Exploration of Beamforming Approach to Enhance the Detection Rate of Underwater Targets in Distributed Multiple Sensor Systems.

We address the problem of distributed detection in multiple sensor systems in the context of underwater target detection. A number of sensor nodes are employed to detect the presence or absence of the signal. The objective is to introduce a method for improving the detection performance with the help of conventional beamforming technique. The sensors are assumed to be independent and observe a common phenomenon. The local data from all the sensors will be collected by a fusion center (FC), which makes a final global decision. The problem of distributed detection is considered with the assumption that the sensor quantized data are independent in the distributed system. The performance matrices considered in our work are probability of detection and false alarm probabilities. We apply decisions rules at the single sensor and beamformer stage and conduct performance analysis of detection using Receiver Operating Characteristic (ROC) curves. The local decisions from sensors will be provided to a fusion center, where fusion rules are used to combine the decisions and provide a global inference. We show that the detection performance increases after each stage of distributed detection, with the aid of beamforming. We have mathematically modeled the scenario and derived the equations for related parameters and verified it with simulation results. [ABSTRACT FROM AUTHOR]