Wireless Sensor Networks (WSNs) are amongst the most important of the new emerging technologies and have shown an explosive growth in recent years for monitoring physical phenomena. Large scale WSNs are known to suffer from coverage holes, large regions of deployment area where no sensing coverage can be provided. This could happen due to hardware failure, costs for deployment or redeployment. Coverage holes can affect the accurate representation of the natural phenomena that is monitored by WSN. This paper introduces a method to overcome the coverage holes problem by using an interpolation in those areas. It is shown that a phenomenon can be interpolated with high level of accuracy by using the available data coming from different nodes. However, due to energy limitations of sensor nodes it is imperative to perform this interpolation in an energy efficient manner that minimizes communications among nodes. An efficient technique for sensor node positioning is developed based on the following steps. First, we build a correlation model of the phenomena being monitored in a distributed manner. Next, a distributed interpolation technique based on the Kriging interpolation phenomena inside coverage holes is proposed. On the basis of estimated and calculated metric the sensor positions are optimized.