Intrusion detection systems (IDSs) detect suspicious activities and possible intrusions in a system or a particular network at the moment at which these happen. To achieve their objectives the different entities that compose the IDS need to communicate with each other. Thus, it is important to keep in mind the integrity of the information, authentication and access control. Agents provide flexibility and distributed processing of the network traffic but increase the risk of possible intrusions. In this paper we present a new security scheme to verify the entities' integrity of an intrusion detection system based on autonomous agents throughout mobile cooperative agents using watermarking software techniques. Furthermore, software fingerprinting is used to assign a different mark to each IDS's entity in order to detect a possible collusion attack