Sensor Selection With Communication Constraints for Linear Measurement Model

This paper proposes a framework for group greedy methods to solve the sensor selection problem with communication constraints based on linear measurement models. Due to the limited communication range of the sensors, if we assume an edge between two sensors which are in the communication range of each other, we would like to be able to connect any two of the selected sensors. As a result, all the data in the selected sensors can be collected to facilitate the data fusion. Thanks to the successive property of the group greedy algorithms, the communication constraints can be accommodated naturally. The simulation results demonstrate the superior performance of the proposed algorithms.