The exponential increase in the number of connected "things" and the proliferation in the usage of Internet of Things (IoT) devices has raised numerous challenges in terms of security, privacy, and interoperability. IoT devices are resource constrained in terms of computational power, onboard memory, network bandwidth, and energy availability which limits the implementation of cryptographic solutions. The heterogeneous nature of IoT devices makes them avenue for an attacker to exploit threats like spoofing, routing, MITM, and DoS attacks. With the current sophistication of threats IoT devices are subjected to, an Intrusion Detection System (IDS) is the preferred solution for IoT devices. An IDS continuously monitors incoming traffic, and analyzes it to detect possible signs of cyber threats. This research proposes a novel intelligent ensemble-based IDS that will reside in the IoT gateway. The uniqueness of our approach is to use an ensemble learning technique which combines multiple machine learning techniques in order to the improve the predictive performance and detection accuracy. Ensemble learning have been studied to increase the detection rate while obtaining better generalization performance due to the combination of several machine learning model also known as base learners. Three popularly known ensemble models (i.e. boosting, stacking, and voting) are used in evaluating the performance of our proposed IDS using three machine learning techniques: Decision Tree, Naive Bayes (NB), and k-Nearest Neighbor (KNN). Lastly, the proposed approach will be evaluated on two publicly available dataset; Intrusion Detection Evaluation Dataset (CIC-IDS2017) and N-BaIoT.