Project Assignments
Team |
Member(s) |
Project Name |
1 |
Taranjeet Singh Bhatia
|
NvrBAlone |
2 |
Philip Shibly
|
SensEvent |
3 |
Vignesh Saravanaperumal
|
The Urban Security and Privacy challenges |
4 |
Lauren Ball
|
FitTogether |
5 |
Michael Betancourt
|
WiFind |
6 |
Brandon Ochs and Albert Park
|
PhotoSharing |
7 |
Sara Gaffar
|
Social distance relating to urban space |
8 |
Travis Cossairt
|
SportTown |
9 |
|
|
10 |
|
|
11 |
|
|
12 |
|
|
Research project submission requirements:
The project report should be written in the form of a conference paper,
you're required to use the ``latex'' (version available for Windows and Linux).
See the definition of Latex
from Wikipedia. You can download the MikTEX.
The
The Not So Short Introduction to LATEX is a good manual that I would
recommend. Please use Bibtex for the
references. Please use the templates from IEEE
publications site.
The project report should include:
- Introduction, motivation, goals, expectations
- Related work, other similar projects, with citations. Evaluation of
conferences, workshops, journals where similar work is presented
- Description of the project, implementation choices, architecture and design.
Challenges and problems encountered
- Experimental results, preferably comparison with other systems
- Conclusions and future work
If you have any questions, please post it to the class emailing address which
would be useful for the other students since there are some students in the
class who have experience with latex.
NOTE If you would like to propose your own project (individual or as a
team), please provide 1 page summary of the proposed project. The project
proposals must be turned in to the instructor by the beginning of the third week
of classes.
Class Projects
Project 1: Pace of life
We use expressions such as "hectic", "slow" etc. to describe the pace of life in
different cities, but is it possible to measure the "pace of life" in a
quantitative manner? Can we perhaps use the average speed at which people walk
in a city to measure the pace of life in a city?
Given that a lot of people carry mobile phones, we can use these mobile phones
to help us to measure the walking speed of people. If many people from many
cities contribute data, we can get an idea about the pace of life from many
cities. We can perhaps use the data to verify whether it is true that the pace
of life in Tokyo is faster than that in Oslo.
A starting point of the project is to look into how to determine whether someone
is walking and his/her walking speed, and the get the data contributed to a web
site. It may be interesting to think about other metrics too, say do people in
one city talk faster than that in another city? How do you measure the speed of
talking? etc.
Project 2: Social distance relating to urban space
Houses, enterprises, universities, temples and churches, governmental
institutions, etc. represent entities that can be understood in different levels
of social proximity to the population. Find a way, how to assign places to
social distances using social interactions and community structures. Related
question: What is the average Euclidean distance between places with the same
social distance in one urban area?
Project 3: Private Space Detection
In dense and crowded cities, private space where people can be alone to relax
are rare. Use mobile sensors to find patterns of spatial-temporal private
spaces, i.e. spaces that at a certain time more likely to be less crowded.
Project 4: Density Probing for Population Estimation
The population of a city changes not only over several years with birthrate and
death rate, but also over several hours. For instance in Luxembourg, the
population is about 500k but increases dramatically over the day when people
come from Germany or Belgium to work there. Similar dynamics can be observed on
Holiday places. Use mobile sensors to locally detect the population density and
try to estimate the dynamics in the population changes by interfering local
data.
Project 5: NovelSee
Participatory sensing often relies on receiving a great number of varied inputs
to provide a useful service. For this project, you are to create a proof of
concept, specifically designed to find novel images in the world. Users will
snap and upload pictures from their mobile devices to the central server. The
server will use an image search system (e.g. Lire:
http://www.semanticmetadata.net/lire/) to determine which images received over a
certain period of time are the most novel (that is, least like the existing
photos in the database). It should then post a photo of the day. This system
will help encourage a stream of interesting and novel shots.
Project 6: FitTogether
Working out is often a boring and isolating experience. People work out on rows
of machines, on treadmills and the like, typically passing time in solitary
pursuits like listening to music or watching TV. Your job is to make a
participatory fitness application. Users at a gym would download this app to
their device, and start it when they start their work out. They could enter
information into it about their current work out (their number of reps, their
personal bests, etc, optionally from a sensor like the one available from Nike).
The application would encourage competition, cooperation and cohesion. Users
should be able to compete in small fitness competitions (e.g. most reps on the
machine during the day). Users who like to lift weights may ask for spotters.
These are just a few suggestions, but the app can include anything that you
think could make the workout more fun and engaging for gym-goers.
Project 7: SensEvent
Special events, including weddings, bar mitzvahs and quinces have long been hot
beds of participatory sensing. People typically bring cameras and share their
pictures with fellow guests. It is popular to give the guests disposable cameras
with which to take pictures. Those who put on the event later mine those photos
for ones that are valuable. Your job is to create an application that supports
participatory sensing of a special event. The event organizers will first be
able to send out invites to the attendees, which they will receive with the
application. At the appointed time, the application will become active on the
mobile device. Users can then take pictures, record video and audio, which will
be timestamped and geocoded. Finally, at the end of the event, the organizers
will be able to login to a website which will assemble all of this content by
various criteria (by user, in chronological order, by area, etc).
Project 8: Disaster Model
Participatory and urban sensing have huge potential applications for the area of
crisis informatics. With platforms like Ushahidi (http://www.ushahidi.com) users
with mobile devices can share information that, if properly aggregated, can make
coping with a disaster easier. However, in order to create worthwhile
applications in this domain, and to be confident that they will do more good
than harm, is is imperative to have a model of how events unfold during a
disaster. You job is to create such a disaster model for a well known disaster
(e.g. the September 11th attacks, Hurricane Katrina, the recent earthquake in
Haiti, etc). Using the best available records and research, assemble a simple
multiagent model of the events and the response of the victims. Your model
should include a simple grid of the area, where grids can be inhabitable,
uninhabitable, dangerous, etc. The model should replay events that changed these
areas over time, and show how agents responded. Then, you should model a
participatory sensing application for these users and gather various simulation
statistics about the application. Models like this will be invaluable to
assessing the efficacy of crisis sensing applications before they are deployed.
Project 9: GoingMyWay
Ride sharing has become a popular way of reducing transportation costs as well
as environmental impact. There are now several web sites that are dedicated to
matching up people to car pool. You job is to make these kinds of ride sharing
match ups even more convenient and useful, as well as extend them to other types
of activities. Build an application that tracks the routes of a user throughout
the day. This information should be uploaded (securely and encrypted) to a
central web server. Then, the system should look at these routes, and find other
users with similar routes. Some work will need to be done to create or adapt a
suitable algorithm for this kind of clustering. The application should then
connect users together who share common routes, whether those routes are during
their daily commute, or part of their recreation time.
Project 10: PhotoSharing
A photo sharing application lets you read the exif data for an image to extract
the gps data and display where the image was taken on the map. The goal is to
create a social networking photo application that friends can use to share
pictures with each other on the iPhone. It will automatically show dots on the
map where each image was sent from, and when they select that image it will show
an image of the user who sent it (an avatar or some kind of icon) as well as an
area for them to chat about the photo.
Project 11: WiFind
Sometimes it is difficult to find open wireless networks to connect to when you
are in a new location. Even worst is when there is something critical that you
must take care of and your own home network connection is down. My proposal is
to make a peer supported map of wireless networks in the area. Data collected
would include the GPS location of where a reading was taken as well as the
following information about the possible connections:
- Service Set Identification (SSID)
- Security (Open, Closed (If possible, the kind of encryption))
- Signal strength
- Current cell phone signal strength
One key function to the system is to allow for storing of wireless network
information locally until a proper network connection can be found. This will
provide the database with a bigger data set and can speed up data collection
over a large area. Since, we are also recording the cell phone signal strength,
a peer based cell phone reception map could also be made. Other applications to
the system include raising awareness of how to properly configure routers.
Public service announcements could be directed in areas where people do not have
security and make sure they are aware of the risks of doing so. Users of the
system could register their own access points to the site as a welcomed network
connection to those in need. They could also set if their internet connection is
down making others in their neighborhood aware of possible outages. This
information could be forwarded to Internet Service Providers (ISPs) for instant
reporting of problems within their networks.
Project 12: Sport Town
Users activate this mobile iPhone app to mark the physical activity/sport they
are doing each day and uses GPS data to record the location and time and type of
activity. Interface will be simplified to popular categories: walk/run, bike,
skate, swim, team sports, gym, etc. and then plotted color-coded on a central
map of the city for others to view. Other user's can use the app to filter out
to see what other parts of the city are the most physically active, and to find
out what kind of activities are available around the area. This would be useful
for other people looking to find better/new places to run/bike/skate/swim and be
around others of the same physical hobby & sport.
Project 13: Survey: The Urban Security and Privacy Challenges
Is there truly a way to be protected? The phone is automatically on because of
911 services. The cell phone company knows where you are traveling- instead of
technological issue, this may be a political or legal issue. People will know
where you are. If it's not you, someone will pick up the bluetooth signal coming
from your phone. Passive monitoring of your presence will be everywhere. Is
there going to be a privacy? Similarly, is our personal data sensed and stored
somewhere in the cloud really secure?
This project should survey the security and privacy issues within urban sensing
research. The survey should provide summaries of the most widely used techniques
and/or applications and evaluate them based on common metrics.
Project 14: Survey: Technical and Engineering Challenges in Urban Sensing
Continuous sensing application access hardware sensors on the phone, they use
signal processing techniques to extract information from raw data and use
classification algorithms or machine learning algorithms software running on the
phone or in the cloud. This pipeline of raw data, classification and
communication with back-end presents a number of different problems.
This project should survey the technical and engineering challenges within
urban sensing research. The survey should provide summaries of the most widely
used techniques and/or applications and evaluate them based on common metrics.
Project 15: Survey: Healthcare Applications in Urban Sensing
This project should survey the healthcare applications within urban sensing
research. The survey should provide summaries of at least 10 applications (from
academia and commercial) and evaluate them based on common metrics.
Project 16: Survey: Environmental Applications in Urban Sensing
This project should survey the environmental applications within urban sensing
research. The survey should provide summaries of at least 10 applications (from
academia and commercial) and evaluate them based on common metrics.