TY - JOUR
T1 - Two Phased Routing Protocol Incorporating Distributed Genetic Algorithm and Gradient Based Heuristic in Clustered WSN
AU - Banerjee, Soumya
AU - Chowdhury, Chandreyee
AU - Chattopadhyay, Samiran
AU - Aslam, Nauman
PY - 2017/12
Y1 - 2017/12
N2 - In wireless cluster networks with a single non mobile sink, finding the optimal cluster assignment is a non-trivial problem. The inherently non centralized nature of wireless sensor networks poses a problem as majority of the learning algorithms are centralized. It is also desirable that single routing algorithm be applicable regardless of whether the sensor network is a dense single-hop network or a sparse multi-hop network. In this paper we present the two phased routing incorporating distributed genetic algorithm and gradient based heuristic (TRIGGER) as an attempt to solve these problems. In the first phase of TRIGGER a distributed (island model) genetic algorithm based clustering is employed to find a spatially optimal cluster assignment. In the second phase a gradient based routing forwards the already aggregated data to the sink. We discuss the rationale behind the two phased nature of TRIGGER. We demonstrate the effectiveness of TRIGGER with extensive simulations and discuss the results.
AB - In wireless cluster networks with a single non mobile sink, finding the optimal cluster assignment is a non-trivial problem. The inherently non centralized nature of wireless sensor networks poses a problem as majority of the learning algorithms are centralized. It is also desirable that single routing algorithm be applicable regardless of whether the sensor network is a dense single-hop network or a sparse multi-hop network. In this paper we present the two phased routing incorporating distributed genetic algorithm and gradient based heuristic (TRIGGER) as an attempt to solve these problems. In the first phase of TRIGGER a distributed (island model) genetic algorithm based clustering is employed to find a spatially optimal cluster assignment. In the second phase a gradient based routing forwards the already aggregated data to the sink. We discuss the rationale behind the two phased nature of TRIGGER. We demonstrate the effectiveness of TRIGGER with extensive simulations and discuss the results.
KW - wireless sensor networks
KW - genetic algorithm
KW - two phased approach
KW - gradient based heuristic
U2 - 10.1007/s11277-017-4786-y
DO - 10.1007/s11277-017-4786-y
M3 - Article
VL - 97
SP - 5401
EP - 5425
JO - Wireless Personal Communications
JF - Wireless Personal Communications
SN - 0929-6212
IS - 4
ER -