IoT systems are now being deployed worldwide to sense phenomena of interest. The existing IoT systems are often independent, which limits the use of sensor data to only one application. Semantic solutions have been proposed to support the reuse of sensor data across IoT systems and applications. This allows the integration of IoT systems for increased productivity by solving challenges associated with their interoperability and heterogeneity. Several ontologies have been proposed to handle different aspects of sensor data collection in IoT systems, ranging from sensor discovery to applying reasoning on collected sensor data for drawing inferences. In this paper, we study and categorize the existing ontologies based on the fundamental ontological concepts (e.g., sensors, context, location, and so on) required for annotating different aspects of data collection and data access in an IoT application. We identify these fundamental concepts by answering the 4Ws (What, When, Who, Where) and 1H (How) identified using the 4W1H methodology.