Numerous application domains use large scale sensor networks and they collect data that are used in decision making for critical infrastructures. Data are transmitted from multiple sources through intermediate nodes which aggregates the information. Sometimes there is a chance of malicious attack in the network and may introduce additional nodes in the network or compromise existing ones. Since this causes problems in correct decision making, providing data trustworthiness plays an important role in wireless sensor networks (WSN). Provenance management can cause several challenging requirements such as storage, bandwidth, energy etc. So here a trust evaluation in detection of provenance forgery is introduced. The proposed technique relies on in packet bloom filter to encode the provenance data. Also the provenance scheme is extended to detect the packet drop attack staged by the malicious nodes. Along with that trust management such as communication trust, energy trust and direct trust are also done to improve the efficiency of system.