Areas of application of sensor networks. Wireless distributed sensor networks

History and scope of use

One of the first prototypes of a sensor network can be considered the SOSUS system, designed to detect and identify submarines. Wireless sensor network technologies began to actively develop relatively recently - in the mid-90s. However, only at the beginning of the 21st century the development of microelectronics made it possible to produce sufficiently cheap elemental base for such devices. Modern wireless networks are mainly based on the ZigBee standard. A considerable number of industries and market segments (manufacturing, various types of transport, life support, security) are ready for the implementation of sensor networks, and this number is constantly increasing. The trend is due to the increasing complexity of technological processes, the development of production, and the expanding needs of individuals in the segments of security, resource control and use of inventory. With the development of semiconductor technologies, new practical tasks and theoretical problems arise related to the applications of sensor networks in industry, housing and communal services, and households. The use of inexpensive wireless sensor-based parameter monitoring devices opens up new areas for the use of telemetry and control systems, such as:

• Timely identification of possible failures of actuators by monitoring parameters such as vibration, temperature, pressure, etc.;
• Real-time access control to remote systems of the monitoring object;
• Automation of inspection and maintenance of industrial assets;
• Commercial asset management;
• Application as components in energy- and resource-saving technologies;
• Monitoring eco-environmental parameters.

It should be noted that despite the long history of sensor networks, the concept of building a sensor network has not finally taken shape and has not been expressed in specific software and hardware (platform) solutions. The implementation of sensor networks at the current stage largely depends on the specific requirements of the industrial task. The architecture, software and hardware implementation is at the stage of intensive technology formation, which draws the attention of developers in order to find a technological niche for future manufacturers.

Technologies

Wireless sensor networks (WSNs) consist of miniature computing devices - motes, equipped with sensors (temperature, pressure, light, vibration level, location sensors, etc.) and signal transceivers operating in a given radio range. Flexible architecture and reduced installation costs distinguish wireless networks of smart sensors from other wireless and wired data transfer interfaces, especially when it comes to a large number of interconnected devices; a sensor network allows you to connect up to 65,000 devices. The constant reduction in the cost of wireless solutions and the increase in their operational parameters make it possible to gradually reorient from wired solutions in systems for collecting telemetric data, remote diagnostics, and information exchange. "Sensor network" is a well-established term today. Sensor Networks), denoting a distributed, self-organizing, resistant to failure of individual elements network of maintenance-free devices that do not require special installation. Each sensor network node may contain various sensors for monitoring the external environment, a microcomputer and a radio transceiver. This allows the device to carry out measurements, independently carry out initial data processing and maintain communication with an external information system.

802.15.4/ZigBee relayed short-range radio technology known as Sensor Networks. WSN - Wireless Sensor Network), is one of the modern trends in the development of self-organizing fault-tolerant distributed systems for monitoring and managing resources and processes. Today, wireless sensor network technology is the only wireless technology that can be used to solve monitoring and control tasks that are critical to the operating time of sensors. Sensors integrated into a wireless sensor network form a geographically distributed self-organizing system for collecting, processing and transmitting information. The main area of ​​application is control and monitoring of measured parameters of physical environments and objects.

• radio path;
• processor module;
• battery;
• various sensors.

A typical node can be represented by three types of devices:

• Network Coordinator (FFD - Fully Function Device);
  • carries out global coordination, organization and installation of network parameters;
  • the most complex of the three types of devices, requiring the largest amount of memory and power supply;

• Device with a full set of functions (FFD - Fully Function Device);
  • 802.15.4 support;
  • additional memory and power consumption allows you to serve as a network coordinator;
  • support for all types of topologies (“point-to-point”, “star”, “tree”, “mesh network”);
  • ability to act as a network coordinator;
  • the ability to access other devices on the network;
• (RFD - Reduced Function Device);
  • supports limited 802.15.4 features;
  • support for point-to-point and star topologies;
  • does not serve as a coordinator;
  • contacts the network coordinator and router;

Developer companies

There are various types of companies on the market:

Notes

Wikimedia Foundation. 2010.

See what “Wireless sensor networks” are in other dictionaries:

- (other names: wireless ad hoc networks, wireless dynamic networks) decentralized wireless networks that do not have a permanent structure. Client devices connect on the fly, forming a network. Each network node tries to forward... ... Wikipedia

It is proposed to rename this page to Wireless Self-Organizing Network. Explanation of the reasons and discussion on the Wikipedia page: To rename / December 1, 2012. Perhaps its current name does not correspond to the standards of modern ... ... Wikipedia

Wireless ad hoc networks are decentralized wireless networks that do not have a permanent structure. Client devices connect on the fly, forming a network. Each network node tries to forward data intended for other nodes. At the same time... ... Wikipedia

Wireless ad hoc networks are decentralized wireless networks that do not have a permanent structure. Client devices connect on the fly, forming a network. Each network node tries to forward data intended for other nodes. At the same time... ... Wikipedia

Architecture of a typical wireless sensor network A wireless sensor network is a distributed, self-organizing network of many sensors and actuators interconnected via a radio channel. Area... ...Wikipedia

To improve this article, would you like to: Rework the formatting in accordance with the rules for writing articles. Check the article for grammatical and spelling errors. Correct the article according to ... Wikipedia

Telemetry, telemetry (from other Greek τῆλε “far” + μέτρεω “I measure”) a set of technologies that allows remote measurements and collection of information to be provided to the operator or user, an integral part ... ... Wikipedia

Ultra-wideband (UWB) signals are radio signals (microwave signals) with “ultra-large” frequency bandwidth. Used for ultra-wideband radar and ultra-wideband radio communications. Contents 1 Definition 2 Regulation ... Wikipedia

The first Open Wireless Data Network Protocol developed for building automation and distributed facility management purposes. One Net can be used with many existing transceivers and... ... Wikipedia

The advantages of wireless sensor network technologies can be effectively used to solve various applied problems related to distributed collection, analysis and transmission of information.

Building automation

In some building automation applications, the use of traditional wired data transmission systems is not practical for economic reasons.

For example, you need to implement a new or expand an existing system in an existing building. In this case, the use of wireless solutions is the most acceptable option, because no additional installation work is required to disrupt the interior decoration of the premises, virtually no inconvenience is caused to employees or residents of the building, etc. As a result, the cost of implementing the system is significantly reduced.

Another example would be open-plan office buildings where it is not possible to specify the exact sensor locations during the design and construction phase. At the same time, the layout of offices can change many times during the operation of the building, therefore, the time and money spent on reconfiguring the system should be minimal, which can be achieved by using wireless solutions.

In addition, the following examples of systems based on wireless sensor networks can be given:

• monitoring temperature, air flow, occupancy and control of heating, ventilation and air conditioning equipment to maintain the microclimate;
• lighting control;
• energy management;
• collection of readings from residential meters for gas, water, electricity, etc.;
• monitoring the condition of load-bearing structures of buildings and structures.

Industrial automation

Until now, the widespread use of wireless communications in the field of industrial automation has been hampered by the low reliability of radio channels compared to wired connections in harsh industrial environments, but wireless sensor networks are radically changing the current situation, because by their nature, resistant to various kinds of disturbances (for example, physical damage to the node, the appearance of interference, changes in obstacles, etc.). Moreover, in some conditions, a wireless sensor network can provide even greater reliability than a wired communication system.

Solutions based on wireless sensor networks fully meet industry requirements:

• fault tolerance;
• scalability;
• adaptability to operating conditions;
• energy efficiency;
• taking into account the specifics of the applied task;
• economic profitability.

Wireless sensor network technologies can find application in the following industrial automation tasks:

• remote control and diagnostics of industrial equipment;
• equipment maintenance based on the current state (forecasting the safety margin);
• monitoring of production processes;
• telemetry for research and testing.

Other applications

The unique features and differences of wireless sensor networks from traditional wired and wireless data transmission systems make their use effective in a wide variety of areas. For example:

• security and defense:
  • control over the movement of people and equipment;
  • means of operational communications and reconnaissance;
  • perimeter control and remote surveillance;
  • assistance in rescue operations;
  • monitoring of property and valuables;
  • security and fire alarm system;
• environmental monitoring:
  • pollution monitoring;
  • Agriculture;
• healthcare:
  • monitoring the physiological state of patients;
  • location control and notification of medical personnel.

Architecture of a typical wireless sensor network

Wireless sensor network is a distributed, self-organizing network of many sensors (sensors) and actuators interconnected via a radio channel. Moreover, the coverage area of ​​such a network can range from several meters to several kilometers due to the ability to relay messages from one element to another.

History and scope of use

One of the first prototypes of a sensor network can be considered the SOSUS system, designed to detect and identify submarines. Wireless sensor network technologies began to actively develop relatively recently - in the mid-90s. However, only at the beginning of the 21st century the development of microelectronics made it possible to produce sufficiently cheap elemental base for such devices. Modern wireless networks are mainly based on the ZigBee standard. A considerable number of industries and market segments (manufacturing, various types of transport, life support, security) are ready for the implementation of sensor networks, and this number is constantly increasing. The trend is due to the increasing complexity of technological processes, the development of production, and the expanding needs of individuals in the segments of security, resource control and use of inventory. With the development of semiconductor technologies, new practical tasks and theoretical problems arise related to the applications of sensor networks in industry, housing and communal services, and households. The use of inexpensive wireless sensor-based parameter monitoring devices opens up new areas for the use of telemetry and control systems, such as:

• Timely identification of possible failures of actuators by monitoring parameters such as vibration, temperature, pressure, etc.;
• Real-time access control to remote systems of the monitoring object;
  • ensuring the protection of museum valuables
  • maintaining records of exhibits
  • automatic audit of exhibits
• Automation of inspection and maintenance of industrial assets;
• Commercial asset management;
• Application as components in energy- and resource-saving technologies;
• Monitoring environmental environmental parameters

It should be noted that despite the long history of sensor networks, the concept of building a sensor network has not finally taken shape and has not been expressed in specific software and hardware (platform) solutions. The implementation of sensor networks at the current stage largely depends on the specific requirements of the industrial task. The architecture, software and hardware implementation is at the stage of intensive technology formation, which draws the attention of developers in order to find a technological niche for future manufacturers.

Technologies

Wireless sensor networks (WSNs) consist of miniature computing devices - motes, equipped with sensors (temperature, pressure, light, vibration level, location sensors, etc.) and signal transceivers operating in a given radio range. Flexible architecture and reduced installation costs distinguish wireless networks of smart sensors from other wireless and wired data transfer interfaces, especially when it comes to a large number of interconnected devices; a sensor network allows you to connect up to 65,000 devices. The constant reduction in the cost of wireless solutions and the increase in their operational parameters make it possible to gradually reorient from wired solutions in systems for collecting telemetric data, remote diagnostics, and information exchange. "Sensor network" is a well-established term today. Sensor Networks), denoting a distributed, self-organizing, resistant to failure of individual elements network of maintenance-free devices that do not require special installation. Each sensor network node may contain various sensors for monitoring the external environment, a microcomputer and a radio transceiver. This allows the device to carry out measurements, independently carry out initial data processing and maintain communication with an external information system.

802.15.4/ZigBee relayed short-range radio technology known as Sensor Networks. WSN - Wireless Sensor Network), is one of the modern trends in the development of self-organizing fault-tolerant distributed systems for monitoring and managing resources and processes. Today, wireless sensor network technology is the only wireless technology that can be used to solve monitoring and control tasks that are critical to the operating time of sensors. Sensors integrated into a wireless sensor network form a geographically distributed self-organizing system for collecting, processing and transmitting information. The main area of ​​application is control and monitoring of measured parameters of physical environments and objects.

The accepted IEEE 802.15.4 standard describes wireless channel access control and the physical layer for low-speed wireless personal area networks, that is, the two lower layers according to the OSI network model. The “classical” sensor network architecture is based on a typical node, which includes an example of a typical RC2200AT-SPPIO node:

• radio path;
• processor module;
• battery;
• various sensors.

A typical node can be represented by three types of devices:

• Network Coordinator (FFD - Fully Function Device);
  • carries out global coordination, organization and installation of network parameters;
  • the most complex of the three types of devices, requiring the largest amount of memory and power supply;

• Device with a full set of functions (FFD - Fully Function Device);
  • 802.15.4 support;
  • additional memory and power consumption allows you to serve as a network coordinator;
  • support for all types of topologies (“point-to-point”, “star”, “tree”, “mesh network”);
  • ability to act as a network coordinator;
  • the ability to access other devices on the network;
• (RFD - Reduced Function Device);
  • supports limited 802.15.4 features;
  • support for point-to-point and star topologies;
  • does not serve as a coordinator;
  • contacts the network coordinator and router;

Notes

1. 1 2 3 Ragozin D.V.. Modeling of synchronized sensor networks. Programming problems. 2008. No. 2-3. Special issue – 721-729 pp.
2. Baranova E. IEEE 802.15.4 and its software add-on ZigBee. // Telemultimedia, May 8, 2008.
3. Levis P., Madden S., Polastre J. and dr. “TinyOS: An operating system for wireless sensor networks” // W. Weber, J.M. Rabaey, E. Aarts (Eds.) // In Ambient Intelligence. – New York, NY: Springer-Verlag, 2005. – 374 p.
4. Algoritmic Considerations of Wireless Sensor Networks. // Miroslaw Kutulowski, Jacek Cichon, Przemislaw Kubiak, Eds. – Poland, Wrozlaw: Springer, 2007.
5. Intelligent systems based on sensor networks. - www.ipmce.ru/img/release/is_sensor.pdf // Institute of Precision Mechanics and Computer Engineering named after. S.A. Lebedev RAS, 2009.
6. Fully completed ZigBee modules from RadioCrafts. - kit-e.ru/articles/wireless/2006_3_138.php // Components and technologies.
7. ZigBee/802.15.4 protocol stack on the Freescale Semiconductor platform - www.freescale.com/files/abstract/global/RUSSIA_STKARCH_OV.ppt, 2004

download
This abstract is based on

Almost all areas of life in the 21st century depend on information and communication technologies (ICT). Data is exchanged not only by people, but also by all kinds of intelligent systems, mobile phones, wearable devices, ATMs, and sensors. At least 5 billion devices are already connected to the Internet of Things. The functioning of any large complexes - industrial, energy, agricultural enterprises, shopping centers, museums, offices, residential buildings - is associated with constant monitoring of the situation on their territory. Sensitive sensors monitor the health of equipment in real time, organize the interaction of devices with each other, and warn about the need to replace them or about emergency situations. With rapidly growing volumes of data, you need a simple and convenient way to share it between devices and data centers.

Print version:

Wireless sensor networks (WSNs, Wireless Sensor Networks), consisting of wireless sensors and control devices and capable of self-organization using intelligent algorithms, demonstrate large-scale prospects for use in monitoring human health, the state of the environment, the functioning of production and transport systems, and accounting for various resources. etc. This issue of the newsletter presents technological trends in the field of WSN related to ensuring the constant operation of wireless sensors and their application in two areas of the modern economy - advanced manufacturing and smart energy.

Self-charging touch devices

For the development of wireless sensor networks, it is important to solve the problem of their power supply. A promising trend is the creation of durable autonomous devices with minimal energy consumption - converted from external sources.

Wireless sensor devices can, for example, be powered by radio energy sent to them from some kind of transmitter (similar to radio frequency identification (RFID) devices or contactless smart cards). This energy is used by the device both to recharge the sensor and to generate a response signal with information about the current state of the monitored object.

Another method is passive conversion of energy from the external environment (energy harvesting): solar (outside the room in fairly clear weather), thermal, mechanical vibration energy (from nearby devices - assembly machines, conveyors, etc.), vibration energy of the sensor itself (in the case of wearable devices), background radio emissions from surrounding electrical appliances (including Wi-Fi).

Effects By reducing highly toxic waste from used power sources, the negative impact on the environment is reduced. The use of self-charging sensor devices will contribute to the development of wireless Internet of Things devices, personalized medicine (wearable and implantable devices that monitor human health indicators), as well as green technologies in the energy sector, industry (especially in hazardous industries) and other areas .

Market estimates 30.1 billion By 2020, there will be 30.1 billion wireless devices in the world. The potential share of self-charging devices will reach 30–65%. The maximum manifestation of the trend is predicted in 2020–2030.

Drivers and barriers The relevance of the trend is constantly growing due to the miniaturization of wireless devices and the reduction in their cost. The development of this area is hampered by the insufficient amount of energy converted from the external environment in current models of wireless sensors, which is necessary to monitor the condition of complex equipment and periodically send information to the data processing center.

Implementation of advanced manufacturing based on wireless sensor networks

Irrational use of resources and production capacity, generation of large amounts of environmentally polluting waste, lack of constant monitoring of the condition of facilities at enterprises - these and other problems of modern industry stimulate the transition to an advanced manufacturing model. It is characterized by the use of new materials and environmentally friendly technologies (green technologies), as well as the widespread use of ICT and intelligent systems, in particular robotics and wireless sensor networks.

Industrial wireless sensor networks (IWSS, Industrial Wireless Sensor Networks) are the most important factor in the implementation of advanced manufacturing. To manage and monitor the state of objects at an enterprise (equipment, conveyors, assembly devices, reactors), a set of interconnected wireless sensors and information systems is used that process data from sensors and interact with controlled objects using control devices. Such an automated system responds to any changes in indicators at the enterprise, notifies personnel about accidents and problem situations, analyzes the efficiency of equipment use, assesses the level of environmental pollution and the volume of waste produced.

Effects The implementation of advanced manufacturing will increase the efficiency, accuracy and flexibility of process management in large enterprises and the quality of products, solve the issues of optimizing resources and identifying bottlenecks. Minimizing risks in production (due to its greater automation) and reducing emissions will significantly reduce the industrial burden on the environment.

Market estimates $3.8 billion will reach the market size of industrial wireless sensor networks in 2017. The widespread spread of ischemic heart disease is predicted by 2017–2020.

Drivers and barriers The development of self-charging sensor devices, the transition to digital control of production equipment, and the mass introduction of robotic systems are stimulating the development of the trend. Insufficient standardization of wireless sensor networks and outdated ICT infrastructure in many industrial enterprises are slowing down the transition to an advanced manufacturing model.

Smart grids

The global problem of irrational use of electricity is especially relevant for Russia. High costs for generating electricity increase the cost of production, which places a double burden on the end consumer. To improve the efficiency and reliability of power systems, many countries are moving to the concept of smart grids.

Such a network controls in real time all generating sources, transmission and distribution networks and facilities that consume electricity connected to it. To manage a smart energy grid, wireless sensor networks are used that monitor the volume of energy production and energy consumption in its different sections. With the help of information systems, the optimal distribution of energy in the network is calculated, forecasts are made for different seasons and periods of the day, energy production and its delivery are synchronized, and the safety of power lines is monitored. To improve the efficiency of the power grid, its non-critical elements are turned off during periods of reduced activity.

Effects Continuous monitoring and control of devices in the power system reduces the number of power outages. Significant cost savings in its production lead to positive effects for industry, the social sphere and the environment (now in Russia the share of energy in air pollution is 26.8% - the maximum value in comparison with other areas).

Market estimates $63 billion by 2020 will amount to the global market for smart power grid technologies with an average annual growth rate of more than 8 %. 2018–2025 - the period of maximum manifestation of the trend.

Drivers and barriers The introduction of stricter environmental regulations and standards is accelerating the trend. The right time to introduce a smart grid in Russia is the planned renewal of electricity infrastructure facilities. The transition to the smart grid concept is complicated by the large scale of the industry, the high cost and significant time spent on its technological update. Periodic shutdowns of power plants during the modernization of energy networks pose a problem for consumer companies.

Monitoring of global technological trends is carried out by the Institute of Statistical Research and Economics of Knowledge of the Higher School of Economics () within the framework of the Fundamental Research Program of the National Research University Higher School of Economics.

The following sources were used in preparing the trendletter: Forecast of scientific and technological development of the Russian Federation until 2030(prognoz2030.hse.ru), scientific journal materials "Foresight"(foresight-journal.hse.ru), data Web of Science, Orbit, idc.com, marketsandmarkets.com, wintergreenresearch.com, greentechmedia.com, greenpatrol.ru, etc.

Wireless Sensor Networks: An Overview

Akuldiz I.F.

Translation from English: Levzhinsky A.S.

annotation

The article describes the concepts of sensor networks, the implementation of which became possible as a result of the combination of microelectro-mechanical systems, wireless communications and digital electronics. The tasks and potential of sensor networks have been studied, and a review of the facts influencing their development has been made. The architecture of building sensor networks, developed algorithms and protocols for each layer of the architecture are also considered. The article explores questions about the implementation of sensor networks.

1. Introduction

Recent advances in micro-electro-mechanical systems (MEMS) technologies, wireless communications, and digital electronics have enabled the creation of low-cost, low-power, feature-rich motes that are small and communicate directly with each other. Sensor networks are based on the collaboration of a large number of tiny nodes, which consist of data collection and processing modules, a transmitter. Such networks have significant advantages over a set of traditional sensors. Here are two key features of traditional sensors: Sensors can be located far from the phenomenon being observed. This approach requires many sensors that use some sophisticated techniques to pick out targets from the noise.
Multiple sensors can be deployed that only collect data. Carefully design sensor positions and topology. They will transmit observations to central nodes, where data collection and processing will be carried out.
A sensor network consists of a large number of nodes (motes) that are densely located close to the observed phenomenon. The position of the motors does not need to be pre-calculated. This allows them to be randomly deployed in hard-to-reach areas or used for relief operations that require a quick response. On the other hand, this means that network protocols and algorithms for the operation of motes must have the ability to self-organize. Another unique feature of sensor networks is the collaboration of individual nodes. Motes are equipped with a processor. Therefore, instead of transmitting the original data, they can process it, perform simple calculations and transmit further only the necessary and partially processed data. The features described above provide a wide range of applications for sensor networks. Such networks can be used in healthcare, military and security applications. For example, physiological data about a patient can be monitored remotely by a physician. This is convenient both for the patient and allows the doctor to understand his current condition. Sensor networks can be used to detect foreign chemical agents in air and water. They can help determine the type, concentration and location of contaminants. In essence, sensor networks enable a better understanding of the environment. We assume that in the future, wireless sensor networks will be an integral part of our lives, more so than modern personal computers. The implementation of these and other projects that require the use of wireless sensor networks requires special methods. Many protocols and algorithms were developed for traditional wireless ad hoc networks, so they are not well suited to the unique characteristics and requirements of sensor networks. Here are the differences between sensor and peer-to-peer networks: The number of nodes in a sensor network can be several orders of magnitude higher than the number of nodes in a peer-to-peer network.
The nodes are densely located.
Nodes are prone to failure.
The topology of sensor networks can change frequently
Peers primarily use broadcast messages, while most peer-to-peer networks are based on point-to-point communications.
Nodes are limited in power, computing power, and memory.
Nodes cannot have a global identification number (ID) due to the large overhead and large number of sensors.
Since the nodes in the network are densely located, neighboring nodes can be very close to each other. Therefore, multi-hop communications in sensor networks will consume less energy than direct communications. In addition, low power data transmission signal can be used, which is useful in covert surveillance. Multi-hop communications can effectively overcome some of the difficulties in long-distance signal propagation in wireless communications. One of the most important constraints for nodes is low power consumption. Motes have limited energy sources. So, while traditional networks focus on achieving high signal quality, mote network protocols must focus primarily on energy conservation. They must have mechanisms that give the user the ability to extend the life of the mote by either reducing the throughput or increasing the latency of data transmission. Many researchers are currently involved in developing circuits that fulfill these requirements. In this article we will review the protocols and algorithms that currently exist for sensor networks. Our goal is to provide a better understanding of current research issues in this area. We will also attempt to explore design constraints and identify tools that can be used to solve design problems. The article is organized as follows: in the second section, we describe the potential and utility of sensor networks. In Section 3 we discuss the factors that influence the design of such networks. We will consider a detailed study of existing methods in this area in section 4. And we will summarize in section 5.

2. Application of wireless sensor networks

Sensor networks can consist of different types of sensors, such as seismic, magnetic field, thermal, infrared, acoustic, which are capable of making a wide variety of measurements of environmental conditions. For example, such as:
temperature,
humidity,
car traffic,
state of lightning,
pressure,
soil composition,
noise level,
the presence or absence of certain objects,
mechanical load
dynamic characteristics such as speed, direction and size of an object.
Motes can be used for continuous sensing, detection and identification of events. The concept of micro sensing and wireless connectivity promise many new applications for such networks. We have classified them into main areas: military applications, environmental research, healthcare, home and other commercial applications. But it is possible to expand this classification and add more categories, such as space exploration, chemical processing and disaster relief.

2.1. Military applications

Wireless sensor networks can be an integral part of military command, communications, intelligence, surveillance and positioning systems (C4ISRT). Fast deployment, self-organization and fault tolerance are the characteristics of sensor networks that make them a promising tool for solving problems. Since sensor networks can be based on a dense deployment of disposable and cheap nodes, destroying some of them during military operations will not affect the military operation as much as destroying traditional sensors. Therefore, the use of sensor networks is better suited for battles. Let's list some other ways to use such networks: monitoring weapons and ammunition of friendly forces, observing battles; location orientation; assessment of damage from battles; detection of nuclear, biological and chemical attacks. Monitoring of friendly forces, weapons and ammunition: Leaders and commanders can constantly monitor the status of their troops, the status and availability of equipment and ammunition on the battlefield using sensor networks. Every vehicle, equipment and critical ammunition can have sensors attached to it that report their status. This data is collected together at key nodes and sent to managers. Data can also be forwarded to higher levels of the command hierarchy to be combined with data from other parts. Battle Observations: Critical areas, paths, routes and straits can be quickly covered by sensor networks to study the activities of enemy forces. During operations or after new plans are developed, sensor networks can be deployed at any time to monitor the battle. Reconnaissance of Enemy Forces and Terrain: Sensor networks can be deployed in critical areas, and valuable, detailed and timely data on enemy forces and terrain can be collected within minutes before the enemy can intercept it. Targeting: Sensor networks can be used in smart munition targeting systems. Post-battle damage assessment: Immediately before or after an attack, sensor networks can be deployed in the target area to collect damage assessment data. Detection of Nuclear, Biological and Chemical Attacks: When using chemical or biological weapons, the use of which is close to zero, timely and accurate detection of chemical agents is essential. Sensor networks can be used as warning systems for chemical or biological attacks, and the data collected in a short time will help to dramatically reduce the number of victims. Sensor networks can also be used for detailed reconnaissance once such attacks are detected. For example, it is possible to carry out reconnaissance in the event of radiation contamination without exposing people to radiation.

2.2. Environmental Application

Some of the areas in ecology where sensor networks are used: tracking the movements of birds, small animals and insects; monitoring the state of the environment in order to identify its impact on crops and livestock; irrigation; large-scale earth monitoring and planetary exploration; chemical/biological detection; detection of forest fires; meteorological or geophysical research; flood detection; and pollution research. Forest fire detection: Because motes can be strategically and densely deployed in a forest, they can relay the exact origin of a fire before a fire becomes uncontrollable. Millions of sensors can be deployed on a continuous basis. They can be equipped with solar panels, as the nodes can be left unattended for months or even years. The motes will work together to perform distributed sensing tasks and overcome obstacles such as trees and rocks that block wired sensors. Mapping the biostatus of the environment: requires sophisticated approaches to integrate information across temporal and spatial scales. Advances in remote sensing technology and automated data collection have significantly reduced research costs. The advantage of these networks is that nodes can be connected to the Internet, which allows remote users to control, monitor and observe the environment. Although satellite and airborne sensors are useful in observing large diversity, such as the spatial complexity of dominant plant species, they do not allow observation of the small elements that make up the majority of an ecosystem. As a result, there is a need for field deployment of wireless sensor network nodes. One example application is the compilation of a biological environmental map of a nature reserve in Southern California. Three sites are covered by a network, each with 25-100 nodes, which are used for continuous monitoring of the environment. Flood detection: An example of flood detection is the warning system in the United States. Several types of sensors placed in the warning system detect rainfall, water levels and weather. Research projects such as the COUGAR Device Database Project at Cornell University and the DataSpace Project at Rutgers University are exploring different approaches to interacting with individual nodes on a network to obtain snapshots and long-running data. Agriculture: Sensor networks also have the advantage of being able to monitor pesticide levels in water, soil erosion levels, and air pollution levels in real time.

2.3. Application in medicine

One application in medicine is devices for the disabled; patient monitoring; diagnostics; monitoring the use of medicines in hospitals; collection of human physiological data; and monitoring doctors and patients in hospitals. Human Physiological Monitoring: Physiological data collected by sensor networks can be stored for long periods of time and can be used for medical research. Installed network nodes can also monitor the movements of older people and, for example, prevent falls. These nodes are small and provide the patient with greater freedom of movement, while at the same time allowing doctors to identify symptoms of the disease in advance. In addition, they help ensure a more comfortable life for patients compared to hospital treatment. To test the possibility of such a system, a “Healthy Smart Home” was created at the Faculty of Medicine of Grenoble–France. . Monitoring doctors and patients in the hospital: each patient has a small and lightweight network node. Each node has its own specific task. For example, one might monitor heart rate while the other takes blood pressure readings. Doctors can also have such a node, it will allow other doctors to find them in the hospital. Monitoring of medications in hospitals: Nodes can be attached to medications, then the chances of dispensing the wrong medication can be minimized. Thus, patients will have nodes that determine their allergies and the necessary medications. Computerized systems as described in have shown that they can help minimize side effects from erroneous drug dispensing.

2.4. Home use

Home automation: Smart nodes can be integrated into household appliances such as vacuum cleaners, microwave ovens, refrigerators and VCRs. They can communicate with each other and with the external network via the Internet or satellite. This will allow end users to easily manage devices at home, both locally and remotely. Smart Environment: Smart environment design can have two different approaches i.e. human centered or technology centered. In the case of the first approach, the smart environment must adapt to the needs of end users in terms of interaction with them. For technology-centric systems, new hardware technologies, network solutions, and middleware applications must be developed. Examples of how nodes can be used to create a smart environment are described in. Nodes can be built into furniture and appliances, they can communicate with each other and the room server. A room server can also communicate with other room servers to learn about services they can offer, such as printing, scanning, and faxing. These servers and sensor nodes can be integrated into existing embedded devices and constitute self-organizing, self-regulating and adaptive systems based on the control theory model as described in Ref.

3. Factors influencing the development of sensor network models.

The design of sensor networks depends on many factors, which include fault tolerance, scalability, production costs, type of operating environment, sensor network topology, hardware limitations, communication model and energy consumption. These factors are considered by many researchers. However, none of these studies have fully taken into account all the factors that influence network design. They are important because they serve as a guideline for developing the protocol or algorithms for sensor networks. In addition, these factors can be used to compare different models.

3.1. fault tolerance

Some nodes may fail due to lack of power, physical damage, or third-party interference. Node failure should not affect the operation of the sensor network. This is a matter of reliability and fault tolerance. Fault tolerance - the ability to maintain the functionality of a sensor network without failure if a node fails. The reliability Rk(t) or fault tolerance of a node is modeled using a Poisson distribution to determine the probability of no node failure in the time period (0; t) It is worth paying attention to the fact that protocols and algorithms can be oriented to the level of fault tolerance required for building sensor networks . If the environment in which the nodes are located is less susceptible to interference, then the protocols may be less resilient. For example, if nodes are embedded in a home to monitor humidity and temperature levels, the requirements for fault tolerance may be low, since these kinds of sensor networks cannot fail and environmental “noise” does not affect their operation. On the other hand, if nodes are used on the battlefield for surveillance, then resiliency must be high since surveillance is critical and nodes may be destroyed during military operations. As a result, the level of fault tolerance depends on the application of sensor networks and models must be designed with this in mind.

3.2. Scalability

The number of nodes deployed to study a phenomenon can be on the order of hundreds or thousands. Depending on the application, the number can reach extreme values ​​(millions). New models should be able to handle this number of nodes. They must also use high density sensor networks, which can range from a few nodes to several hundred in an area that can be less than 10 m in diameter. Density can be calculated according to,

3.3. Production costs

Since sensor networks consist of a large number of nodes, the cost of one node must be such as to justify the total cost of the network. If the cost of the network is higher than the deployment of traditional sensors, then it is not economically justified. As a result, the cost of each node should be low. Now the cost of a node using a Bluetooth transmitter is less than $10. The price for PicoNode is around $1. Therefore, the cost of a sensor network node must be much less than $1 to make their use economically justifiable. The cost of a Bluetooth node, which is considered a cheap device, is 10 times higher than the average prices for sensor network nodes. Note that the node also has some additional modules, such as a data acquisition module and a data processing module (described in section 3.4.) In addition, they can be equipped with a location system or a power generator depending on the application of sensor networks. As a result, the cost of the node is a difficult issue, given the amount of functionality even at a price of less than $1.

3.4. Hardware Features

A sensor network node consists of four main components, as shown in Fig. 1: data acquisition unit, processing unit, transmitter and power supply. The availability of additional modules depends on the application of the networks, for example, there may be location modules, a power generator and a mobilizer (MAC). The data acquisition module usually consists of two parts: sensors and analog-to-digital converters (ADC). The analog signal generated by the sensor based on the observed phenomenon is converted into a digital signal using an ADC and then fed to the processing unit. The processing module, which uses integrated memory, manages procedures that allow it to collaborate with other nodes to perform assigned monitoring tasks. The transmitter unit (transceiver) connects the node to the network. One of the most important components of the node is the power supply. The power supply can be recharged, for example, using solar panels.

Most nodes transmitting and collecting data need to know their location with high accuracy. Therefore, a location determination module is included in the overall scheme. Sometimes a mobilizer may be needed to move the unit as needed to perform the assigned tasks. All of these modules may need to be housed in a matchbox-sized enclosure. The size of the knot can be less than a cubic centimeter and is light enough to remain in the air. Besides size, there are some other hard restrictions on nodes. They have to :
consume very little energy,
work with a large number of nodes at short distances,
have low production costs
be autonomous and work without supervision,
adapt to the environment.
Since nodes can become unavailable, the life of a sensor network depends on the power supply of individual nodes. Food is a limited resource and due to size restrictions. For example, the total energy reserve of a smart node is about 1 J. For Wireless Integrated Network of Sensors (WINS), the average charge level must be less than 30 LA to ensure long operating time. It is possible to extend the life of sensor networks by using rechargeable batteries, for example, by obtaining energy from the environment. Solar panels are a prime example of the use of recharging. The node data module can be a passive or active optical device, as in a smart node, or a radio frequency (RF) transmitter. RF transmission requires a modulation module that uses a certain bandwidth, filtering module, demodulation, which makes them more complex and expensive. In addition, there may be losses in data transmission between two nodes due to the fact that the antennas are located close to the ground. However, radio communication is preferred in most existing sensor network designs because data transmission frequencies are low (typically less than 1 Hz) and transmission cycle rates are high due to short distances. These characteristics allow the use of low radio frequencies. However, designing energy-efficient and low-frequency radio transmitters is still a technical challenge, and existing technologies used to manufacture Bluetooth devices are not efficient enough for sensor networks because they consume a lot of power. Although processors are constantly becoming smaller and more powerful, the processing and storage of the node is still its weak point. For example, the smart node processing module consists of a 4 MHz Atmel AVR8535 processor, a microcontroller with 8 KB of instructions, flash memory, 512 bytes of RAM, and 512 bytes of EEPROM. This module, which has 3500 bytes for the OS and 4500 bytes of free memory for code, uses the TinyOS operating system. The processing module of another lAMPS node prototype has a 59-206 MHz SA-1110 processor. IAMPS nodes use the multi-threaded L-OS operating system. Most data collection tasks require knowledge of the node's position. Since nodes are typically located randomly and without supervision, they must cooperate using a location system. Location sensing is used in many sensor network routing protocols (more details in Section 4). Some propose that each node should have a global positioning system (GPS) module that operates with an accuracy of up to 5 meters. The paper argues that equipping all nodes with GPS is not necessary for sensor networks to work. There is an alternative approach, where only some nodes use GPS and help other nodes determine their position on the ground.

3.5. Network topology

The fact that nodes can become unavailable and are subject to frequent failures makes network maintenance a challenging task. From hundreds to several thousand nodes can be located on the territory of a sensor network. They deploy ten meters from each other. The node density can be higher than 20 nodes per cubic meter. The dense location of many nodes requires careful network maintenance. We will consider issues related to maintaining and changing the network topology in three stages:

3.5.1. Pre-deployment and deployment of nodes itself can consist of a massive scattering of nodes or installing each one separately. They can be expanded:

Scattered from an airplane,
by placing it in a rocket or projectile
thrown out by catapult (for example, from a ship, etc.),
plant location
each node is placed individually by a person or robot.
Although the sheer number of sensors and their automated deployment usually precludes placement according to a carefully designed plan, designs for initial deployment should:
reduce installation costs,
eliminate the need for any pre-organization and pre-planning,
increase placement flexibility,
promote self-organization and fault tolerance.

3.5.2. Post-network phase

After the network is deployed, a change in its topology is associated with a change in the characteristics of the nodes. Let's list them:
position,
accessibility (due to interference, noise, moving obstacles, etc.),
battery charge,
malfunctions
change in assigned tasks.
Nodes can be expanded statically. However, device failure is common due to battery depletion or destruction. Sensor networks with high node mobility are possible. In addition, nodes and networks perform different tasks and may be subject to intentional interference. Thus, the structure of a sensor network is prone to frequent changes after deployment.

3.5.3. Additional Node Deployment Phase

Additional nodes can be added at any time to replace faulty nodes or due to changing tasks. Adding new nodes creates the need to reorganize the network. Dealing with frequent changes in the topology of a peer-to-peer network, which contains many nodes and has very strict power consumption restrictions, requires special routing protocols. This issue is discussed in more detail in section 4.

3.6. Environment

The nodes are densely located very close to or directly within the observed phenomenon. Thus, they operate unattended in remote geographic areas. They can work
at busy intersections,
inside big cars
at the bottom of the ocean,
inside a tornado,
on the surface of the ocean during a tornado,
in biologically and chemically contaminated areas
on the battlefield,
in a house or large building,
in a large warehouse,
attached to animals
attached to fast moving vehicles
in a sewer or river along with a stream of water.
This list gives an idea of ​​the conditions under which nodes can operate. They can operate under high pressure on the ocean floor, in harsh environments, among debris or on the battlefield, in extreme temperatures, such as in an aircraft engine nozzle or in arctic regions, in very noisy places where there is a lot of interference.

3.7. Data transfer methods

In a multi-hop sensor network, nodes communicate wirelessly. Communication can be carried out via radio, infrared or optical media. In order to use these methods globally, the transmission medium must be available throughout the world. One radio option is to use the Industrial, Scientific and Medical (ISM) bands, which are available without licenses in most countries. Some of the types of frequencies that may be used are described in the International Frequency Table contained in Article S5 of the Radio Regulations (Volume 1). Some of these frequencies are already used in wireless telephony and wireless local area networks (WLANs). For small size and low cost sensor networks, a signal amplifier is not required. According to , hardware limitations and trade-offs between antenna efficiency and energy consumption impose certain restrictions on the choice of transmission frequency in the ultrahigh frequency range. They also offer 433 MHz ISM in Europe and 915 MHz ISM in North America. Possible transmitter models for these two zones are discussed in. The main advantages of using ISM radio frequencies are the wide range of frequencies and worldwide availability. They are not tied to a specific standard, thereby providing greater freedom to implement energy-saving strategies in sensor networks. On the other hand, there are various rules and restrictions, such as different laws and interference from existing applications. These frequency bands are also called unregulated frequencies. Most of today's node equipment is based on the use of radio transmitters. The IAMPS wireless nodes described in , use Bluetooth-compatible 2.4 GHz transmitters and have an integrated frequency synthesizer. The design of low-power nodes is described in the work; they use one radio transmission channel, which operates at a frequency of 916 MHz. The WINS architecture also uses radio communications. Another possible method of communication in sensor networks is infrared. Infrared communication is available without a license and is protected from interference from electrical devices. IR transmitters are cheaper and easier to manufacture. Many of today's laptops, PDAs and mobile phones use an IR interface to transmit data. The main disadvantage of such communication is the requirement of direct visibility between the sender and the recipient. This makes IR communication undesirable for use in sensor networks due to the transmission medium. An interesting transmission method is used by smart nodes, which are modules for automatic monitoring and data processing. They use an optical medium for transmission. There are two transmission schemes, passive using a corner-cube retroreflector (CCR) and active using a laser diode and controlled mirrors (discussed in). In the first case, an integrated light source is not required; a configuration of three mirrors (CCR) is used to transmit the signal. The active method uses a laser diode and an active laser communication system to send light beams to the intended receiver. The unusual requirements for sensor network applications make the choice of transmission media challenging. For example, marine applications require the use of aquatic transmission media. Here you need to use long-wave radiation that can penetrate the surface of water. In difficult terrain or on the battlefield, errors and more interference may occur. In addition, it may turn out that the antennas of the nodes do not have the required height and radiation power for communication with other devices. Therefore, the choice of transmission medium must be accompanied by reliable modulation and coding schemes, which depend on the characteristics of the transmission channel.

3.8. Power consumption

A wireless node, being a microelectronic device, can only be equipped with a limited power supply (

3.8.1. Connection

A node spends maximum energy on communication, which involves both transmitting and receiving data. It can be said that for short distance communications with low radiation power, transmission and reception require approximately the same amount of energy. Frequency synthesizers, voltage control oscillators, phase-locking (PLL) oscillators, and power amplifiers all require energy, which has limited resources. It is important that in this case we do not only consider active power; we also consider electricity consumption when starting transmitters. Starting up the transmitter takes a fraction of a second, so it consumes a negligible amount of energy. This value can be compared to the PLL blocking time. However, as the transmitted packet decreases, the startup power begins to dominate the energy consumption. As a result, it is ineffective to constantly turn the transmitter on and off, because Most of the energy will go towards this. Currently, low power radio transmitters have standard Pt and Pr values ​​of 20 dBm and Pout close to 0 dBm. Note that PicoRadio directed to Pc is -20 dBm. The design of small-sized, low-cost transmitters is discussed in the source. Based on their results, the authors of this paper, taking into account budget and energy consumption estimates, believe that the values ​​of Pt and Pr should be at least an order of magnitude smaller than the values ​​​​given above.

3.8.2. Data processing

Energy consumption during data processing is significantly less compared to data transmission. The example described in the work actually illustrates this discrepancy. Based on Rayleigh's theory that a quarter of the power is lost during transmission, it can be concluded that the energy consumption of transmitting 1 KB over a distance of 100 m will be approximately the same as executing 3 million instructions at a speed of 100 million instructions per second (MIPS )/W processor. Therefore, local data processing is critical to minimize energy consumption in a multi-hop sensor network. Therefore, nodes must have built-in computing capabilities and be able to interact with the environment. Cost and size limitations will lead us to select semiconductors (CMOS) as the core technology for microprocessors. Unfortunately, they have limitations on energy efficiency. CMOS requires power every time it changes state. The energy required to change states is proportional to the switching frequency, capacitance (depending on area) and voltage fluctuations. Therefore, reducing the supply voltage is an effective means of reducing power consumption in the active state. Dynamic voltage scaling, discussed in , seeks to adapt processor power and frequency according to the workload. When the computational load on the microprocessor is reduced, simply reducing the frequency gives a linear reduction in energy consumption, however, reducing the operating voltage gives us a quadratic reduction in energy consumption. On the other hand, all possible processor performance will not be used. This will work if we take into account that peak performance is not always required and therefore the operating voltage and frequency of the processor can be dynamically adapted to the processing requirements. The authors propose workload prediction schemes based on adaptive processing of existing load profiles and on the analysis of several already created schemes. Other strategies for reducing processor power are discussed in . It should be noted that additional circuits may be used to encode and decode the data. Integrated circuits may also be used in some cases. In all these scenarios, the sensor network structure, operating algorithms and protocols depend on the corresponding energy consumption.

4. Sensor network architecture

The nodes are usually located randomly throughout the observation area. Each of them can collect data and knows the route for transmitting data back to the central node, the end user. Data is transmitted using multi-hop network architecture. The central node can communicate with the task manager via the Internet or satellite. The protocol stack used by the central node and all other nodes is shown in Fig. 3. The protocol stack includes power information and routing information, contains information about network protocols, helps communicate efficiently over the wireless medium, and facilitates node collaboration. The protocol stack consists of the application layer, transport layer, network layer, data link layer, physical layer, power management layer, mobility management layer, and task scheduling layer. Depending on the data collection tasks, different types of application software can be built at the application level. The transport layer helps maintain data flow if required. The network layer provides routing of data provided by the transport layer. Because the environment is noisy and nodes may move, the MAC protocol must minimize the occurrence of collisions when transmitting data between neighboring nodes. The physical layer is responsible for the ability to transfer information. These protocols help nodes perform tasks while saving energy. The power management layer determines how the node should use energy. For example, a node may turn off the receiver after receiving a message from one of its neighbors. This will help you avoid receiving a duplicate message. In addition, when a node has low battery power, it informs its neighbors that it cannot participate in message routing. It will use all the remaining energy to collect data. The mobility control (MAC) layer detects and records the movement of nodes, so there is always a route for data to travel to the central node and nodes can determine their neighbors. And knowing its neighbors, a node can balance energy consumption by working together with them. The task manager plans and schedules information collection for each region separately. Not all nodes in the same region are needed to perform sensing tasks at the same time. As a result, some nodes perform more tasks than others, depending on their power. All these layers and modules are necessary for the nodes to work together and strive for maximum energy efficiency, optimize the data transmission route in the network, and also share each other’s resources. Without them, each node will work individually. From the point of view of the entire sensor network, it is more efficient if the nodes work together with each other, which helps to extend the life of the network itself. Before discussing the need to include modules and control layers in the protocol, we will review three existing works on the protocol stack, which is shown in Figure 3. The WINS model, discussed in the source, in which nodes are connected in a distributed network and have access to the Internet. Since a large number of WINS network nodes are located at a short distance from each other, multi-hop communications reduce energy consumption to a minimum. The environmental information received by the node is sequentially forwarded to the central node or WINS gateway through other nodes as shown in Figure 2 for nodes A, B, C, D and E. The WINS gateway communicates with the user through normal network protocols such as the Internet . The WINS network protocol stack consists of an application layer, a network layer, a MAC layer, and a physical layer. Smart nodes (or motes). These nodes can be attached to objects or even float in the air due to their small size and weight. They use MEMS technology for optical communication and data acquisition. Motes may have solar panels for recharging during the day. They require line of sight to communicate with the optical transmitter base station or other speck. Comparing the architecture of the mote network with that presented in Figure 2, we can say that smart nodes typically communicate directly with the base station transmitter, but one-to-one communication is also possible. Another approach to developing protocols and algorithms for sensor networks is driven by the requirements of the physical layer. Protocols and algorithms must be designed in accordance with the choice of physical components, such as the type of microprocessors, and the type of receivers. This bottom-up approach is used in the IAMPS model and also considers the dependence of the application layer, network layer, MAC layer, and physical layer on the host hardware. The IAMPS nodes interact with the end user in the same way as in the architecture shown in Figure 2. Various schemes, such as time division channel (TDMA) or frequency division channel (FDMA) and binary modulation or M-modulation are compared in the source. The bottom-up approach means that the node's algorithms must know the hardware and use the capabilities of microprocessors and transmitters to minimize power consumption. This may lead to the development of different assembly designs. And different node designs will lead to different types of sensor networks. Which in turn will lead to the development of various algorithms for their operation.

Literature

1. G.D. Abowd, J.P.G. Sterbenz, Final report on the interaction workshop on research issues for smart environments, IEEE Personal Communications (October 2000) 36–40.
2. J. Agre, L. Clare, An integrated architecture for cooperative sensing networks, IEEE Computer Magazine (May 2000) 106–108.
3. I.F. Akyildiz, W. Su, A power aware enhanced routing (PAER) protocol for sensor networks, Georgia Tech Technical Report, January 2002, submitted for publication.
4. A. Bakre, B.R. Badrinath, I-TCP: indirect TCP for mobile hosts, Proceedings of the 15th International Conference on Distributed Computing Systems, Vancouver, BC, May 1995, pp. 136–143.
5. P. Bauer, M. Sichitiu, R. Istepanian, K. Premaratne, The mobile patient: wireless distributed sensor networks for patient monitoring and care, Proceedings 2000 IEEE EMBS International Conference on Information Technology Applications in Biomedicine, 2000, pp. 17–21.
6. M. Bhardwaj, T. Garnett, A.P. Chandrakasan, Upper bounds on the lifetime of sensor networks, IEEE International Conference on Communications ICC’01, Helsinki, Finland, June 2001.
7. P. Bonnet, J. Gehrke, P. Seshadri, Querying the physical world, IEEE Personal Communications (October 2000) 10–15.