SkiScape Sensing Shane B. Eisenman

1
SkiScape Sensing Shane B. Eisenman and Andrew
T. Campbell Electrical Engineering, Columbia
University Computer Science, Dartmouth College
SkiScape Simulation
Physical Components Description
Motivation
Physical drawing of the Dartmouth Skiway is
translated to a reachability graph, where trails,
lodges and lifts compose the set of vertices

• Skier Node
• Mote-class devices embedded in ski equipment
  (e.g., ski boot heels)
• Powered with rechargeable batteries
• Sensors temperature, accelerometer, microphone

• Pole Node
• Line powered, mounted on trail-side light poles
• Gateway between Skiers and backhaul network
• Sensors webcam, microphone, radar

Compared to traditional applications for sensor
networks (environmental and industrial
monitoring), recreational sports is a domain that
will spark more general interest in wireless
sensor network technology by bringing people into
the loop. The SkiScape is part of our broader
goal to push People-Centric Sensing.

• Lift Node
• Line-powered, mounted on lift structures
• Gateway between Skiers and backhaul network
• Sensors temperature, accelerometer

• Lodge Node
• Line powered, mounted inside lodge structures
• Gateway between Skiers and backhaul network
• Sensors temperature, photogate, webcam

Attributes for each of the vertices are assigned.
Trail dimensions, difficulty, pole location.
Lifts speed, length. Lodge dimensions,
locations of Lodge Nodes. Attributes for each
skier are assigned skill level.
Opportunistic Sensing and Collection
Data Usage and Presentation
Sparse deployment of mobile and static sensors
give a more complete picture over time of the
trail conditions. The mobility of Skier Nodes
reduces the cost of data transport to the
ultimate data sinks.
Provide sensed data feedback to skiers about
trail conditions.
Set tx power for Node radios, set radio duty
cycles, beacon rates.
Time-stamped data allows for automatic speed
policing of skiers (AutoPatrol feature).
Skier nodes physically carry locally sensed data
and may also mule the data of other Skiers for
robustness or timeliness.
Location-stamped data allows for FriendFinder
feature.
Initial Results Skier Node / Pole Node
Connectivity
Data transfer between Skiers and gateway Nodes
depends on radio duty cycle and beacon rate.
Data archives allow for skiers to analyze
long-term traces on their skiing habits.

• - 300 Skiers
• - 8 hour simulated time
• Random placement of Pole
• Nodes along trails edge at
• average 300 ft spacing.

Nominal values suggest 10 hours between battery
recharge is possible, if operating all Skier node
components at full duty cycle.
Safety/emergency workers track skiers speed and
location in case of accident
Management monitors skier flow statistics to
estimate wear and enact preventative trail
maintenance (snow making)
With support from the Institute for Security
Technology Studies (ISTS) and Intel Corp. More
information on the MetroSense Project, including
publications, technical reports, and source code
from http//metrosense.cs.dartmouth.edu/.