Intrusion detection and border surveillance constitute a major application category for wireless sensor networks. A major goal in these applications is to detect intruders as they cross a border or as they penetrate a protected area. WSN is usually composed of small, low-cost devices that communicate wirelessly and have the capabilities of processing, sensing and storing. It typically consists of large number of resource-limited sensor nodes working in a self-organizing and distributed manner. Due to the ad hoc working style, once deployed, the inner structures and interactions within a WSN are difficult to observe from the outside. Intrusion detection using three-tier accelerometer sensors detect intrusion ships. The sensors deployed on the sea surface get tossed by ocean waves which makes them move randomly. This random movement of the node makes it difficult for most sensors to detect an intrusion. Network data processing with spatial and temporal correlations between nodes estimates the speed of a passing ship. Using signal processing and cooperative signal processing techniques the ocean waves and ship-generated waves are differentiated accordingly with their respective different energy spectrums. Though the algorithm detects multiple ships travelling along distances in different geographical areas it requires a relatively dense network especially to achieve a high detection ratio due to larger attenuation. To resolve such issues the proposed approach introduces the concept of Adaptive self-organizing localization algorithm. This is included in sink level detection to deal with invasion detection ships.