<p>Cooperative mobile target tracking has gained significant attention due to its wide range of applications and inherent robustness. This work addresses the problem of tracking a mobile target using a team of robots with limited sensing range and mobility constrained within designated boundaries. Two communication policies, event-triggered and timer-based are introduced, along with two tunable parameters for adaptive coordination, to ensure a harmonious trade-off between tracking quality and communication costs. A decentralized Extended Kalman Filter (EKF) is developed to estimate and predict the target’s state using minimal inter-robot data sharing, including multi-step prediction based on neighbor observations. To ensure fault tolerance, a novel reconfiguration strategy based on Mixed Integer Semi-definite Programming (MISDP) is proposed for maintaining network connectivity when a robot undergoes sensing degradation. A heuristic handover method is also implemented to enable seamless target transfer between robot regions, resulting in a significant improvement in tracking accuracy. The proposed framework is evaluated through simulations in the Gazebo environment using the Robot Operating System (ROS), and is compared against centralized EKF and other state-of-the-art approaches. Performance is assessed based on estimation error, message count, and sensitivity to tunable parameters, demonstrating improved robustness and reduced communication overhead.</p>