Bistatic simultaneous transmitter localization and mapping

In a bistatic scenario, we consider the task of simultaneously estimating the positions of a non-cooperative transmitter and landmarks that reflect the transmitted signal, generating bistatic measurements at the receiver. Neither the position of the transmitter, nor the timing of transmissions are known to the receiver. With the use of the horizontal multi-path measurements caused by multiple reflection at the landmarks, the estimation of the unknown landmark locations is possible conditional upon the knowledge of the transmitter position. Thus two interdependent estimation problems arise in a similar fashion to the simultaneous localization and mapping (SLAM) problem. An approach using a factorized density in a similar way to FastSLAM 2.0 is used. In contrast to the SLAM problem, the differing receiver and transmitter positions lead to a measurement function that is not only non-linear but also non-invertible. Due to the nature of the measurements, a particle filter is used to estimate the landmark as well as the transmitter receiver system state. A practical example for simulations consists of an autonomous underwater vehicle (AUV) equipped with a towed hydrophone antenna array.