Radio Tomographic Imaging Data Set

Published Jan 21, 2010
Joey Wilson and Neal Patwari
SPAN Lab: Sensing and Processing Across Networks
Department of Electrical and Computer Engineering
The University of Utah

LICENSE and CREDIT

This data is free to use for academic purposes. In publications using this
data set, you must cite:

Commercial use of this data is strictly prohibited without written consent from the SPAN lab.

If you use this data set, we'd appreciate hearing how you used it. Please contact Joey Wilson and Neal Patwari.

DOWNLOAD HERE

OVERALL EXPERIMENT DESCRIPTION

28 nodes were deployed in a square perimeter of 21x21 feet. Each node was
spaced 3 feet from the neighboring nodes. The perimeter of the network
surrounded two trees with approximately 1 foot diameter trunks. See the
photo directory for pictures and a map.

Crossbow TelosB nodes were used for the network (www.xbow.com). They
transmit with the IEEE 802.15.4 protocol, in the 2.4GHz frequency band.

DATA FORMAT

Only one node transmits at a time. The first value of each line in the
data files indicates the node that is reporting its data. The 2nd column
to the 29th (since there are 28 total nodes) are the received signal
strength (RSS) values from each of the nodes to the reporting node.
For example, if the first column is 3, then the 2nd column would represent
the RSS from node 0 to node 3. Remember that the node ID's are 0-based.

Example:

0, -45, -52, -45, -85, -33, -43 ... 10 2 46
1, -74, -45, -55, -45, -56, -48 ... 10 2 46
2, -34, -52, -45, -85, -66, -23 ... 10 2 46
3, -44, -52, -55, -45, -76, -93 ... 10 2 46

It's not possible for the nodes to transmit to themselves (ie, node 0 to 0),
so if the first column is 3, the 5th column is meaningless and will always
report "-45".

The last 3 values are the time the measurement was received, in the format
of HR MIN SEC. Notice that the data collection is fast, and many RSS vectors
are captured every second.

Important note: Sometimes data from a previous capture remains in the standard
output buffer, which causes "old" data to be written to the beginning of a file.
To be safe, you may want to ignore the first 30 or so lines of data in each file.

FILE DESCRIPTIONS (whats happening in the data for each file)

Format is "filename: description"

empty.csv: The network is completely vacant.

s-*x-*y.csv: A stationary human stands at coordinate (*x,*x).
Example: s-3-3.csv: A stationary human stands at coordinate (3,3).

one-*x1-*y1-two-*x2-*y2.csv: Two stationary humans stand in the network
at locations (*x1,*y1) and (*x2,*y2).

m-1.csv: One person walks at constant velocity in 2 loops around the following
square loop path: (3,6) to (3,15) to (18,15) to (18, 6) and back to (3,6).
m-2.csv: Two people move in the same path as m-1.csv. One person starts the path
at (3,6) and the other across the diagonal at (18,15).
m-3.csv: Two people move in the same path as m-1.csv. One person starts the path
at (3,6) and the other across the diagonal at (3,15).

treehugger.csv: One person "hugs" the tree located at (8,2).
treehugger2.csv: One person "hugs" the tree located at (19,19).
runner.csv: A human runs at sprinting speed through the network.

NODE LOCATIONS

This section indicates the position of each node in the network.
The coordinates are in feet. Format is "nodeID: x_coordinate,y_coordinate"

0: 0,0
1: 0,3
2: 0,6
3: 0,9
4: 0,12
5: 0,15
6: 0,18
7: 0,21
8: 3,21
9: 6,21
10: 9,21
11: 12,21
12: 15,21
13: 18,21
14: 21,21
15: 21,18
16: 21,15
17: 21,12
18: 21,9
19: 21,6
20: 21,3
21: 21,0
22: 18,0
23: 15,0
24: 12,0
25: 9,0
26: 6,0
17: 2,0