Super MDS with heterogeneous information

We consider the simultaneous localization of multiple sources under non-line of sight (NLoS) conditions. The contribution here is an extension of the super multidimensional scaling (MDS) technique, to allow for both distance and angle information to be processed algebraically (without iteration) and simultaneously. Due to the flexibility of the method, the algorithm can be executed relying on the Nyström approximation, reducing computational complexity to that of a few matrix multiplications. Simulations demonstrate the superiority of the proposed solution compared to the standard C-MDS and the SMACOF algorithms.