|
|
 |
| |
|
|
|
A Collaborative Composite Event Detection Approach in Wireless Sensor Network Using Fuzzy Assisted Decision System |
|
|
|
PP: 567-577 |
|
|
|
doi:10.18576/amis/120311
|
|
|
|
Author(s) |
|
|
|
S. Nalini,
A. Valarmathi,
|
|
|
|
Abstract |
|
|
| Recently, inWireless Sensor Networks (WSNs), several energy efficient event detection algorithms have been proposed that
aim to minimize battery storage, energy consumption and to maximize the network lifetime by framing reduced fuzzy rules related to
spatio-temporal properties for determining the event. Since sensors are prone to intermittent fault it would result to fault susceptible
reading that generates false alarm event, which would lead to lack of conforming the event at a higher confidence rate. Hence, during
decision making, the confidence level parameter of the sensor node is integrated by considering the spatio-temporal properties. In
this paper, we demonstrate through collaborative techniques by setting hypothesis to confirm the composite event collected from the
neighborhood nodes later the intelligent fuzzy decision system evaluates the rules with the composition of novel parameter confident
factor, which results in higher event detection accuracy. This work is implemented in MATLAB and simulations are carried out under
different network scenarios. The algorithm is evaluated with various metrics such as event detection accuracy, false positive rate, error
rate of the event and energy consumption. Based on the results of the simulations, we conclude that our intelligent hypothesis based on
the fuzzy decision system outperform than the well-established J48 decision tree classification algorithm |
|
|
|
|
|
|
|
