WSNs are the temporary networks which are dynamic and self-maintainable. As being dynamic in nature and without any fixed infrastructure, the security becomes a great concern in WSNs. There are different types of security attacks like black hole, worm hole and grey hole but in this paper we are only focusing on the detection of malicious nodes and false reporting. We propose and implement a new intrusion-detection system named Enhanced Adaptive Acknowledgment (EAACK) specially designed for WSNs. Compared to contemporary approaches, EAACK demonstrates higher malicious behavior detection rates in certain circumstances while does not greatly affect the network performances.