Dissertation for receiving a master's degree in electrical engineering
Electronics orientation
Abstract:
Wireless sensor networks consist of a large number of small sensor nodes, which have limitations in energy level, bandwidth, and processing power. and memory. Therefore, routing, clustering, reducing energy consumption and increasing network lifetime are the main challenges of routing in wireless sensor networks, which have been studied a lot. Many algorithms for clustering and routing have been presented, and all their efforts are to reduce energy consumption. We use the Bloom filter, which is one of the types of scrambling functions. When the Bloom filter is used, the bandwidth consumption can be reduced because the Bloom filter has a random data structure that occupies less space and the time to send data in the Bloom filter is short, and it reduces the network traffic in the wireless sensor network. And it provides a tool to accelerate and simplify packet routing protocols. In this thesis, using a suitable distributed hashing table algorithm, the relationship between the system data is established directly with minimal dependence on the hashing table, and this makes it possible to process queries with less time. The superior performance of this protocol has been proven in terms of increasing the useful life of the network compared to previous protocols such as LEACH and LEA2C, as well as the effect of the proposed cost function on its performance (by simulation). rtl;">
1-1. Introduction
In recent years, we have seen a lot of growth in the field of wireless sensor networks, and one of the most important tools for obtaining information and understanding the environment, which has focused extensive research, is wireless sensor networks. A wireless sensor network is a wireless network consisting of a large number of very small devices called sensor nodes [1]. Sensor nodes are generally equipped with sensing, processing and communication capabilities. Sensor nodes are spatially distributed and measure the conditions of their surrounding environment. The main task of the sensor node is to collect data points at regular time intervals and convert it into an electronic signal and broadcast the signal to the sink node or base station through reliable wireless communication media. With the emergence and development of microelectronics technology in the 70s, new sensors were noticed. Using microelectronic technology, cheap sensors with small dimensions and light weight were produced. New raw materials for making sensors were discovered and known, and subsequently, new principles were proposed for the practical purposes of information collection. The integration of sensor and signal-modifying electronic circuits has created significant opportunities for a wide range of applications. Nowadays, reducing the size and weight of sensors and increasing their sensitivity is the main goal of many research laboratories and different companies. But the shrinking of the size of the sensor nodes meant the shrinking of their energy generating batteries. The most important reason for the emergence and development of wireless sensor networks has been the applications of continuous monitoring of environments where it is difficult or impossible to reach and have a permanent human presence in them. Applications such as monitoring the eruption of an active volcano, monitoring difficult border areas, monitoring the strength of dams, bridges and roads, monitoring the battlefield or sensitive military areas, and so on. As a result, it is usually not possible to recharge or replace dead nodes (disabled due to the exhaustion of the energy source), because, as mentioned, these nodes are usually placed in harsh, harsh and inaccessible environments and conditions, and are often randomly scattered in the environment. Therefore, two points are of particular importance in the efficiency of sensor networks: one is the lifetime and the other is the amount of network coverage of these networks.Such a large network usually includes a number of distributed sensor nodes, which sensor nodes are actually small computers that have a user interface and limited components, which are important modules of wireless sensor networks, routers, and base stations that focus their organization on several hubs. But the completion of wireless sensor networks has its own theoretical and practical problems, such as energy consumption, reliability, fault tolerance, and scalability. Despite the progress made in this type of networks, sensor nodes due to their large number, small size, and contingent placement methods still rely on low-power batteries for their energy supply. The average power of each sensor node is about 1 milliwatt, and each sensor node has an embedded processor with low radio power and usually works with batteries. Each sensor has limited energy resources and its capability continues until it exhausts its energy. The practice is that the energy for the sensor network must be managed and the lifetime of the sensor must be carefully controlled [1][2]. Therefore, one of the most important issues in wireless sensor networks, which requires the problem of severe energy limitation and reliability for the design of network protocols. Therefore, it is vital to include distributed scrambler algorithms in the design of sensor networks with a long life, because it is possible to add new data in the scrambler tables in a short time. The time required for searching and adding are both functions of the table type and the amount of data. This time can reach the time order of 1 by choosing the appropriate table. Nowadays, dynamic power management methods that reduce the energy consumption of sensor networks after their design and deployment are of the highest importance. In recent years, for the dynamic management of power, attention has been paid to smart and capable tools such as distributed hashing algorithm. A distributed hashing algorithm is a large system consisting of parallel or distributed processing elements that includes a distributed hashing table, which is also a class of decentralized distributed system that provides a service reference like a hashing table [3]. The key and value pairs are stored in the distributed hashing table, and each efficient node can retrieve the values ??associated with a keyword, and the responsibility of mapping to maintain the key value is distributed among the nodes, and also to improve the spatial resolution of data collection. instead of a powerful single sensor, this multiple approach of distributing simple and inexpensive sensors in an area is useful. The data is not stored separately from the processing, because the data itself is connected. Considering the bandwidth of the sensor, which is between 50 and 250 kilobytes per second, the limited energy for calculations is an important obstacle to achieve success. Therefore, the distributed hashing algorithm as the backbone for processing on the network can be a suitable tool to be used in sensor networks and have a significant effect on reducing energy consumption and reliability of sensor networks and increasing their lifespan. Our goal of this research is to provide an optimal method to reduce energy consumption and reliability in sensor networks by using the features of the distributed hashing algorithm, which includes the multi-part and single-part distribution of nodes in scale, as well as File sharing and distribution system content [5]. In general, the hashing algorithm is called a function that converts a large amount of data (indeterminate amount of data) into a natural number so that data compression can be done at a high speed [6]. Hash performs a digest operation on the input stream, turning the input data stream into a small digest. When the hash value of two different inputs are the same, we say that a collision has occurred. This comes from the fact that the number of values ??of a hashing algorithm is very large. This is a one-way operation (irreversible) and whatever the volume of their input data stream, the output is a fixed value [7]. When we distribute the sensors, we find out what information each message we give to the sensor contains, which forms the distribution algorithm [13].