DigitalCommons@WPI Copyright (c) 2013 Worcester Polytechnic Institute All rights reserved. http://digitalcommons.wpi.edu Recent documents in DigitalCommons@WPI en-us Tue, 18 Jun 2013 01:39:00 PDT 3600 Culture for Radical Innovation: What can business learn from creative processes of contemporary dancers? http://digitalcommons.wpi.edu/oa/vol2/iss1/7 http://digitalcommons.wpi.edu/oa/vol2/iss1/7 Fri, 14 Jun 2013 08:25:25 PDT Organizational culture is considered by several scholars to have a significant impact on the organization's capacity for innovation. However, there is little known about the specific aspects of organizational culture that facilitates radical innovation. This article investigates in what ways contemporary dancers´ creative practice may contribute to our understanding as well as to the development of radical innovation in business. By interviewing twenty contemporary dancers and choreographers from different countries, we found five key elements that support their creative processes from idea to performance. These elements or categories are improvisation, reflection, personal involvement, diversity, and emergent supportive structures.

An interesting finding is the dancers´ approach to work and their mindset, characterized by iteration between improvisation and reflection, rather than working with pre-planned goals and structures. We argue that this approach imprints their working environment and the culture for radical innovation emerges through their way of thinking, acting and relating. This study presents a systematic framework that will provide the basis for long-term strategic artistic interventions in business in order to enable cultural transformation towards radical innovation.

]]> Nina Bozic et al. Traffic Sensitive Active Queue Management http://digitalcommons.wpi.edu/computerscience-pubs/66 http://digitalcommons.wpi.edu/computerscience-pubs/66 Wed, 12 Jun 2013 13:00:26 PDT Internet applications have varied Quality of Service (QoS) Requirements. Traditional applications such as FTP and email are throughput sensitive since their quality is primarily affected by the throughput they receive. There are delay sensitive applications such as streaming audio/video and IP telephony whose quality is more affected by the delay. The current Internet however does not provide QoS support to the applications and treats the packets from all applications as primarily throughput sensitive. Delay sensitive applications can however sacrifice throughput for delay to obtain better quality. We present a Traffic Sensitive QoS controller (TSQ) which can be used in conjunction with many exisiting Active Queue Management (AQM) techniques at the router. The applications inform the TSQ enabled router about their delay sensitivity by embedding a delay hint in the packet header. The delay hint is a measure of an application's delay sensitivity. The TSQ router on receiving packets provides a lower queuing delay to packets from delay sensitive applications based on the delay hint. It also increase s the drop probability of such applications thus decreasing their throughput and preventing any unfair advantage over throughput sensitive applications. We have also presented the quality metrics of some typical Internet applications in terms of delay and throughput. The applications are free to choose their delay hints based on the quality they receive. We evaluated TSQ in conjunction with the PI-controller AQM over the Network Simulator (NS-2). We have presented our results showing the improvement in QoS applications due to the presence of TSQ.

]]> Mark Claypool et al. Adaptive Video Streaming using Content-Aware Media Scaling http://digitalcommons.wpi.edu/computerscience-pubs/65 http://digitalcommons.wpi.edu/computerscience-pubs/65 Mon, 10 Jun 2013 11:46:14 PDT Streaming video applications on the Internet generally have very high bandwidth requirements and yet are often unresponsive to network congestion. In order to avoid congestion collapse and improve video quality, these applications need to respond to congestion in the network by deploying mechanisms to reduce their bandwidth requirements under conditions of heavy load. In reducing bandwidth, video with high motion will look better if all the frames are kept but the frames have low quality, while video with low motion will look better if some frames are dropped but the remaining frames have high quality. Unfortunately, current video applications scale to fit the available bandwidth without regard to the video content. In this paper, we present a content -aware scaling mechanism that reduces the bandwidth occupied by an application by either dropping frames (temporal scaling) or by reducing the quality of the frames transmitted (quality scaling). We have designed a streaming video client and server with the server capable of quantifying the amount of motion in an MPEG stream and scaling each scene either temporally or by quality as appropriate, maximizing the quality of each video stream. We have evaluated our setup by conducting a user study wherein the subjects rated the quality of the video clips that were first scaled temporally and then scaled by quality in order to establish the optimal mechanism for scaling a particular stream. We find that our content-aware scaling can improve perceived video quality by as much as 50%. We have also evaluated the practical impact of adaptively scaling the video stream by conducting a user study for longer video clips with varying amounts of motion and available bandwidth. We find that for such clips the improvement in perceptual quality on account of adaptive content-aware scaling is as high as 30%.

]]> Mark Claypool et al. On an anti-Ramsey problem of Burr, Erdos, Graham and T. Sós http://digitalcommons.wpi.edu/computerscience-pubs/64 http://digitalcommons.wpi.edu/computerscience-pubs/64 Mon, 10 Jun 2013 11:46:12 PDT Gábor N. Sárközy et al. Aggregate Rate Control for Efficient and Practical Congestion Management http://digitalcommons.wpi.edu/computerscience-pubs/63 http://digitalcommons.wpi.edu/computerscience-pubs/63 Mon, 10 Jun 2013 11:46:10 PDT Active queue management (AQM) promises to overcome the current limitations of end-host only congestion control by providing congestion feedback information before router queue buffers overflow. Many emerging AQM approaches use proportional integral (PI) controller design because of PI’s simplicity and effectiveness. Unfortunately, these promising AQMs still face a critical deployment challenge since there are no simple and effective PI control parameter configurations available for time-delayed systems (i.e. the Internet). As a solution, we present the Aggregate Rate Controller (ARC), a reduced parameter PI controller for Internet traffic. ARC, founded on both classical control theory and a sound understanding of Internet congestion control, uses a low frequency rate-based approach to detect congestion that minimizes control noises and provides more flexible link Quality of Service (QoS) compared with queue-based approaches. In addition, we provide practical configuration guidelines for ARC that produce efficient and resilient performance over a wide range of traffic conditions. Simulations verify that ARC effectively handles network congestion over a range of network and traffic conditions, overall outperforming other mechanisms in terms of queue dynamics, link utilization, data loss rate and object response time for Web traffic.

]]> Jae Chung et al. Inferring Queue Sizes in Access Networks by Active Measurement http://digitalcommons.wpi.edu/computerscience-pubs/62 http://digitalcommons.wpi.edu/computerscience-pubs/62 Mon, 10 Jun 2013 11:46:08 PDT Mark Claypool et al. Distributing vertices along a Hamiltonian cycle in Dirac graphs http://digitalcommons.wpi.edu/computerscience-pubs/61 http://digitalcommons.wpi.edu/computerscience-pubs/61 Mon, 10 Jun 2013 11:46:05 PDT Gábor N. Sárközy et al. The Effects of Latency on Online Madden NFL Football http://digitalcommons.wpi.edu/computerscience-pubs/60 http://digitalcommons.wpi.edu/computerscience-pubs/60 Mon, 10 Jun 2013 11:46:03 PDT With the growth in interactive network games comes increased importance in a better understanding of the effects of latency on game performance. While previous work has measured the effects of latency on first-person shooters and real-time strategy games, there has been no systematic investigation of the effects of latency on sports games. In this work, we study the effects of latency on online Madden NFL football, one of the most popular online sports games, through a series of carefully designed experiments in which we systematically control the latency between players. Our experiments illustrate the mechanisms Madden NFL uses to compensate for latency. Our user studies show there is little impact from latency on user performance in Madden NFL over typically low Internet latencies. However, for latencies higher than 500 ms, there is a significant impact on user performance, degrading performance by almost 30%. Our network measurements show periodic data rates during game-play with significant command aggregation at higher latencies.

]]> James Nichols et al. Common Tutor Object Platform - an e-Learning Software Development Strategy http://digitalcommons.wpi.edu/computerscience-pubs/59 http://digitalcommons.wpi.edu/computerscience-pubs/59 Mon, 10 Jun 2013 10:48:43 PDT The Common Tutor Object Platform (CTOP) was designed as a lightweight component framework for creating and deploying applications relating to Intelligent Tutoring Systems and e- Learning. The CTOP supports a runtime for intelligent tutoring system content deployment, a content development environment, an extensive reporting tool, and other smaller applications. The CTOP was designed with future development in mind, allowing easy specification of new base objects and extension points for future development. It has been used as the foundation of the Assistments Project, a wide scale server based ITS deployment. This paper documents the software engineering side, and has been submitted in conjunction with a second paper detailing the educational results [5]. The Assistments Project is capable of supporting a quarter of targeted students in Massachusetts, and optimistically scalable to the entire state and beyond.

]]> Goss Nuzzo-Jones et al. Measuring the Performance Gains from Directional Antennas in an Unplanned 802.11b Mesh Network http://digitalcommons.wpi.edu/computerscience-pubs/58 http://digitalcommons.wpi.edu/computerscience-pubs/58 Mon, 10 Jun 2013 10:48:41 PDT Wireless mesh networks are becoming increasingly popular. Most proposed mesh algorithms are evaluated using simulation. Simulations frequently oversimplify real world scenarios and can lead to results that are significantly different. Thus, although it is more difficult to run controlled experiments in already deployed real-world networks, it is important to understand how proposed improvements perform under these realistic scenarios. Several authors have suggested the use of directional antennas, but their merits have only comprehensively evaluated in simulators. Roofnet is an unplanned 802.11b wireless mesh network deployed by MIT and is constructed primarily of omni-directional antennas. We use measurements to evaluate performance gains from using a directional antenna on this network. Other factors such as user location and number of hops to the gateway are considered. Using throughput, latency, and streaming video tests, this paper examines the end-to-end performance of the network. Five working locations were chosen in total and three permitted directed performance comparisons between directional and omni-directional antennas. The directional antennas improved throughput by 21% reduced roundtrip times by 15% quicker round trip times. The directional antenna also provided more consistent video quality, and stabilized faster.

]]> Jason Wilson et al. On Latency and Player Actions in Online Games http://digitalcommons.wpi.edu/computerscience-pubs/57 http://digitalcommons.wpi.edu/computerscience-pubs/57 Mon, 10 Jun 2013 10:48:40 PDT Mark Claypool et al. WBest: a Bandwidth Estimation Tool for Multimedia Streaming Application over IEEE 802.11 Wireless Networks http://digitalcommons.wpi.edu/computerscience-pubs/56 http://digitalcommons.wpi.edu/computerscience-pubs/56 Mon, 10 Jun 2013 10:48:38 PDT Multimedia streaming applications can benefit from bandwidth estimation techniques to perform media scaling and buffer optimization efficiently. However, most current techniques were designed for wired networks and produce relatively inaccurate results and long convergence times on wireless networks where capacity and contention for the capacity can vary dramatically. Therefore, it is difficult to apply current bandwidth estimation tools to multimedia streaming applications in wireless network. This paper presents a newWireless Bandwidth estimation tool (WBest) designed for fast, non-intrusive, accurate estimation of available bandwidth in IEEE 802.11 networks. WBest applies a two-step algorithm: 1) a packet pair technique to estimate the effective capacity of the wireless networks; 2) a packet train technique to estimate the achievable throughput and report the inferred available bandwidth. Using an analytic model, the possible error sources are explored and WBest parameters are optimized given the tradeoffs of accuracy, intrusiveness and convergence time. The advantage of WBest is that it does not depend upon search algorithms to detect the available bandwidth but instead, statistically detects the available fraction of the effective capacity, mitigating estimation delay and the impact of random wireless channel errors. WBest is implemented and evaluated on an 802.11 wireless testbed. Comparing WBest with other popular bandwidth estimation tools shows WBest to have higher accuracy, lower intrusiveness and faster convergence times. Thus, WBest demonstrates the potential for improving the performance of applications that need bandwidth estimation, such as multimedia streaming, on wireless networks.

]]> Mingzhe Li et al. 2006 Computer Science Department MQP Review http://digitalcommons.wpi.edu/computerscience-pubs/55 http://digitalcommons.wpi.edu/computerscience-pubs/55 Mon, 10 Jun 2013 10:48:36 PDT This report presents results of a peer review of MQPs conducted within the Computer Science Department during the Summer of 2006 as part of a campus-wide MQP review. The intent of the report is to assess whether the department MQPs are accomplishing their educational goals. The report identifies problems that need to be addressed and trends that need to be continued to make the MQPs a worthwhile learning experience. It re ects data and evaluations for 44 MQPs, involving 85 computer science students, that were completed between the Summer of 2005 and the Spring of 2006. The report also makes comparisons to similar reviews done in the past. Overall, the large majority of the projects are meeting the educa- tional goals of the department. The reviews indicate that 88% of the projects were evaluated with an overall quality of at least adequate with 47% better than adequate. However, these measures are down from similar measures in the 2001 MQP Review. In contrast, the grades for this year's projects are significantly higher than the grades for the 2001 projects. This report examines these issues as well as drawing a number of conclusions about the success of the projects based upon the data collected and evaluations done for this review. The report concludes with recommendations for future reviews as the department continues to use the MQP Review as part of a larger department assessment effort.

]]> George T. Heineman et al. An Efficient Approximate Algorithm for the k-th Selection Problem http://digitalcommons.wpi.edu/computerscience-pubs/54 http://digitalcommons.wpi.edu/computerscience-pubs/54 Mon, 10 Jun 2013 10:48:34 PDT We present an efficient randomized algorithm for the approximate k-th selection problem. It works in-place and it is fast and easy to implement. The running time is linear in the length of the input. For a large input set the algorithm returns, with high probability, an element which is very close to the exact k-th element. The quality of the approximation is analyzed theoretically and experimentally.

]]> D. Cantone et al. Analysis of An Approximate Median Selection Algorithm http://digitalcommons.wpi.edu/computerscience-pubs/53 http://digitalcommons.wpi.edu/computerscience-pubs/53 Mon, 10 Jun 2013 10:48:30 PDT We present analysis of an efficient algorithm for the approximate median selection problem that has been rediscovered many times, and easy to implement. The contribution of the article is in precise characterization of the accuracy of the algorithm. We present analytical results of the performance of the algorithm, as well as experimental illustrations of its precision.

]]> Domenico Cantone et al. On the Distribution of a Saddle Point Value in a Random Matrix http://digitalcommons.wpi.edu/computerscience-pubs/52 http://digitalcommons.wpi.edu/computerscience-pubs/52 Mon, 10 Jun 2013 10:48:28 PDT Micha Hofri A Study of Video Motion and Scene Complexity http://digitalcommons.wpi.edu/computerscience-pubs/51 http://digitalcommons.wpi.edu/computerscience-pubs/51 Mon, 10 Jun 2013 10:48:25 PDT Huahui Wu Measuring the Queue Sizes of IEEE 802.11 Wireless Access Points http://digitalcommons.wpi.edu/computerscience-pubs/50 http://digitalcommons.wpi.edu/computerscience-pubs/50 Mon, 10 Jun 2013 10:48:23 PDT The increasing power, lower cost and improved convenience of wireless networks is fueling the spread of IEEE 802.11 wireless access points (APs) in the residential and commercial environments. While there has been considerable study of wireless performance, little is publicly known about the queue sizes of wireless APs. This paper presents the QFind methodology for measuring wireless AP queue sizes. QFind determines the wireless saturation point, measures the baseline delay, induces the saturation rate and measures the delay with queuing, and computes the queue size. The accuracy of QFind is validated in a wireless testbed with a Host AP, where the controlled queue size is compared to the QFind measured queue size. QFind is then used to measure the queue sizes of seven wireless APs, three commercial class APs and four residential class APs targeted for residences. The results show that wire- less AP queue sizes are packet-based and vary considerably across vendor and AP classes, with queue sizes ranging 50 packets to over 350 packets. This suggests additional engineering and science to determine the best mechanisms for providing AP queue sizes for a variety of traffic types.

]]> Feng Li et al. Knowledge Engineering for Intelligent Tutoring Systems: Using machine learning assistance to help humans tag questions to skills based upon the words in the questions. http://digitalcommons.wpi.edu/computerscience-pubs/49 http://digitalcommons.wpi.edu/computerscience-pubs/49 Mon, 10 Jun 2013 10:21:13 PDT Building a mapping between items and their related knowledge components, while difficult and time consuming, is central to the task of developing affective intelligent tutoring systems. Improving performance on this task by creating a semi-automatic skill encoding system would facilitate the development of such systems. The goal of this project is to explore techniques involved in text classification to the end of improving the time required to correctly tag items with their associated skills.

]]> Kevin Kardian et al. Towards Enabling Collaboration in Intelligent Tutoring Systems http://digitalcommons.wpi.edu/computerscience-pubs/48 http://digitalcommons.wpi.edu/computerscience-pubs/48 Mon, 10 Jun 2013 10:21:11 PDT Intelligent Tutoring Systems have historically been shown [Koedinger, Anderson, Hadley & Mark, 1997; Morgan & Ritter, 2002] to be an effective means of educating an audience. While there is great benefit from such systems they are generally very costly to build and maintain. Currently it is estimated that 200 hours of time is required to produce one hour of Intelligent Tutoring Systems content [Murray, 2002; Anderson, 1993]. For Intelligent Tutoring Systems to be widely accepted in the classroom environment there needs to be a tool set that allows for the most novice user to maintain and grow the system with minimal cost. The goal of this work is to create such a tool set targeted towards the Assistments Project [Razzaq, Feng, Nuzzo-Jones, Heffernan, & et. al., 2005] and enable teacher collaboration within the system. A goal of the Assistments Project is to provide a means for teachers to obtain meaningful data from the system that they utilize in the classroom environment thus enabling a comprehensive learning solution.

]]> Kevin Kardian et al.