The mobile adhoc network is highly dependent on the energy parameter of the mobilenodes, it suffers from different threats wich target the service performance of thed network. Thepackets transferred through intermediate nodes are subject to attack and the routing informationhasbecome a subject for malicious threats. In this paper a trust model is presented based on thecomputation of different parameters to improve the performance of intrusion detection in mobileadhoc network. The trust model computes the direct trust between different agents. Similarly themethod computes the indirect trust which is propagated by the direct trust which is computedbetween third party. The method monitors the network and stores the logs about the data packet beingforwarded. At each time when a node has a packet it performs neighbor discovery and routediscovery. Then for each route the method compute the trust weight and each intermediate node hasbeen verified for its location factor. Based on trust weight computed the method selects an efficientforwarding neighbor to send the data packet.Finally the method computes the recommendation trustbetween different agents to perform intrusion detection. Using all these measures the methodperforms intrusion detection on the mobile adhoc network.