All interpretation of the surrounding world is based on models, verified or dismissed by new, possibly uncertain, information. This study highlights critical aspects of uncertainty with special emphasis on sensor networks in urban environment. Main topics deal with how to find, master and deliver uncertain information in dynamically changing sensor nets involving many different sensor types distributed over large geographic areas with difficult or unstable ways of communication. Multisensor data fusion offers methods to combine this wealth of information into a more manageable content, exemplified by tracking vehicles in an acoustic sensor net. Hosting a particularly complex milieu, urban environment provides great challenges to information retrieval, dealt with in the final chapter.