Joint source and sensor location estimation

Expressions are presented for the minimum attainable bounds on the mean squared error for estimators of the range and bearing to a source and a parameter vector which specifies the receiving array sensor positions. It is assumed that noise corrupted measurements for estimating tne sensor positions are available. For simple sensor array geometries, these results are compared to the corresponding performance when no source measurements are available. The results show that having a source present at an unknown range and bearing can do little to improve the sensor position estimator performance. A polynomial model for the locus of the sensors in the receiving array is assumed.