The greatest accomplishment seems unfinished, yet its applications are endless. The greatest fullness seems empty, yet its applications are never exhausted. --- Tao Teh Ching, Lao Tzu
| Models | Architecture | Utilities | Home | Download |
Introduction:
Rmase has been developed for dealing with the challenge of comparing different routing algorithms for sensor networks. More than ten routing algorithms have been integrated into the Rmase repository. Those algorithms were developed by different groups in the sensor network community, and some were originally written in TinyOS/NesC and were used in real applications. A presentation on application scenarios and performance comparisons can be found here.
Algorithm Repository:
Following is a (partial) list of routing algorithms that has been integrated into the Rmase repository.
|
Name |
Institute |
Characteristics |
Layers |
|
Gradient Routing (Grad) |
Vanderbilt University (VU) |
flooding-based |
aggregate_queue |
|
Mint Routing (Mint) |
UC. Berkeley (UCB) |
tree-based |
neighborhood |
| Grid Routing (Grid) | Ohio State University (OSU) |
geo-tree-based asymmetric link periodic maintenance top down |
neighborhood check_duplicate grid_routing |
| Broker Routing (Broker) | University of Notre Dame (ND) |
tree-based symmetric link periodic adaptive maintenance top down |
neighborhood check_duplicate nd_hood nd_broker |
| Backbone Routing (Backbone) | University of Virginia (UVA) |
tree-based symmetric link no maintenance top down |
neighborhood check_duplicate SD backbone |
| Real-time search (LRTA) | Palo Alto Research Center (PARC) |
search-based asymmetric link no maintenance learning |
neighborhood confirm_transmit check_duplicate mcbr_search init_hello (or init_backward) |
| Constrained Flooding (CF) | PARC |
flooding-based delayed transmission learning |
neighborhood delay_transmit mcbr_flood init_hello (or init_backward) |
| Adaptive Tree (AT) | PARC |
tree-based asymmetric link no maintenance learning top down |
neighborhood confirm_transmit check_duplicate mcbr_tree init_backward |
Application Scenarios:
Four routing scenarios have been modeled after real applications.
|
Scenarios |
Defined in | Model Parameters (non-default) |
Characteristics |
|
Intruder Detection (LIS) |
testlis/lis_model.m |
size: (nx, ny)=(10,10),
offset: (ox, oy)=(0.1,0.1), hole: (x, y, rx, ry, q)=(6.5,4.5,2,9,0), source: user defined events, destination: (static, (0,0), 1, 1, true, (0,0,0)) |
convergecast |
|
Tracking (RFT) |
testrft/rft_model.m |
size: (nx, ny)=(5,10),
offset: (ox, oy)=(0.1,0.1), source: (dynamic, (4,9), 1, 1, true, (-0.2,-0.2,0.1)), destination: (static, (0,0), 1, 1, true, (0,0,0)), source rate: 1p/s |
dynamic to static |
| Pursuer Evader Game (PEG) | testpeg/peg_model.m |
offset: (ox, oy)=(0.1,0.1),
source: (dynamic, (3,3), 1, 1, true, (-0.2,-0.2,0.1)), destination: (mobile, (6,6), 1, 1, true, (-0.2,-0.2,0.1)), source rate: 1p/s |
dynamic to mobile |
| Sound Source Localization (SL) | testsl/sl_model.m |
topology: user defined topology, source: user defined events, destination: (static, (39.42,1.9), 10, 1, true, (0,0,0)), strength = 20 |
convergecast |
Performance Comparisons (click the picture icon to see the full size picture):
| Experiment | LIS | RFT | PEG | SL |
| Topology |
 |
 |
 |
 |
| Event Trace |
 |
 |
||
| Latency |
 |
 |
 |
 |
| Throughput |
 |
 |
 |
 |
| Success rate |
 |
 |
 |
 |
| Efficiency |
 |
 |
 |
 |