In Wireless Sensor Networks (WSN), the measurements that deviate from the normal behavior of sensed data are taken to be as outliers. It is also known as anomaly. An outlier is an observation, which departs so much from other analysis. Sources of outliers can be noise and errors, events, and malicious attacks on the network. The paper exploits the outlier detection techniques for wireless-sensornetwork based problem and proposes an outlier or anomaly detection scheme. This technique either learns or predicts a normality model of data, to classify the newly arrived measurements as normal or anomalous. The outlier detection schemes to deal with abnormal RS (Received Signal) data, but the RS measurements are known to be sensitive to the change of the environment.It allows the integration of results from neighboring network areas to detect correlated anomalies/attacks that involve multiple groups of nodes.An outlier detection approach that fuses data gathered from different nodes in a distributed sensor network is evaluated.