Abstract:
|
Locating an animal helps researchers to track migration, monitor resource utilization, and predict animal behavior. For animals that are too small to use a GPS collar on, triangulation (measuring multiple angles) is the most common method to locate an animal in the field. An alternative is trilateration, measuring multiple distances using the received signal strength (RSS). The use of angle in addition to distance provides valuable information but is rarely discussed in models for RSS data. This study constructed RSS models with both distance and angle effect for a better resolution. The engineering literature suggested multiple models for the relationship between RSS and angle. A truncated normal distribution was assumed because the receiver only recorded signals above a minimum. Locations were estimated by maximum likelihood and least sum of squares. We illustrate this approach using simulation data and data from a study of Monarch butterflies. This approach combined with state-space movement model will be used to connect multiple locations and trace animal movement.
|