A deep learning approach to predict the number of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si3.svg" display="inline" id="d1e920"><mml:mi>k</mml:mi></mml:math>-barriers for intrusion detection over a circular region using wireless sensor networks

Wireless Sensor Networks (WSNs) is a promising technology with enormous applications in almost every walk of life. One the crucial WSNs intrusion detection and surveillance at border areas defence establishments. The are stretched over hundreds to thousands miles, hence, it not possible patrol entire region. As result, an enemy may enter from any point absence cause loss lives or destroy military can be feasible solution for problem areas. Detection nearby critical such as cantonments time-sensitive task delay few seconds have disastrous consequences. Therefore, becomes imperative design systems that identify detect soon comes within range deployed system. In this paper, we proposed deep learning architecture based on fully connected feed-forward Artificial Neural Network (ANN) accurate prediction number k-barriers fast prevention. We trained evaluated ANN model using four potential features, namely area circular region, sensing sensors, transmission sensor Gaussian uniform distribution. These features extracted through Monte Carlo simulation. doing so, found accurately predicts both distribution correlation coefficient (R = 0.78) Root Mean Square Error (RMSE 41.15) former R 0.79 RMSE 48.36 latter. Further, approach outperforms other benchmark algorithms terms accuracy computational time complexity.