Dynamic State Estimation for Large Scale Systems Based on a Parallel Proximal Algorithm

In this paper, a novel method for parallel dynamic state estimation of large scale systems is presented. Since this task requires a high amount of computational resources, a novel solution is presented based on a minimization problem, including spatial and temporal constraints solved with a parallel proximal dual approach. In order to evaluate the performance of the proposed method, experiments are carried out to dynamically estimate sparse brain activity resulting from a large scale real brain model. To this end, simulated and real signals are used in the state estimation process. Results show that the temporal and spatial constraints consider the dynamic state evolution in time and the sparseness inherent to the estimated activity, respectively. Besides, the parallel solution significantly reduces the computational burden required to perform the task. It is worth noting that, for real electroencephalographic signals of each subject, the estimated activity into the brain is located in the areas removed during the successful surgery.