The world around us is going wireless, the unwired enterprise has long been a goal in most organizations, and the advent of 802.11 now makes achieving that goal a practical reality. A Mobile Adhoc Network (MANET) is nothing but collection of independent mobile nodes that can communicate to each other via radio waves. The mobile node communicate with each other by using two ways one is the mobile nodes communicate directly if they are in radio range of each other, whereas others needs the aid of intermediate nodes to route their packets. Due to its primary characteristics, such as wireless medium, dynamic topology, distributed cooperation, MANETs is susceptible to different kinds of security attacks like worm hole, black hole, rushing attack etc. Intrusion detection for wireless infrastructure is complex and filled of complexity mainly due to the dynamic character of MANETs, their highly forced nodes, and the lack of central monitoring points. Old IDSs are not applied easily to wireless network. Researchers have to develop new approaches or else adapt existing methodologies/approaches for MANETs. In this paper, we proposed and fully implement a new methodology for intrusion-detection system named Enhanced Adaptive Acknowledgment (EAACK) with Elliptic Curve Algorithm (ECC) specially designed for MANETs. Compared to contemporary approaches, EAACK demonstrates higher malicious-behavior-detection rates in certain circumstances while does not greatly affect the network performances.