PhD Candidate: Optimization of SW/HW Architectures, (Noord-Brabant)

Department
Department of Electrical Engineering

Date off
21/11/2010

Reference number
V36.1091

The Department of Electrical Engineering concerns the research and education of the Electrical Engineering discipline. Electrical Engineering covers the application of electrical phenomena with respect to energy transfer, telecommunication, and calculation and processing of information and the technology involved. Both hardware, in the form of electronic circuits and accessories, and software, in the form of system software for electro-technical application, are the subject of study. Existing and new electrical components and systems are analyzed, designed and realized. In addition, the maintenance of these systems is the subject of research, as is the relevance for society of electrical engineering and informatics.

Organization


Video Coding and Architectures (VCA) research group The Video Coding and Architectures research group (http://vca.ele.tue.nl) is one of the leading groups in the Netherlands in designing real-time video analysis and coding systems. It is involved in various EU projects such as ITEA – Information Technology for European Advancement, as well as in industrial delivering of advanced video technology.
VCA group is a part of the Signal Processing Systems department at the Eindhoven University of Technology in the Netherlands. The group carries out long-term research on image-processing algorithms and also applied research on video-processing hardware architectures. Research topics include, among others, video analysis, multiprocessor video systems, video segmentation, 3D scene reconstruction, multi-view video compression and 3D video streaming.
Job description


The Ph.D. candidate will work in the framework of the OMECA project jointly led by University of Eindhoven, Leiden University, Prodrive, CycloMedia, and other industrial partners. The project focuses on the problem of optimization (enhancement) of multimedia processing architectures with respect to multiple properties, e.g. performance, robustness, reliability and power consumption.
The research challenge of optimization comes from the fact that current systems should satisfy multiple requirements that contradict each other. For instance, a power consumption constraint and a processing speed requirement are normally contradicting each other, i.e. architecting for higher speed may increase power consumption and vice versa. Another simple example of such contradicting properties is reliability vs production cost. System reliability can be increased by duplication of SW or HW blocks, which raises the cost of materials.


The main architecture optimization challenge can be specified as follows. Based on predefined system requirements on multiple quality attributes, an architect should be able to construct a SW/HW architecture, which, firstly satisfies all the requirements and, secondly, features most balanced values of these contradictory quality attributes.

This challenge can be addressed by defining a method that caters the following steps:
(a) rapid construction of several architecture alternatives, (b) analysis of multiple quality attributes of each alternative, (c) identification of transformation rules to enhance these predefined alternatives, and (d) automated generation of new alternatives, based on the transformation rules. These steps form an iteration that can be run several cycles, until an optimal architecture alternative is found.
In this context, the goal of the Ph.D. candidate is twofold:
• investigate and develop a method that features the above-mentioned optimization steps,
• apply and validate the optimization method on a number of computer vision systems.

Requirements


University Graduate.
The ideal candidate should hold a MSc degree in computer science with emphasis on software/hardware architectures and embedded. The candidate should have an expertise in the Software Architecture domain and good programming skills, preferably in Java. The knowledge of Eclipse is a plus.


Conditions of employment

The appointment is for four years. As an employee of the university you will receive a competitive salary as well as excellent employment conditions (including excellent sport facilities and child care). The research in this project must be concluded with writing a PhD thesis. A salary is offered starting at EUR 2042 per month (gross) in the first year and increasing up to EUR 2612 per month (gross) in the last year. Moreover 8% bonus share (holiday supplement) is provided annually. Assistance for finding accommodation can be given. The university offers an attractive package of fringe benefits such as excellent technical infrastructure, child care, savings schemes, and excellent sports facilities.
TU/e offers you also the opportunity for personal development by developing your social and communication skills. We do this by offering every PhD student a series of courses that are part of the Proof program as an excellent addition to your scientific education (Proof program).
Additional Information

Additional information about the vacancy can be obtained from:

PDEng. E.Bondarev
E-mail address: e.bondarev@tue.nl
Or additional information can be obtained through one of the following links.

· About the organization: http://vca.ele.tue.nl/
· Eindhoven University of Technology: http://w3.tue.nl/en/