Optimization of energy consumption in wireless sensor networks by ant colony algorithm

Computer Engineering Master's Thesis

Trend: Software

Abstract

The current use of wireless sensor networks (Wireless Sensor Network) is in an expanded form. Due to the predominant use of a battery to supply the energy consumption of these sensors and also the lack of easy access to the sensors in many of these applications, engineers and researchers are looking for the design of routing protocols with low energy consumption characteristics and increasing the network lifetime. have been encouraged This research presents a distributed routing protocol based on the ant colony algorithm in order to improve the mentioned parameters. The ant colony algorithm is a metaheuristic algorithm that was introduced by Dorrigo and his colleagues to solve some combined optimization problems such as the traveling salesman problem. The ant colony algorithm has a series of features such as distributed computation, self-organization and positive feedback, which is used for path searching in communication networks. 

Finally, we implemented our project with the powerful MATLAB language and presented our simulations. The test results show a 40% reduction in energy consumption and a 3-fold increase in network life compared to the normal state. Keywords: wireless sensor networks, ant colony algorithm, energy consumption Chapter 1

General

1 Introduction

Today, the discussion of remote control and monitoring systems is one of the most challenging topics in the field of electronic computer science, so research is always looking for a solution to meet specific conditions and expected expectations; In the conditions of the same quality of work, the lower the ratio of cost to efficiency, the more popular that method will be. In order to be aware of the changes in the surrounding environment or the condition of each set, we need a series of equipment known as sensors, and these provide the intended changes (physical or chemical changes) in the form of a response, in order to measure the amount of changes or its existence. After collecting the required information, other operations can be performed based on the provided answer. Recent advances in the field of electronics and wireless communication have made it possible to have multifunctional sensor nodes with low power consumption and low cost, which are very small in terms of size and can communicate with each other for short distances. According to the theory of sensor networks, these small sensor nodes have sensing, data processing and communication equipment, as well as data-oriented energy sources. The main difference between sensor networks and other networks is in their very limited data-oriented nature, which makes the proposed methods for data transmission in other networks and even networks that have a structure similar to sensor networks to a large extent (such as ad hoc networks) not applicable in these networks. The development process of these networks is such that surely these networks will play an important role in our daily life in the near future. Among the applications that are currently proposed for the sensor network and their number is increasing day by day, we can mention applications such as tracking in wide geographic environments, security systems, monitoring large structures, monitoring patients with sensitive conditions, monitoring environmental parameters in areas where human presence is dangerous, and many other applications. Sensor networks are actually a collection of a large number of sensor nodes that are scattered in the environment and each of them pursues a specific goal autonomously and in cooperation with other nodes. The nodes are close to each other, and each node can communicate with another node and provide its information to another node, and finally, the status of the monitored environment is reported to a central node. The techniques and methods used in such networks are strongly dependent on the nature of the network application and the structure of the network topology, atmospheric and environmental conditions, limitations, etc. They are effective factors in network efficiency and cost parameters. Therefore, nowadays, wireless sensor networks are considered a very attractive and popular research field throughout the prestigious universities and computer, electronic and especially telecommunication research centers. Many researches and proposals have been presented in various topics and the amount of research in this field is still increasing. The main goal of all these efforts and providing solutions is to have a system with simple, easy and low-cost control methods.The main goal of all these efforts and providing solutions is to have a system with simple, easy and low-cost control methods that can finally meet our needs and withstand the limitations (bandwidth, energy, environmental interference, feeding, etc.) to extract optimal and efficient ideas. These ideas can be analyzed using mathematical laws and theoretical theories. For the reasons mentioned above, the discussion of sensor networks is currently increasing day by day in the IEEE. In recent years, several prestigious conferences have been held in the field of sensor networks is collected in the nodes of these networks. Since these networks are limited in terms of the amount of available energy and available processing resources, the methods proposed for other networks cannot be used in sensor networks. The Ant Colony algorithm is a metaheuristic algorithm that was introduced by Durigo and his colleagues to solve some optimization problems such as the traveling salesman problem. The ant colony algorithm has a series of features such as distributed computation, self-organization and positive feedback, which is used for path searching in communication networks. 6[

1-2 statement of the topic

Wireless sensor networks are collections of small sensor nodes that have the ability to monitor and sense their surroundings and send sensed data to a main station (well) [3].

As stated in [4], routing protocols in wireless sensor networks can be divided into three categories of flat, hierarchical, and location-based routing from the point of view of the network structure. In the flat model, all the nodes have the same role or work, but in the hierarchical model, the nodes play different roles in the network. In the location-based model, the position of the sensor nodes is used for data routing in the network. SPIN [1], flood, rumor, direct and all protocols are in flat routing category. LEACH protocols [2] are placed in a hierarchical category and GEAR and [4] GAF protocols are also placed in a location-based category. In [5], it is stated that sensors provide their energy with limited energy sources, including batteries, which leads to processing, communicating and storing information. According to the environmental conditions of the deployment of these networks, it is impossible and very expensive to replace the batteries of the sensor nodes. The main challenge [5] of WSN is the limitation of the energy available in the nodes, which has affected the survival of the network and hindered its progress.

In [6], colony life includes a mass of organisms that live together is A large number of elements or living organisms live together and all their behaviors are orderly and in the direction of the survival of the colony. The ant colony method is derived from the actual behavior of ants. Ants are social insects that live in colonies and their behavior is more towards the preservation and survival of the colony. Ants use the secretion of substances called pheromone to find food and for their navigation, and they choose the right and return path through tentacles that are able to detect pheromone. Ant colony algorithm was proposed for the first time in [6] by Marker Dorigo and his colleagues as a solution to find the shortest path. The proposed solution is affected by the real life of ants. Ants are social creatures and have a colony life. They are not able to live outside their colony. They recognize and choose their path by secreting pheromone. Artificial ants are used in this algorithm. The algorithm starts with the initial production of ants. The ants are placed in the initial state and the initial pheromone is quantified. Then all ants should be evaluated based on the objective function. After the evaluation, the pheromone and the initial path are assigned to them. They are distributed based on distance and pheromone level. When the ants are finished, the process must produce new ants, of course, if we have not reached the optimal solution