Survey on Recent Malicious Data Injection Detection Techniques

Wireless Sensor Networks are used to monitor the environmental parameters, for quick response of event detection. It is used to predict the happening of upcoming events such as, fire alarm system, intrusion detection, heart attack detection systems, military applications etc. But most of the time sensors are compromise by external entities, which arises a serious issue of security in such networks. Such Compromised sensors can report the false readings, which produce the inappropriate and most of the time dangerous responses. So there is a need of system that can appropriately identify the malicious data injected by unauthorized entities at sensor nodes. It is necessary because it is very difficult to detect such malicious data injection attacks on sensor node, if it is occur at multiple sensor nodes simultaneously in network. This paper makes survey of some recent approaches or techniques that are used for malicious data injected nodes in WSN. Also compare these approaches based on various parameters such as technique used, their advantages and limitations. At the end we suggest some research directions which will be used for further study in same field.