Neural network approach to the prediction of percentage data packet loss for wireless sensor networks

Wireless sensor networks are used in the field of communications and have gained enormous popularity in recent times. Depending upon the environment in which the wireless sensor network operates, the amount of noise level would differ and hence the data packet loss in wireless communication would vary. This paper presents a solution to the prediction of percentage data packet loss in the wireless sensor network in indoor and outdoor environment. It uses the artificial neural network (ANN) to predict the data packet loss and the Erasure Coding technique to find the actual percentage data packet lost in wireless sensor network. The results obtained from the ANN are compared to the respective ones yielded by the Erasure Coding technique and are found to exhibit satisfactory accuracy.