Quantitative analysis and visualization of memory access patterns

Cost Effective Modular Adders for RNS-based Processors

RNS can distribute the computation on long operands over small word-width RNS functional units able to operate in parallel. This property is the ground to develop fast arithmetic units. While RNS can boost addition and multiplication performance, other arithmetic operations like division, magnitude comparison, and sign detection are more...

Longest Prefix Match and Incremental Updates for Range Tries

Address lookup has become a major bottleneck in the performance of today's network routers with rapid increase in the urge for faster and high bandwidth communication. The most crucial metrics for a lookup algorithm being lookup speed, scalability and memory usage, the Range Trie algorithm provided the much needed solution to the diminishing...

Splash: Data synchronization in unmanaged, untrusted peer-to-peer networks

Peer-to-peer networks rely on gossip algorithms to spread information about the peer activity and the network status. State-of-the-art gossip algorithms are not sufficient to spread the information widely, as the size and the complexity of the unmanaged networks grow. They suffer from high bandwidth utilization and lack mechanisms to verify the...

RNS support for RSA cryptography

The majority of currently established public-key cryptosystems, e.g., RSA, ECC, requires modular multiplication in finite fields as their core operation. As a result, the throughput rate of such cryptosystem depends upon the speed of modular multiplication and upon the number of performed modular multiplications. Montgomery algorithm is one...

Exploring Tile-Based Rasterization Alternatives for Mobile Devices

Investigation on Tile-Based Rasterization on Mobile Devices. External traffic on mobile devices is a major part of power consumption. We start with a critical evaluation of previous work on Tile-Based Rasterization where software was used to reduce external traffic. The added software is now killing the performance of the microprocessor. We...

Hardware implementation of digital signal processing algorithms for long distance pipeline inspections

Inspections on transmission pipelines in the petrochemical industry are regularly conducted in order to guarantee safety of operations. Inspection devices are sent through the lines to record sensor data, this data needs to be post-processed in order to be able to make an analysis on the state of the pipeline. In this thesis, existing software...

Measurement Study of Multi-point Videoconferencing

From the 80s, with the rapid development of the Internet, videoconferencing made significant progress and played an important role in economics, education and remote healthcare. A lot of applications and models appeared. More and more companies and people get used to communicate through a video conferencing service. Therefore, it is important to...

Automated Implant-Processor Design

As we move towards an aging population, it is likely that an increasing number of people will require an increasing diversity of implants, but at a lower cost to the society. Also, as computer technology progresses, smaller, more powerful, and less battery intensive implants can be designed. However, present implant design methodology is highly...

Accelerating Computational Electromagnetic Diffraction Model on Programmable Graphics Processors

This thesis is done in the framework of wafer metrology to accelerate an electromagnetic diffraction model (EDM), which is used to measure the profile on a wafer, on programmable graphics processors (GPU). The main focus of this thesis is to investigate the suitability of EDM to GPU architectures. This study contributes with statistical data and...

System performance analysis and fixed-point architecture of a gradient-based optical flow algorithm

Optical flow algorithms present a way for computers to estimate motion from the real world. Applications like cloud motion, surveillance and robot eyesight are examples of this. The focus of existing research is mainly on either fast, but poor solutions, or slow but good solutions. In this thesis an approach to improve performance through...

Design of a Portable and Customizable Microprocessor for Rapid System Prototyping

Due to the increasing number of processors which are integrated in System On Chips (SOCs) the need for robust, highly configurable processors emerged. Preliminary research showed that commercial processors are not suited for many research projects due to the fact that they are closed source and they can not be modified. Open source processors on...

Implementing Texture Feature Extraction Algorithms on FPGA

Feature extraction is a key function in various image processing applications. A feature is an image characteristic that can capture certain visual property of the image. Texture is an important feature of many image types, which is the pattern of information or arrangement of the structure found in a picture. Texture features are used in...

A Quantitative Model for Hardware/Software Partitioning

Heterogeneous System Development needs Hardware/Software Partitioning performed early on in the development process. In order to do this early on predictions of hardware resource usage and delay are necessary. In this thesis a Quantitative Model is presented that can make early predictions to support the partitioning process. The model is based...