SENIOR DESIGN - Potential Projects

Project #

Project

Description

Outside Stakeholder

Supervising Faculty

1
Play-by-play software development with Jazz

It is great people - working together - who develop great software. Effective collaboration support is nowadays the cutting edge in software development environments, and shown by tools like Jazz(see http://jazz.net ), which builds upon the experience of the Eclipse Individual Development Environment (IDE) to push the envelope, and deliver comprehensive, integrated and natural support for team work and collaboration.

The goal of this project is to enhance the open Jazz platform with facilities to collect, examine and analyze the activities of software developers within a team, as well as innovative mechanisms to orchestrate their team work in an emergent, bottom up fashion. The team will leverage the Jazz repository that keeps track of individual and collaborative development activities to design and implement mechanisms for transforming that data into bits of actionable knowledge that capture the play-by-play operation of the team. That knowledge is going to be pushed back in real time onto the environment, and used to improve the developers' awareness of their peers and their work, as well as enhance in a friendly and unobtrusive way their ability to coordinate and collaborate.

None

Peppo Valetto
valetto@cs.drexel.edu

2
Virtual Classroom

Develop a virtual classroom that can be used to teach online classes. Video/audio feeds of students and instructor, a presentation layer, and an area to take notes are parts of the key interface. The instructor controls the "room," but can temporarily give control to the students.

This project has three students already Chris, Cannon, Chris Gore, Michael Sullivan, and Kyle Gorlick. We are looking for a few more.

None Jeff Salvage
jsalvage@cs.drexel.edu
3
Intelligent Tactical Mesh Networks Supporting Situation Awareness

Situation awareness tools facilitate a first responder’s decision-making process by assisting in the perception, comprehension and prediction of an environment. These tools require a wireless network so that first responders can communicate and share information. Unfortunately, in many environments, such as a train tunnel or a inside a building, access to wireless networks is frequently not available.

The goal of this senior design project is to design and develop a tactical network to support situation awareness capabilities for first responders. Such tactical networks can be rapidly deployed in environments with no pre-existing network infrastructure.

Students will work with technologists at Drakontas (www.drakontas.com) in this project to help develop real world use cases and requirements, as well as design, implement and test a proof-of-concept. Students will be exposed to and get hands on experience with various cutting edge technologies such as: 802.11, mobile ad-hoc network routing protocols, instant messaging and presence protocols, embedded Linux-based operating systems, Java Micro Edition, etc.

 

Alan Kaplan
kaplan@drakontas.com

Bill Regli
regli@drexel.edu
4
Precision-Guided Robot Platform

Robotic land explorers (e.g., lawn mowers, metal detectors) typically work either by following a buried guide wire or by randomly traversing a small space, e.g. 3,000 square feet maximum. It is impractical to run that much guide wire in a large space, especially considering the presence of utility lines, tree roots, and other obstacles along the path. Likewise, a “bouncing wanderer” protocol is prone to missing large portions of the yard and potentially getting stuck on various obstacles. A prototype robotic metal detection system has been designed and partially implemented, making use of a high-precision GPS. Continue last year's project by improving the hardware, software, user interface, usability, safety and traversal strategies.

Frank Jacquette, Jacquette Consulting

Jeff Popyack
jpopyack@cs.drexel.edu

5
Electronic Version of Philadelphia's Deadman's Game
Deadman was a game that was literally played in the streets of Philadelphia in the 40’s 50’s and 60’s that was a combination of hopstoch and a board game. The object of the game is to go from number to number in sequence. The fun is that six players are trying to do the same thing while no getting stuck i the deadman box. The strategy is how to move your piece around to obtain victory. Team play enhances the competition.

Minister David Williams
wllmsdve1@msn.com

None

6
4D Archaeological Data Exploration

In 2000 construction began on what would become the National Constitution Center, the first museum dedicated to the study of the US Constitution. Before construction began the National Park Service in conjunction with the National Constitution Center undertook what would become known as the richest colonial era archaeological site excavated in an urban area. Working with the National Park Service (NPS) on projects like ``3-D Colonial Philadelphia'' in which Drexel students have reconstructed rich 3D models of the block at the time, Drexel has helped make this history accessible to students and researchers alike.

Past work has focused on building historically inspired visual models of the buildings at the time. Since then millions of artifacts have been cataloged, soil samples have been detailed, and biographies of the area's residents have been pieced together. The data, although well cataloged, is unwieldy and uninviting. We intend to modernize the presentation of this data, and create an exciting visual exploration environment that enables archaeologists and real/virtual visitors to intuitively study our city's history.

By building on Drexel's past work, and the work done by NPS archaeologists we will create a rich, four dimensional environment where museum visitors and NPS researchers alike can explore the area's historical data: three-dimensional spatial/geographical data augmented with temporal/historical information. With the detailed geo-tagged and time-stamped data, we will empower researchers to ask questions like: How did pottery fabrication develop from 1750 to 1790? and; Does the number of artifacts discarded have any correlation to the socioeconomic profile of the inhabitants?

Archaeologist Jed Levin (NPS)

Ko Nishino
kon@drexel.edu

Glen Muschio
glen.muschio@gmail.com

7 Visual Language for Processor Design and Testing
Visual Language for Processor Design and Testing
This project builds on a previous project which designed a visual
language that compiles to VHDL (a hardware description language).  The
goal of the previous project was to design and implement a visual
language that would be capable of using a block diagram, like the ones
in the Patterson and Hennesy (PH) text used in Systems Architecture, to
design and implement a processor like the one in PH.  See the V2HDL
project at

https://www.cs.drexel.edu/SeniorDesign/2008Material/2008Groups.html

for details of the previous project.  While successful - the V2HDL
project made came in third place in the senior design competition -
there is more work to be done to get the project to the state where it
can be incorporated into Systems Architecture and related classes.  In
particular, there needs to be a mechanism for testing and verifying the
produced hardware and a direct link so that the hardware can be
simulated in the inside the visual environment rather than having to
write a separate test bench in VHDL and simulate using VHDL wavetools
like modelsim.  Ideally, the VHDL code produced by the tool would be
capable of producing working FPGA (Field Programmable Gate Array) code
and hence a complete hardware implementation.
This project is well suited to a multidisciplinary project involving CS
and CE students.
  Jeremy Johnson
jjohnson@cs.drexel.edu
8 Experimental Material Sample Archival Tool A significant problem in the field of materials science is the inability of researchers to meaningfully compare and share data.  Typically, a material is characterized by techniques such as scanning electron microscopy and tomography that capture the internal structure of a material as either an image or a volume array.  The  Mechanics of Microstructures Group (MMG) has developed techniques to quantify and compare these images via higher order statistical correlations. 

The goal of this project is to: 1) design a database and retrieval system for images, volumes and their higher order correlations; 2) create a multi-user, collaborative user interface that uses the retrieval-system as either a desktop application or a web site, 3) establish an extensible framework for comparing the statistics of the images in a lower dimensional feature space and retrieve similar images based on their representation in the space.

The candidate team does not need any knowledge on materials science, characterization techniques, or higher order statistical representations. Ideally the candidate team should have an interest in image processing, DSP, numerical analysis and database design.  
While there are no constrains on the technologies used, large amounts of existing code and libraries are written in C#.  Some familiarity with or willingness to learn Matlab will be helpful on this project. 

For more information about the MMG: http://mmg.materials.drexel.edu

Dr. Kalidindi, Department of Materials Science and Engineering
skalidin@coe.drexel.edu

 

 
9 Automated system for Nanoindentation maps Nanoindentation, where a diamond indenter is pressed on to a softer
material, is a versatile tool for measuring the mechanical properties from
small material volumes. At the Mechanics of Microstructures Group (MMG) in
the Department of Materials Science and Engineering at Drexel University we
have recently developed novel data analysis procedures that can capture a
wealth of information about the mechanical properties of the sample from a
single indentation experiment. However in order to generate maps of
indentation (maps containing 1000 indentation points or more) and also to
commercialize our analysis techniques, there is an urgent need for
automating the analysis system.

The goal of the senior design team in this project will be to develop a
fully automated system that will gather the raw data on a square grid on a
sample, and analyze it as per the recently developed and validated models in
our research group, and produce maps of various indentation properties.
There isn't a commercial system available in the market at this time for
this purpose. Given its importance to materials community, this product will
have significant commercial value if it is properly developed and
demonstrated to work.

For more information about the MMG: http://mmg.materials.drexel.edu
Dr. Kalidindi, Department of Materials Science and Engineering
skalidin@coe.drexel.edu
 
10 2nd Generation CRM Platform Development

Ciright Software has developed a comprehensive end-to-end CRM (Customer Relationship Management), ERP (Enterprise Resource Planning), ERM (Enterprise Resource Management) computing application/platform for use in managing corporate intelligence data used to drive additional sales. The application was developed using Microsoft Visual Basic/ Microsoft Access front-end with a Microsoft SQL Server back-end. The application is currently being used as the internal data management and operations platform for a construction capital goods sales/service company. Ciright would like to port the application from the desktop environment to an Application Service Provider model (central client/server application) so that the application runs on the .NET architecture of ported to a JAVA front-end with a MY-SQL back-end. The application is rather involved as it has been developed over the past 15 years or so.

Company is a local Philadelphia startup.

George J. Awad
george@iamg.biz

 
11 Knowledge Mining of Maintenance Data: Systems Integration, Data Management, & Discovery This project could involve prognostics and artificial intelligence discovery to mine a complex, interconnected pool of maintenance data. The project could begin at various stages in the full system life cycle: definition of the requirements, identification of the systems, definition of the database, tuning of the database, identification of the prognostic and diagnostic queries, application of knowledge discovery to answer various asset lifecycle questions.

Bonnie Harvey
Raytheon Technical Services Company LLC
bonnie_l_harvey@raytheon.com

 
12 Business Process Re-Engineering / Enterprise Architecture, Visualization, Analysis, Taxonomy / Ontology This could encompass creating a model of business topics to support their further visualization and analysis. Efforts could include defining the model structure, data, taxonomy, ontology, and how the data is inter- and intra-connected. Once the model was created, ways to surface deeply embedded concepts and relationships would come into play. Bonnie Harvey
Raytheon Technical Services Company LLC
bonnie_l_harvey@raytheon.com
 
13 Tools to Improve Awareness about the State of the Operator
A major challenge for battle command systems for tactical edge
networks is how to connect the situation awareness and state
information from the soldier/operator with the underlying controls for
the network.  For example, users who are under duress may need
improved situation awareness feeds, hence requiring changes to
underlying QoS on the network to information more rapidly to where it
is needed.  Currently, there are few ideas as to how to best bridge
the "cognitive state" of the operator with so-called "cognitive
networks".  This project will assess tools and software for extracting
cognitive state from the human user.
  Bill Regli
regli@drexel.edu
14 Tools to Support Small Team Operations Warfighters, first responders and other tactical teams (drug
enforcement, SWAT, bomb squad etc) usually use operational concepts
based on small groups (3-to-12 people).  Improving the performance and situation awareness of these small groups is an area of extensive
research, however there are few tools specifically designed for such
focused teams---most collaboration and situation awareness tools are
very generic (i.e. VoIP or a shared whiteboard).  Under this project,
Senior Design Teams will identify a specific team context (i.e. SWAT
team
or a dismounted platoon), perform a task analysis of the needs of
the team, and design/implement information sharing, collaboration and
awareness prototypes to help the team perform these specific tasks.  A
specific focus will be on activities that require the group to work
together or monitor their group performance.  The senior design teams
will also implement tools for assessing and monitoring the performance
of the group using the tools.  This task is particularly suitable to
being assigned to multiple senior design teams, each of which can work
on a different specific context or scenario.  In addition, where
appropriate (i.e. if the team is a team consisting entirely of US
Citizens), the ACIN Program can provide the team with doctrine and
materials specific to a number of missions and scenarios for
dismounted warfighters.
  Bill Regli
regli@drexel.edu
15 WiMAX Software defined radio implementation Software defined radios implement all transmitter and receiver operations in a flexible, software reconfigurable architecture. They will be a critical component of next generation communications systems. The IEEE 802.16 WiMAX standard exists to provide broadband wireless data for a variety of applications. A senior design team will use a flexible software defined radio architecture in the Drexel Wireless Systems Laboratory (wireless.ece.drexel.edu) to implement the WiMAX standard and create a platform through which future enhancements to the standard can be prototyped and analyzed. Dr. Dandekar, dandekar@ece.drexel.edu  
16 Controlling virtual machines in large-scale service environments The team will design, implement and experiment with provisions
for the monitoring and analysis of the behavior of virtual machines, and
the applications that run on them. Virtualization is a hot technology
that enables to instantiate multiple separate and heterogeneous
operating systems and computing environments on top of the same
hardware. Virtual machines (VMs), such as those supported by VMWare, are
widely used, for example, in large-scale data centers, to simplify the
provisioning and management of services with disparate resource and
quality requirements.

In that context, it becomes critical to keep under control the
performance and the health of running VMs: a malfunctioning VM may
compromise the service(s) it is currently hosting, and possibly also
other "adjacent" VMs running on the same physical machine, for example
by making wasteful use of hardware resources.

Currently, it is hard to differentiate between faults and other
incidents that occur at the application level, or the VM level or on the
underlying physical host. Therefore, we want to design a monitoring and
analysis suite for determining at any given time the utility of a
running VM, under a number of different experimental circumstances, and
compare it with the utility of its hosted applications. The development
work and the experiments will be carried out on a cluster of blade
servers in the ECE department.


Desired Deliverable: A monitoring and analysis suite (software and
models) for diagnosing VM failures.

Nagarajan Kandasamy
kandasamy@coe.drexel.edu

Peppo Valetto
valetto@cs.drexel.edu