The paper deals with the problem of computing optimal or suboptimal motion for a network of mobile sensors. The use of moving sensors means that for each point of the field under measurement asynchronous discrete time measures are given instead of continuous time ones, being possible to fix in advance the maximum time interval between two consecutive measures for the same point. The constraints here considered are on the full coverage of the fleld, with respect to the measurements, within the prefixed time interval and on the communication connections, between any pair of moving sensors, at any time. A solution, based on a local distributed approach, is proposed and compared with a centralized approach previously proposed, and here recalled, by the same authors. Some simulations show the effectiveness of the both solutions, putting in evidence advantages, disadvantages and differences.