The Mobile Wireless Sensor Networks (MWSN), classified within MANETS, have multiple applications for critical situations management such as target monitoring and tracking in conflict zones, supporting urban security, critical infrastructure monitoring, remote locations exploration (i.e. aerospace exploration), and patients monitoring and care in health facilities, among others. All of these applications have requirements of certain intelligence in the network that can be used for network's self-configuration in order to find targets, guarantee connectivity and information availability until its reception. This paper proposes a MWSN architecture with an initial random distribution in a specific work area, and a centralized management to perform autonomous decision making about the movement and connectivity of the sensors. The work area presents mobile targets with interesting events which must be covered by the mobile sensors, and thus, send the collected information through the network to any base station available. Our work shows a dynamic mathematical model used to maximize targets' coverage and send its sensed information to the base stations available, while minimizing system's power consumption and maximizing operation time. The heuristic algorithm we used to construct and find a feasible solution is also shown.