Signal processing algorithms and architectures can use dynamic reconfiguration to exploit variations in signal statistics with the objectives of improved performance and reduced power. Parameters provide a simple and formal way to characterize incremental changes to a computation and its computing mechanism. This paper develops a framework for dynamic parameterization and applies it to MPEG-4 motion estimation.A novel motion estimation architecture facilitates the dynamic variation of parameters to achieve power-compression tradeoffs. Our work shows that parameter variation in motion estimation helps achieve power reduction by an order of magnitude, trading off higher ii

For motion estimation (ME) and discrete cosine transform (DCT)

No abstract

No abstract

No abstract

No abstract

No abstract

We have all seen the formation of tribes and cliques on the latest edition of the television hit Survivor. Has Survivor mentality invaded your senior projects as well? For the last ten years or so, engineering programs nationwide have, with varying degrees of success, tried to incorporate the ABET-required outcome of "ability to function on multidisciplinary teams." 1 While recognizing that all engineers must function on such teams in the real world, implementing them in the academic context poses substantial challenges. Nonetheless, I would argue that it is either extremely difficult or impossible to evaluate how well we have accomplished this outcome without putting students into the situation where they must actually work with members of other disciplines to accomplish a significant design. Even after such teams have been formed, the vagaries of academia, and the real world, interfere with our ability to measure student function, or dysfunction, within these teams. This paper presents a short history of efforts at the United States Air Force Academy, focusing on the projects sponsored by the Department of Electrical and Computer Engineering and the assessment techniques used to evaluate our students" ability to function on multidisciplinary teams. One method that we and others have used to assess the performance of students on teams is a peer rating scheme. There are a variety of approaches. 2-10 One component is usually a weighting system used to enforce reasonable grades and prevent the inevitable effect of team members all wanting to get good "grades" in the component of team performance. This is usually accomplished by limiting the average grade given to some value or renorming the student submissions to establish the desired average and variance. 2, 3 This can be effective, but if the variance is changed significantly, it may hide problems in team members working effectively together. A particularly difficult problem to deal with is the formation of subgroups within the project team. In our experience this most commonly occurs with engineers from various disciplines forming a clique or tribe and rating the members of their tribe highly while penalizing those from other disciplines. When this occurs, the team cohesiveness and focus on the goal of completing the design successfully can be compromised, as team members try to insure that the majors in their own discipline get better grades than those in other disciplines. We have named this phenomenon the "Survivor" effect. We will discuss our approach to defeating this divisive aspect of assessing the performance of students on multidisciplinary teams. Overview In the early 1990s, many engineering programs accomplished the ABET requirement for design experience using a senior design project involving a single student. Students were typically enrolled in a single "Design Project" course and completion of the design and course constituted success in the required element. As ABET attention to industrial concerns increased, the importance of working as teams of ...