#Toy Simulator 2014
Authors: Rodny Perez Emily Dolson Josh Nahum
This projects aims to develop a simulator for wireless sensors.
We will something else later
##Todo
- Implementing features:
- Sensor dependancy (Done)
- Time occilation (The mean changes over time) (Done)
- Days or seasons (comming soon)
- Sensors that varies in mean and standard deviation
- Sensor pattern
- Partition sensors by groups (Done)
- Multiple sensor type (Done)
- Temperature drifts globally
- Adding correlations
- Change temperature over time (moran process)
- Change station to correlated sensor group (Done)
- Change mean in station to a periodically changin value such as sine wave.
##Errors and anomalies
- Abrupt changes (added)
- Large deviation (added)
- Unusually low variance (change in result from predicted value)
- Deviation from neighbors (variance in result from neigbors) (added)
- Drift towards a particular value (0 or something else) (change over time in unforeseen ways) (added)
- increasingly mean difference
##Ways of representing anomalies
- Change in mean value among multiple sensors. Either within a group or an entire field
- Adjusting the correlation between sensor modalities