Search

The subject of this dissertation’s research is the integration of data obtained from sensors that monitor the sea surface beyond the line of horizon. The primary sensor network is composed of OTH (Over the Horizon) radars, while the secondary data source is the AIS system (Automated Information System), both satellite and terrestrial. With the integration of all collected data, a unique operational picture is formed in order to control maritime traffic on the high seas. The emphasis is on the control of the Exclusive Economic Zone in order to increase the safety of navigation, preserve natural resources and prevent illegal activities. In order to solve the integration problem in an efficient way, two new algorithms for data integration have been proposed and implemented. The dissertation first presents an algorithm for efficient data fusion within network consisting of 2 or more OTH radars. Then, an algorithm for efficient integration of OTH radar tracks and other naval sensors, primarily AIS systems, was presented. For both algorithms, the results obtained through exploitation in real working environment (the Gulf of Guinea) are presented. At the end of this dissertation, the architecture of a hybrid empirical-neural model is proposed to estimate the number of false targets caused by the strong atmospheric disturbances, including even meteo-tsunamis, and to reduce false alarms in the OTH sensor network. The architecture of this model is based on the PNN (Probabilistic Neural Network).