Multidimensional scaling-based passive emitter localisation from range-difference measurements

The problem of passive emitter localisation from range difference measurements has received considerable attentions. Unlike most of ordinary hyperbolic location methods focusing on minimising a loss function with respect to the range-difference vector, a simple estimator is proposed by exploiting a new multidimensional scaling analysis. On the basis of the optimisation of a loss function related to a scalar product matrix, it is robust for large measurement noise. The bias and variance of the proposed estimator is also derived. It is shown that the new method attains the Cramér–Rao lower bound for Gaussian range-difference noise at moderate noise level. Moreover, the threshold effect occurs latter than the conventional methods as the measurement noise increases. Additionally, they have comparable computational complexity.