This paper proposes a novel multi-level Distributed Intrusion Detection System in a Smart Home environment. The proposed approach aims to detect unexpected behaviors of a network component by exploiting the collaboration between the different IoT devices. The problem has been addressed by implementing an architecture based on a distributed hash table (DHT) that allows sharing network and system information between nodes. A distributed Intrusion Detection System, located in each node of the network, represents the core component to detect malicious behavior. The proposed Intrusion Detection system implements a binary classifier, based on a machine learning mechanism, which analyzes, in a novel way, the aggregation of features extracted from data coming from kernel, network and DHT level. In this work we present our idea with some preliminary experiments performed in order to compare different classifiers results on this kind of data with respect to a specific malicious behavior.