Research On Routing Protocol For WSN Based On Ant-colony Algorithm

At present the research area of wireless sensor network has attracted a lot of attention in the international academic field for WSN. The rapid technological development of the WSN has improved the power for human to obtain data and we can predict that human beings can also get the most direct, most effective and promptest message by connecting the outside physical data with the next generation network. As we know that the WSN has been widely used in the area of battle, city management, biomedical treatment, environmental monitoring and emergencies and disasters. The WSN is often deployed in the extreme environment which the human can not get close to. And each sensor node in the network can only carry limited power which can not be supplemented. The sensor nodes in a WSN will organize the network by themselves and perform their jobs after they have been placed in the specific sites. So the problem on how to design a self-organized routing protocol which can perfectly get used to the character of WSN has become a hot research spot in the area of WSN.The ant colony algorithm can find out the optimal routing of a network by using the agent usually called ant in ant colony algorithm and this kind of algorithms are self-adapted, robust and flexible. So now more and more academic scholars in such fields are devoting themselves to researching the routing algorithms based on ant colony algorithm for wireless sensor network. But it has been found that the WSN routing protocol based on ant colony algorithm may lead to the problem of energy balance and low speed of convergence to the optimal routing. In order to solve the two problems mentioned above this thesis put up the ACO-QEE algorithm which is based on the basic ant colony algorithm and the thoughts of Q-learning. First of all, this algorithm applied the thoughts of Q-learning to the basic ant colony algorithm and this kind of method successfully speeds up the convergence while comparing with the basic ant colony algorithm. Secondly, the ACO-QEE algorithm improved the level of energy balance of the whole sensor network by considering the left energy level of each sensor node in the computational formula of transmission probability of the ant in the ant colony algorithm.