Multi Sensor Data Fusion Methods Using Sensor Data Compression and Estimated Weights

When data fusion is performed in a distributed environment, it improves accuracy but the constraints are limited communication bandwidth and limited processing capability at the fusion center. So, it is crucial to compress the data at the fusion center. This is accomplished by reducing the dimension of the data. Based on the Linear Estimation and weighted least square fusion results, a method is presented for compressing data at each local sensor to improve the accuracy of the fused estimates. Another method for multi sensor data fusion with estimated weights is also suggested in this paper which also improves the accuracy of the fused estimates.