Abstract—Internet of Things technology, or IoT, is changing people’s lifestyles. Smartwatches, smart cars, smart homes, smart farms, and more – IoT has already been incorporated into a variety of products and services. To efficiently manage interactions between currently deployed smart things and applications, IoT vendors worldwide continually introduce different middleware platforms to meet application development requirements. Therefore, finding a suitable IoT middleware is a major issue faced by developers, especially when the system contains heterogeneous smart things and generates a vast amount of heterogeneous data. Most existing IoT middleware models do not satisfy all functional requirements and are tailored to specific system layers. To address these issues, this paper proposes a middleware model based on semantic web technologies and context-aware computing as an enhancement of the previously developed middleware MSOAH-IoT (A Middleware based on Service Oriented Architecture for Heterogeneity Issues within the Internet of Things). It uses a low-level ontology to automatically register, classify and then identify heterogenous smart things. The developed middleware provides a search engine to determine the appropriate smart object to respond to incoming requests of real-time measurements from the user/application layer.