A robust algorithm for target tracking in presence of impulsive noise

This study addresses the challenge of target tracking in systems affected by non-Gaussian noise, with a particular focus on impulsive disturbances. Traditional tracking approaches often rely on minimizing the mean square error (MSE), which can be overly sensitive to outliers commonly introduced by impulsive noise. To overcome this limitation, we propose an alternative method based on minimizing the ℓ1 norm of the estimation error. This formulation provides a more robust criterion that treats all errors equitably, significantly reducing the influence of outliers. The proposed ℓ1-norm-based algorithm is especially effective in scenarios involving sharp transitions or boundary disruptions—typical sources of impulsive noise—ensuring reliable tracking performance. Simulation results confirm that the method offers strong resilience against impulsive noise, making it well-suited for real-world target tracking applications in harsh or unpredictable environments.