Finally, from these conversations, a model was developed to illustrate the progression of a dialogue in these groups. These results have many implications for program evaluators, focus group leaders, and other practitioners in the field.

NVIDIA developed a technique of executing general program on GPU, named CUDA (Compute Unified Device Architecture), in 2006. The CUDA programming model allows a group of same instructions to execute on multi-thread simultaneously, which has advantage of parallel programs in reducing the execution time significantly. Although CUDA provides a series of C-like APIs (Application Programming Interface) so that programmers can easy use CUDA language, it still costs certain efforts to be familiar with the development. In this thesis, we propose a tool to automatically translate C programs into corresponding CUDA programs which reduce program development time effectively.

In this thesis, a multi-rate LDPC (Low-Density Parity-Check code) decoder circuit is proposed for IEEE 802.16e standard. In the proposed circuit, we modify the overlapping structure for different code rate of the LDPC decoder to enhance the hardware utilization ratio and provide flexible parametric design. LDPC decoding is completed by the recursive operations between variable nodes and check nodes. We use Beneš network to implement the wire-routing of the operations between variable nodes and check nodes. However, the decoders with different code rates may result in different Beneš networks and increase the hardware complexity. We propose a modified overlapping structure to reduce the complexity of parallelized Beneš network and to increase the hardware utilization ratio.

The GPU significantly improves the computing power by increasing the number of cores in recent years. The design principle of GPU focuses on the parallism of data processing. Therefore, there is some limitation of GPU application for the better computing power. For example, the processing of highly dependent data could not be well-paralleled. Consequently, it could not take the advantage of the computing power improved by GPU. Most of researches in GPU have discussed the improvement of computing power. Therefore, we try to study the cost effectiveness by the comparison between GPU and Multi-Core CPU. By well-applying the different hardware architectures of GPU and Multi-Core CPU, we implement some typical algorithms, respectively, and show the experimental result. Furthermore, the analysis of cost effectiveness, including time and money spending, is also well discussed in this paper.