Introduction to Wireless Sensor Networks PDF

Introduction to Wireless Sensor Networks Unit I / Overview of WSN Prepared By B S Bhatt 1  Syllabus / Unit -I Overview of WSN: Single-Node Architecture - Hardware Components - Network Characteristics - Unique cons...

Introduction to Wireless Sensor Networks Unit I / Overview of WSN Prepared By B S Bhatt 1  Syllabus / Unit -I Overview of WSN: Single-Node Architecture - Hardware Components - Network Characteristics - Unique constraints and challenges - Enabling Technologies for Wireless Sensor Networks- Types of wireless sensor networks. 2  Topic 1 Introduction to WSN 3  Introduction A Sensor is a device used to gather information about a physical process and translate into electrical signals that can be processed, measured and analyzed. The physical process can be any real-world information like temperature, pressure, light, sound, motion, position, flow, humidity, radiation etc. A Sensor Network is a structure consisting of sensors, computational units and communication elements for the purpose of recording, observing and reacting to an event or a phenomenon. The events can like physical world, an industrial environment, a biological system while the controlling or observing body can be a consumer application, government, civil, military, or an industrial entity. 4  Such Sensor Networks can be used for remote sensing, medical telemetry, surveillance, monitoring, data collection etc. 5 Wireless Sensor Networks A typical sensor network consists of sensors, controller and a communication system. If the communication system in a Sensor Network is implemented using a Wireless protocol, then the networks are known as Wireless Sensor Networks. 6  According to technologists, Wireless Sensor Networks is an important technology for the twenty first century. Recent developments in MEMS Sensors (Micro Electro Mechanical System) and Wireless Communication has enabled cheap, low power, tiny and smart sensors, deployed in a wide area and interconnected through wireless links for various civilian and military applications. A Wireless Sensor Network consists of Sensor Nodes deployed in large quantities and support sensing, data processing, embedded computing and connectivity. 7  Motivation for WSN The recent developments in engineering, communication and networking led to new sensor designs, information technologies and wireless systems. Such advanced sensors can be used as a bridge between the physical world and the digital world. Sensors are used in numerous devices, industries, machines and help in avoiding infrastructure failures, accidents, conserving natural resources, preserving wildlife, increase productivity, provide security etc. The use of distributed sensor network contributed by the technological advances in VLSI, MEMS and Wireless Communication. 8  With the help of modern semiconductor technology, powerful microprocessors can be developed, smaller in size when compared to the previous generation products. This miniaturization of processing, computing and sensing technologies led to tiny, low- power and cheap sensors, controllers and actuators. 9  Elements of WSN A typical wireless sensor network can be divided into two elements. They are: – Sensor Node – Network Architecture A Sensor Node in a WSN consists of four basic components. They are: – Power Supply – Sensor – Processing Unit – Communication System 10 Fig 2 / Basic Components of WSN 11  Elements of WSN (Cont) The sensor collects the analog data from the physical world and an ADC converts this data to digital data. The main processing unit a microprocessor or a microcontroller, performs an intelligent data processing and manipulation. Communication system consists of radio system, a short-range radio for data transmission and reception. As all the components are low-power devices, a small battery like CR-2032, is used to power the entire system. A Sensor Node consists of not only the sensing component but also other important features like processing, communication and storage units. 12  With and all these features, components enhancements, a Sensor Node is responsible for physical world data collection, network analysis, data correlation and fusion of data from other sensor with its own data. 13 Network Architecture When a large number of sensor nodes are deployed in a large area to monitor a physical environment, the networking of these sensor nodes is equally important. A sensor node in a WSN not only communicates with other sensor nodes but also with a Base Station (BS) using wireless communication. 14  The base station sends commands to the sensor nodes and the sensor node perform the task by collaborating with each other. The sensor nodes in turn send the data back to the base station. A base station also acts as a gateway to other networks through the internet. Afterreceivingthedatafromthesensornodes,abase station performs simple data processing and sends the updated information to the user using internet. Ifeachsensornodeisconnectedtothebasestation,it is known as Single-hop network architecture. Although long distance transmission is possible, the energy consumption for communication will be significantly higher than data collection and computation. 15 Fig 4 / Single Hop Architecture 16  Multi-hop Architecture Hence, Multi-hop network architecture is usually used. Instead of one single link between the sensor node and the base station, the data is transmitted through one or more intermediate node. 17  This can be implemented in two ways. Flat network architecture and Hierarchical network architecture. In flat architecture, the base station sends commands to all the sensor nodes but the sensor node with matching query will respond using its peer nodes via a multi-hop path. In hierarchical architecture, a group of sensor nodes are formed as a cluster and the sensor nodes transmit data to corresponding cluster heads. The cluster heads can then relay the data to the base station 18 Fig 6 / Flat and Hierarchical Network Architectures 19  Network Topologies in WSN A WSN can be either a single-hop network or a multi- hop network. The following are a few different network topologies that are used in WSNs. StarTopology In star topology, there is a single central node known as hub or switch and every node in the network is connected to this hub. Star topology is very easy to implement, design and expand. The data flows through the hub and plays an important role in the network and a failure in the hub can result in failure of entire network. 20  Tree Topology A tree topology is a hierarchical network where there is a single root node at the top and this node is connected to many nodes in the next level and continues. energy consumption The processing is highest power at the root and node and keeps on decreasing as we go down the hierarchical order. Mesh Topology In mesh topology, apart from transmitting its own data, each node also acts as a relay for transmitting data of other connected nodes. Mesh topologies are further divided into Fully Connected Mesh and Partially Connected Mesh. In fully connected mesh topology, each node is connected to every other node while in partially connected mesh topology, a node is connected one or more neighboring nodes. 21 Fig 7 / Network Topologies in WSN 22  Applications of WSN Air Traffic Control (ATC) Heating Ventilation and Air Conditioning (HVAC) Industrial Assembly Line Automotive Sensors Battlefield Management and Surveillance Biomedical Applications Bridge and Highway Monitoring Disaster Management Earthquake Detection Electricity Load Management 23  Environment Control and Monitoring Industrial Automation Inventory Management Personal Health Care Security Systems 24  Topic 2 Single Node Architecture – Hardware Components 25  Introduction Building a wireless sensor network requires the constituting nodes to be developed. These nodes have to meet the requirements from a given application. They have to be small, cheap, energy efficient, equipped with the right sensors, memory resources and sufficient communication facilities. The hardware components of the functioning node are explained as follows. 26  Overview of Sensor Node A basic sensor node comprises five main components are shown in the Figure. Controller: To process allrelevant data Memory: To store programs and intermediate data. Sensors and actuators: Actual interface to the physical world to observe or control physical parameters of the environment. Communication: Device for sending and receiving information over a wireless channel Power supply: Some form of batteries necessary to provide energy and some form of recharging by obtaining energy from the environment as well. 27 Fig 8 / Basic Components of a Sensor Node 28  Controllers The controller is the core of a wireless sensor node. It is the Central Processing Unit (CPU) of the node It collects data from sensors, processes this data, receives data from other sensor nodes, and decides on the actuator’s behavior. It has to execute various programs, ranging from time- critical signal processing and communication protocols to application programs. Such a variety of processing tasks can be performed on various controller architectures, representing trade-offs between flexibility, performance, energy efficiency, and costs. 29  Microcontrollers are suitable for WSNs since they can reduce their power consumption by going into sleep states where only parts of the controller are active. One of the main differences to general-purpose systems is that microcontroller-based systems do not include a memory management unit – for example, protected or virtual memory is difficult. In a wireless sensor node, DSP can be used to process incoming data. But the advantages of a DSP are not required in a WSN node and they are usually not used. Another option for the controller is to use Field- Programmable Gate Arrays (FPGAs) or Application- Specific Integrated Circuits (ASICs) instead of 30 microcontrollers.  An FPGA can be reprogrammed in the field to adapt to a changing set of requirements , but this can take time and energy. An ASIC is a specialized processor, designed for a given application such as high-speed routers and switches. The typical trade-off here is loss of flexibility in return for a considerably better energy efficiency and performance. 31  Memory There is a need for Random Access Memory (RAM) to store intermediate sensor readings, packets from other nodes etc. RAM is fast, but it loses its contents if power supply is interrupted. The program code can be stored in Read-Only Memory (ROM) or in Electrically Erasable Programmable Read- Only Memory (EEPROM) or flash memory. Flash memory can also serve as intermediate storage of data when the power supply goes off for some time. The long read and write access delays of flash memory should be taken into account as well as the high required energy. 32  Communication Module 1. Choice of transmission medium The first choice is the transmission medium and usual choices include radio frequencies, optical communication, and ultrasound. Radio Frequency (RF)- based communication is vital requirement of most WSN applications. It provides long range and high data rates, acceptable error rates at reasonable energy expenditure, and does not require line of sight between sender and receiver. For a practical wireless, RF-based system, the carrier frequency The wireless has tosensor be carefully networks chosen. use communication frequencies between about 433 MHz and 2.4 GHz. 33  2. Transceivers For actual communication, both a transmitter and a receiver are required in a sensor node to convert a bit stream coming from a microcontroller and convert them to and from radio waves. Such combined devices are called transceivers. Usually, since half-duplex operation is realized transmitting and receiving at the same time on a wireless medium is impractical in most cases. A range of low-cost transceivers is available that incorporate all the circuitry required for transmitting and receiving, modulation, demodulation, amplifiers, filters, mixers etc.. 34  3. Transceiver tasks and characteristics The following are the some of the important characteristics of a transceiver which should be taken into account. – Service to upper layer – Power Consumption and Energy Efficiency – Carrier Frequency & Multiple channels – Transmission Power Control – Data Rates – Modulation – Noise Figure – Power Efficiency – Frequency Stability etc 35  4. Transceiver States Transmit State: The transmit part of the transceiver is active and the antenna radiates energy. Receive State: The receive part is active. Idle State: A transceiver that is ready to receive but not currently receiving anything is said to be in an idle state. Sleep State: The significant parts of the transceiver are switched off. There are transceivers offering several different sleep states. 36  Sensors & Actuators Sensors can be categorized into the following three categories - 1. Passive Omni-directional sensors: They can measure a physical quantity at the point of the sensor node without manipulating the environment by active probing. They obtain the energy directly from the environment – energy is only needed to amplify their analog signal. There is no notion of “direction in these measurements. include thermometer, Typical light sensors, examples vibration, microphones, humidity, chemical sensors etc 37 2. Passive narrow-beam sensors: They are passive but have a well-defined notion of direction of measurement. A typical example is a camera, which can “take measurements” in a given direction, but has to be rotated if need be. 3. Active sensors: They probe the environment, for example, a sonar or radar sensor or some types of seismic sensors, which generate shock waves by small explosions. 38  Power Supply of Sensor Nodes 1. Traditional batteries The power source of a sensor node is a battery, either non-rechargeable (primary batteries) or, if an energy scavenging device is present on the node, also rechargeable (secondary batteries). In some form or other, batteries are electro-chemical stores for energy – the chemicals being the main determining factor of battery technology. 39  2. Energy scavenging Some of the unconventional energy sources like fuel cells, micro heat engines and radioactivity – convert energy from stored secondary form into electricity in a easy way than a normal battery would do. The entire energy supply is stored on the node itself – once the fuel supply is exhausted, the node fails. The energy from a node’s environment must be tapped into and made available to the node – energy scavenging should take place. 40 3. Photo-voltaics The solar cells can be used to power sensor nodes. The available power depends on whether nodes are used outdoors or indoors, and on time of day. The resulting power ranges between 10 mW/cm2 indoors and 15 mW/cm2 outdoors. Single cells achieve a fairly stable output voltage of about 0.6 V. Hence, solar cells are used to recharge secondary batteries. 4. Temperature gradients Differences in temperature can be directly converted to electrical energy. Theoretically, even small difference for example, 5 K can produce considerable power, but practical devices fall very short of theoretical upper limits. 41 5. Vibrations Walls or windows in buildings are resonating with cars or trucks passing in the streets, machinery often has low- frequency vibrations, ventilations also cause it, and so on. The available energy depends on amplitude and frequency of the vibration and ranges between 0.1 mW/cm3 and 10, 000 mW/cm3 for some extreme cases. 42 Challenges of WSN 51 Introduction To realize the characteristics requirements, the innovative mechanisms for a communication network have to be found. The find particular challenge is the need to mechanisms specific to the idiosyncrasies of a given application to support the specific quality of service, and maintainability requirements. These mechanisms also have to generalize to a wider range of applications and implementation of a WSN becomes necessary for every individual application. Someofthemechanismsthatwillformtypicalpartsof WSNs are: 52  1. Multi-hop Wireless Communication Since wireless communication is a core technique, a direct communication between a sender and a receiver is faced with limitations. In particular, communication over long distances is only possible using high transmission power. The use of intermediate nodes as relays can reduce the total required power. Hence, for many forms of WSNs, multi-hop communication will be a necessary ingredient. 53  Energy Efficient Operation & Auto-configuration 2. Energy-efficient Operation: It is a key technique for supporting long life time. The other options include energy-efficient data transport between two nodes or the energy-efficient determination of requested information. The non-homogeneous energy consumption – the forming of “hotspots” is an issue. 3. Auto-configuration: A WSN will have to configure most of its operational parameters, independent of external configuration. As an example, nodes should be able to determine their geographical positions only using other nodes of the network so- called “self- location”. The network should be able to tolerate failing nodes or to integrate new nodes. 54 4. Collaboration & In-network Processing In some applications, a single sensor is not able to decide whether an event has happened but several sensors have to collaborate to detect an event and only the joint data of many sensors provides enough information. Information is processed in the network in various forms to achieve this collaboration. This is opposite to having every node transmit all data to an external network and process it “at the edge” of the network. An example is to determine the highest or the average temperature within an area and to report that value to a sink. To solve such tasks, readings from individual sensors can be aggregated reducing the amount of data to be transmitted and hence improving the energy efficiency. 55  5. Data Centric Traditional communication networks are centered around the transfer of data between two specific devices, each equipped with one network address – the operation of such networks is thus address-centric. In a WSN, the nodes are deployed to protect against node failures or to compensate for the low quality of a single node’s actual sensing equipment. Hence, switching from an address-centric paradigm to a data- centric paradigm in designing architecture and communication protocols is promising. An example for such a data-centric interaction will be to request the average temperature in a given location area, as opposed to requiring temperature readings from individual nodes. 56  6. Locality The principle of locality will have to be embraced to ensure in particular, scalability. Nodes with limited should attempt to limit the state that they accumulate during protocol processing to only information about their direct neighbors. This will allow the network to scale to large numbers of nodes without having to depend on powerful processing at each single node. 57  7. Exploit Trade-offs Similar to locality principle, WSNs will have to depend to a large degree on exploiting various trade-offs between contradictory goals, both during system design and runtime. Examples for such trade-offs are - higher energy expenditure allows higher result accuracy, longer lifetime of the entire network trades off against lifetime of individual nodes and node density. If there is a depart from an address-centric view of the network, it may require new programming interfaces beyond the simple semantics of the conventional socket interface and allow concepts like required accuracy, energy/accuracy trade-offs etc. 58  Topic 5 Enabling Technologies for WSN 59  Introduction It has only become possible to build wireless sensor networks with some fundamental advances in enabling technologies. First and foremost among these technologies is the miniaturization of hardware. Smaller feature sizes in chips have driven down the power consumption of the basic components of a sensor node to a level that the constructions of WSNs can be contemplated. This is particularly relevant to microcontrollers and memory chips and the radio modems responsible for wireless communication. 60  Reduced chip size and improved energy efficiency is accompanied by reduced cost, which is necessary to make redundant deployment of nodes affordable. The actual sensing equipment is the third relevant technology next to processing and communication. However,itisdifficulttogeneralizebecauseofthevast range of possible sensors. 61 Fig 9 / Enabling Technologies for WSN 62  Energy Scavenging These three basic parts of a sensor node have to be accompanied by power supply. This requirement depends on application, high capacity batteries lasting for long times and can efficiently provide small amounts of current. A sensor node also has a device for energy scavenging, recharging the battery with energy gathered from the environment – solar cells or vibration-based power generation are conceivable options. Such a concept requires the battery to be efficiently chargeable with small amounts of current, which is not a standard ability. The counterpart to the basic hardware technologies is software. 63  The architecture of the operating system or runtime environment has to support simple re-tasking, cross- layer information exchange and modularity to allow for simple maintenance. This software architecture on a single node has to be extended to a network architecture, where the division of tasks between nodes is considered. The third part to solve is how to design appropriate communication protocols. Figure 9 shows the enabling technologies for WSN. 64  Topic 6 Types of Wireless Sensor Networks 65  Introduction The types of networks are decided based upon the environment so that they can be deployed underwater, underground, on land and so on. Different types of WSNs include: – Terrestrial WSNs – Underground WSNs – Underwater WSNs – Multimedia WSNs – Mobile WSNs 66  Terrestrial WSN’s Terrestrial WSNs are capable of communicating base stations efficiently and consist of hundreds to thousands of wireless sensor nodes deployed either in an unstructured or structured manner. In an unstructured mode, the sensor nodes are randomly distributed within the target area dropped from a fixed plane. The preplanned or structured mode considers optimal placement, grid placement, and 2D, 3D placement models. In this WSN, the battery power is limited but equipped with solar cells as a secondary power source. TheenergyconservationoftheseWSNsisachievedby using low duty cycle operations, minimizing delays, and optimal routing, and so on. 67 68  Underground WSN The underground wireless sensor networks are more expensive than the terrestrial WSNs in terms of deployment, maintenance, and equipment cost considerations and careful planning. The WSNs networks consist of several sensor nodes hidden in the ground to monitor underground conditions. To relay information from the sensor nodes to the base station, additional sink nodes are located above the ground. The underground wireless sensor networks deployed into the ground are difficult to recharge. 69  The sensor battery nodes equipped with limited battery power are difficult to recharge In addition to this, the underground environment makes wireless communication a challenge due to the high level of attenuation and signal loss. 70 Fig 10 / Underground WSN 71 Under Water WSN More than 70% of the earth is occupied with water. These networks consist of several sensor nodes and vehicles deployed underwater. Autonomous underwater vehicles are used for gathering data from these sensor nodes. A challenge of underwater communication is a long propagation delay, and bandwidth and sensor failures. Underwater, WSNs are equipped with a limited battery that cannot be recharged or replaced. The issue of energy conservation for underwater WSNs involves the development of underwater communication and networking techniques. 72 Fig 11 / Underwater WSN 73 Multimedia WSN Multimedia wireless sensor networks have been proposed to enable tracking and monitoring of events in the form of multimedia suchasimaging, video,andaudio. These networks consistof low-cost sensornodes equipped with microphones and cameras. These nodes are interconnected with each other over a wireless connection for data compression, data retrieval, and correlation. The challenges with the multimedia WSN include high energy consumption, high bandwidth requirements, data processing, and compressing techniques. In addition to this, multimedia contents require high bandwidth for the content to be delivered properly and easily. 74 Fig 12 / Multimedia WSN 75 Mobile WSN These networks consist of a collection of sensor nodes that can be moved on their own and can be interacted with the physical environment. The and mobile nodes can compute sense communicate. Mobile wireless sensor networks are much more versatile than static sensor networks. The advantages of MWSN over static wireless sensor networks include better and improved coverage, better energy efficiency, superior channel capacity, and so on. 76 77  Classification of WSN’s The classification of WSNs can be done based on the application but its characteristics mainly change based on the type. Generally, WSNs are classified into different categories like the following. – Static & Mobile – Deterministic & Nondeterministic – Single Base Station & Multi Base Station – Static Base Station & Mobile Base Station – Single-hop & Multi-hop WSN – Self Reconfigurable & Non-Self Configurable – Homogeneous & Heterogeneous 78  1. Static & Mobile WSN All the sensor nodes in several applications can be set without movement so these networks are static WSNs. Especially in some applications like biological systems uses mobile sensor nodes which are called mobile networks. The best example of a mobile network is the monitoring of animals. 2. Deterministic & Nondeterministic WSN In a deterministic type of network, the sensor node arrangement can be fixed and calculated. This sensor node’s pre-planned operation is possible in simply some applications. In most applications, the location of sensor nodes cannot be determined because of different factors like hostile operating conditions and harsh environment, so these networks are called non-deterministic. 79  3. Single Base Station & Multi Base Station In a single base station network, a single base station is used and it can be arranged very close to the region of the sensor node. The interaction between sensor nodes can be done through the base station. In a multi-base station type network, multiple base stations are used and a sensor node is used to move data toward the nearby base station. 4. Static Base Station & Mobile Base Station Base stations are either mobile or static similar to sensor nodes. The static type base station includes a stable position close to the sensing area whereas the mobile base station moves in the region of the sensor so that the sensor nodes load can be balanced. 80  5. Single-hop & Multi-hop WSN In a single-hop type network, the arrangement of sensor nodes can be done directly toward the base station whereas, in a multi-hop network, both the cluster heads and peer nodes are utilized to transmit the data to reduce the energy consumption. 6. Self Reconfigurable & Non-Self Configurable Inanon-selfconfigurablenetwork,thearrangementof sensor networks cannot be done by them within a network and depends on a control unit for gathering data. In wireless sensor networks, the sensor nodes maintain and organize the network and collaboratively work by using other sensor nodes to accomplish the task. 81  7. Homogeneous and Heterogeneous In a homogeneous wireless sensor network, all the sensor nodes mainly include similar energy utilization, storage capabilities and computational power. In heterogeneous network, some sensor nodes include high computational power as well as energy necessities as compared to others. The processing and communication tasks are separated consequently. 82 Model Question Bank 83  PART A 1. What is asensor? 2. What is asensor network? 3. Givetheelements ofWSN. 4. What arethe basic components of asensor node? 5. Differentiate between single hop and multi-hop networks. 6. Differentiate between flat and hierarchical network architectures. 7. What arethe various topologies used in WSN? 8. Giveanyfour applications of WSN. 9. How does wireless sensor network work? 10. What is the need for wireless sensor network? 84 11. Mention the challenges of wireless sensor networks. 12. What is event detection? 13. What is energy scavenging? 14. Differentiate between sensor and actuator. 15. What is quality of service? 16. List the types of WSN. 17. What is Multi-hop wireless communication? 18. What is data centric? 19. What is an active sensor? 20. State the deployment options. 85  PART B 1. Discuss briefly the various hardware components used in Single node architecture of WSN. 2. Explain the characteristics, constraints and challenges of WSN. 3. Write a short note on enabling technologies for wireless sensor networks. 4. Describe the types of wireless sensor networks in a brief manner. 86 Wireless Sensor Networks Architectures 1  Syllabus / Unit 2 Network Networks- Scenarios- Architecture- Design Sensor Principle, Physical Layer and Transceiver Design Considerations, Optimization Goals and Figures of Merit, Gateway Concepts, Operating Systems and Execution Environments- Introduction to TinyOS and nesC- Internet to WSN Communication. 2  Topic 1 Sensor Networks Scenario 3  Types of Sources & Sinks A source is any entity in the network that provide information typically a sensor node and also be an actuator operation. node that provides feedback about an A sink is the entity where information is required. There are three options for a sink - it can belong to the sensor network or just another sensor/actuator node or can be an entity outside this network. For the second case, the sink can be an actual device to interact with the sensor network or can also be a gateway to another larger network such as Internet. These main types of sinks are shown in Figure 2.1, showing sources and sinks in direct communication. 4 Fig 1 / Three Types of Sinks 5  Single Hop Vs Multiple Hop Networks Because of limited distance, the simple direct communication between source and sink is not possible in WSN, which are intended to cover a lot environmental or agriculture applications. To overcome such limited distances, the relay stations are used with the data packets taking multi hops from the source to the sink. The multi-hopping is a working solution to overcome problems with large distances and can also improve the energy efficiency of communication. It consumes less energy to use relays instead of direct communication. 6  The energy is actually wasted if intermediate relays are used for short distances and for large distance, the radiated energy dominate the fixed energy costs consumed in transmitter and receiver electronics. Moreover multi-hop networks operate in a store and forward fashion. 7 Fig 2 / MultihopNetwork 8  Multiple Sinks & Sources So far, only networks with a single source and a single sink have been explained. In many cases, there are multiple sources and/or multiple sinks present. In the most challenging case, multiple sources should send information to multiple sinks, where either all or some of the information has to reach all or some of the sinks. Figure 2.3 illustrates these combinations. 9 Types of Mobility One of the main virtues of wireless communication is its ability to support mobile participants. In wireless sensor networks, mobility can appear in three main forms: 1. Node mobility: The wireless sensor nodes can be mobile. The meaning of such mobility is highlyapplication dependent. In node mobility, the network has to reorganize itself frequently enough to be able to function correctly. There are trade- offs between the frequency and speed of node movement on one hand and the energy required to maintain a desired level of functionality in the network on the other hand. 10 2. Sink mobility: The information sinks can be mobile. The important aspect is the mobility of an information sink that is not part of the sensor network, for example, a human user requested information via a PDA while walking in an intelligent building. In a simple case, such a requester can interact with the WSN at one point and complete its interactions before moving on. In many cases, consecutive interactions can be treated as separate unrelated requests. 11 Fig 4 / Mobile Sink through Sensor Network 12  3. Event mobility: In applications like event detection and in tracking, the cause of the events or the objects to be tracked can be mobile. In such scenarios, the observed event is covered by a sufficient number of sensors at all time. Hence,sensorswillwakeuparoundtheobject,engaged in higher activity to observe the present object, and then go back to sleep. As the event source moves through the network, it is accompanied by an area of activity within the network. This is called as Frisbee Model as shown in Figure 2.4 13  Topic 2 Design Principles for WSN 14  Distributed Organization Both the scalability and the robustness optimization goal are required to organize the network in a distributed fashion. When organizing a network in a distributed fashion, it is necessary to know potential shortcomings of this approach. many cases, a centralized approach can produce In solutions that perform better or require fewer resources. One possibility is to use centralized principles in a localized fashion by electing, out of set of equal nodes. Such elections result in a dynamic hierarchy. The election process should be repeated continuously until the elected node runs out of energy 15  In Network Processing Techniques 1. Aggregation: The simplest in-network processing technique is aggregation. The term aggregation means that information is aggregated into a condensed form in nodes intermediate between sources and sinks out of information provided by nodes further away from the sink. The aggregation function must be applied in the intermediate nodes as shown in Figure 2.5. 16 Fig 5 / Aggregation as an Example 17  2. Distributed Source Coding and Distributed Compression: The objective is to encode the information provided by several sensors by using traditional coding schemes, which may be complex for simple sensor nodes. The readings of adjacent sensors are going to be quite similar and correlated. Suchcorrelationcanbeexploitedinsteadofsendingthe sum of the data so that the overhead can be reduced. 18  3. Distributed and collaborative signal processing When complex computations on a certain amount of data is to be done, it can be more energy efficient to compute these functions on the sensor nodes using Fast Fourier Transform (FFT). In principle, this is similar to algorithm design for parallel computers. However the energy and consumption of communication computation are relevant parameters to decide between various algorithms. 4. Mobile code/Agent-based networking The idea of mobile code is to have a small, compact representation of program code to be sent from node to node. This code is executed locally for collecting measurements and then decides where to be sent next. This idea has been used in various environments 19  Adaptive Fidelity & Accuracy The idea of making fidelity of computation depends upon the amount of energy available for that particular computation. This concept can be extended from a single node to an entire network. As an example, consider a function approximation application. When more sensors participate in the approximation, the function is sampled at more points and the approximation is better. But more energy has to be invested. Hence, it is up to an application to define the degree of accuracy of the results and the task of the communication protocols to achieve this accuracy. 20  Data Eccentricity In traditional communication networks, the focus will be on the pair of communicating peers, the sender and the receiver of data. In a wireless sensor network, the interest of an application is actual information reported about the physical environment. This is applicable when a WSN is redundantly deployed such that any given event can be reported by multiple nodes. This method of concentrating on the data rather than identity of nodes is called data-centric networking. For an application, this means that an interface is exposed by the network where data only is addressed in requests. 21  Exploit Local Information Another location information useful technique in the communication is to exploit protocols when-ever such information is present. Since the location of an event is crucial information for many applications, mechanisms must be available to determine the location of sensor nodes. Itof communication can simplify the protocols designandandcan operation improve their energy efficiency. 22  Exploit Activity Patterns Activity patterns in a wireless sensor network are quite different from that of traditional networks. The data rate averaged over a long time can be very small. This can be detected by a larger number of sensors, breaking into a frenzy of activity, causing a well-known event shower effect. Hence, the protocol design should be able to handle such bursts of traffic by switching between modes of quiescence and of high activity. 23  Exploit Heterogeneity Sensor nodes can be heterogeneous by constructions, that is, they have larger batteries, farther-reaching communication devices, or more processing power. They can also be heterogeneous by evolution, that is, they started from an equal state, but scavenge energy from the environment due to overloading. Heterogeneity in the network is both a burden and an opportunity. The opportunity is an asymmetric assignment of tasks, giving nodes with more resources or more capabilities the more demanding tasks. The burden is asymmetric task assignments cannot be static but have to be reevaluated. 24 Component Based Protocol Stacks The concept is a collection of components which can form a basic “toolbox” of protocols and algorithms to build upon. All wireless sensor networks will require some form of physical, MAC, Link layer protocols, routing and transport layer functionalities. Moreover, “helper modules” like time synchronization, topology control can be useful. On top of these basic components, more abstract functionalities can then be built. The set of components active on a sensor node can be complex and will change from application to application. Protocol components will also interact with each other either by using simple exchange of data packets or by exchange of cross-layer information. 25  Topic 3 Physical Layer and Transceiver Considerations 26  Introduction Some of the crucial points influencing the Physical Layer design in wireless sensor networks are - –Low power consumption –Small transmit power and a small transmission range –Low duty cycle –Low data rates in the order of tens to hundreds kilobits per second –Low implementation complexity and costs –Low degree of mobility –Small form factor for the overall node 27 Energy Usage Profile The choice of a small transmission power leads to an energy consumption profile different from other wireless devices like cell phones. The radiated energy is small and the overall transceiver consumes much more energy than actually radiated. Then for small transmit powers, transmit and receive modes consume more or less the same power depending on the transceiver architecture. To reduce average power consumption in a low-traffic wireless sensor network, the transceiver must go into sleep state instead of just idling. During this startup time, no transmission or reception of data is possible. 28  The third key observation is the relative costs of communications versus computation in a sensor node. A comparison of these costs depends for the communication part on BER requirements, range, transceiver type etc. 29  Choice of Modulation Scheme The following factors have to be balanced for the choice of modulation scheme - – Required data rate and symbol rate – Implementation complexity – Relationship between radiated power and target BER – Expected channel characteristics To maximize the time of transceiver in sleep mode, the transmit times should be minimized. The higher the data rate offered by a modulation, the smaller the time needed to transmit a given amount of data and the smaller the energy consumption. Moreover, the power consumption of a modulation scheme depends much more on the symbol rate than on the data rate. 30  Dynamic Modulation Scaling To adapt the modulation scheme to the current situation, an approach called dynamic modulation scaling is employed. For the case of m-ary QAM, a model has been developed with the symbol rate ‘B’ and the number of levels per symbol ‘m’ as parameters. This model expresses the energy required per bit and also the achieved delay per bit, taking into account the higher levels of modulation. Hence the bit delay decreases for increasing values of ‘B’ and ‘m’. The energy per bit depends much more on ‘m’ than on ‘B’. 31  For the particular parameters chosen, both energy per bit and delay per bit can be minimized for the maximum symbol rate. With modulation scaling, a packet is equipped with a delay constraint, from which directly a minimum required data rate can be derived. 32  Antenna Considerations The desired small form factor of the overall sensor nodes restricts the size and the number of antennas. If the antenna is much smaller than the carrier’s wavelength, it is difficult to achieve good antenna efficiency. In case of small sensor node cases, it will be difficult to place two antennas with suitable distance to achieve receive diversity. The antennas should be spaced apart at least 40–50% of the wavelength used to achieve good effects from diversity. 33  Inaddition,radiowavesemittedfromanantennaclose to the ground are faced with higher path-loss coefficients than the common value α = 2 for free-space communication. Moreover,dependingontheapplication,antennasmust not protrude from the casing of a node to avoid possible damage to it. These restrictions limit the quality and characteristics of an antenna for wireless sensor nodes. 34  Topic 4 Optimization of Goals & Figure of Merit 35 Introduction The following techniques will optimize a network, compare solutions, decide a better approach for a given application, and turn optimization goals into measurable figures of merit. 36  Quality of Service WSNs differ from other conventional communication networks in the type of services they offer. These networks only move bits from one place to another. Such QoS can be regarded as a low-level, networking- device attributes like bandwidth, delay, jitter or as a high-level, user attributes like perceived quality of a voice communication or a video transmission. But high-level QoS attributes in WSN highly depend on the application. Some generic possibilities are: 37 1. Event detection/reporting probability The probability of an event that actually occurred is not detected or not reported to an information sink 2. Event classification error If events are to be both detected and classified, the error in classification must be small. 3. Event detectiondelay The delay between detecting an event and reporting to all interested sinks 4. Missing reports The probability of undelivered reports should be small in periodic reporting applications. 38 5. Approximation accuracy For function approximation applications, the average/maximum absolute error with respect to the actual function. 6. Trackingaccuracy In Tracking applications, the reported position should be as close to the real position and the error should be small. 39  Energy Efficiency The most commonly considered aspects of energy efficiency are: 1. Energy per correctly received bit The average amount of energy to transport one bit of information from the source to the destination. 2. Energy per reported event The average energy spent to report one event 3. Delay/energy trade-offs The notion of “urgent” events to justify the increased energy investment for a speedy reporting of events. 40 4. Network lifetime The time for which the network is operational to fulfill its tasks starting from a given amount of stored energy. – Time to first node death: First node in the network run out of energy and stop operating – Network half-life: When 50% of the nodes run out of energy and stopped operating. – Time to partition: First partition of the network in two or more disconnected parts occur – Time to Loss of Coverage: For the first time any spot in the deployment region is no longer covered by any node’s observations. – Time to failure of first event notification: The unreachable part of the network does not want to report any events in the first place. 41  Scalability The ability to maintain performance characteristics irrespective of the size of the network is called scalability. Scalability requires consistent state such as addresses or routing table entries to be maintained. Hence, the need to restrict such information is enforced with the resource limitations of sensor nodes with respect to memory. The need for extreme scalability has direct consequences on the protocol design. implement appropriate scalability Architectures support rather and protocols than trying to be should as scalable as possible. 42  Robustness Related to QoS and scalability requirements, wireless sensor networks should also exhibit an appropriate robustness. They should not fail just because a limited number of nodes run out of energy, or because their environment changes. These failures have to be compensated by finding other routes. A precise evaluation of robustness is difficult in practice and depends mostly on failure models for both nodes and communication links 43  Topic 5 Gateway Concepts 44  Need for Gateways For practical deployment, the sensor network has to interact with other information devices. The standard example is to read the temperature sensors in one’s home while traveling. Figure 2.6 shows the networking scenario. The WSN has to exchange data with such a mobile device or with some sort of gateway which provides the physical connection to the Internet. The first option is to regard a gateway as a simple router between Internet and sensor network. This will entail the use of Internet protocols within the sensor network. The next option is to design the gateway as an actual application-level gateway on the basis of the application-level information. 45 Fig 6 / WSN with Gateway Node 46  WSN to Internet Communication For example, a sensor node wants to deliver an alarm message to some Internet host. The first problem to solve is to find the gateway from within the network. If several gateways are available, the selection of the particular route and gateway for a given destination have to be done. To handle several gateways the option is to build an IP overlay network on top of the sensor network. Figure 2.7 shows the mapping of Alice to a concrete IP address. The sensor node has to include sufficient information such as IP address and port number in its own packets. The gateway in turn will extract this information and translate it into IP packets. 47 Fig 7 / WSN to Internet Communication 48  Topic 6 Operating Systems & Execution Environment 53 Embedded Operating Systems The traditional tasks of an operating system are controlling and protecting the access to resources, managing their allocation to users and support for concurrent execution of processes. These tasks are only partially required in an embedded system and these systems do not have required resources to support a full-blown operating system. In particular, the need for energy-efficient execution requires support for energy management or Dynamic Voltage Scaling (DVS) techniques. Also, external components like sensors, the radio modem, or timers should be handled easily and efficiently. All this requires an appropriate programming model to structure a protocol stack and explicit support for energy management. 54  Programming Paradigms 1. Concurrent Programming The support for concurrent execution is crucial for WSN nodes to handle data coming from arbitrary sources like multiple sensors or the radio transceiver at arbitrary points in time. For example, a system can poll a sensor to decide whether data is available and process the data, then poll the transceiver to check whether a packet is available and then immediately process the packet and so on. 55 2.Process Based Concurrency Most general-purpose operating systems support concurrent execution of multiple processes on a single CPU. Hence such a process-based approach can be used to support concurrency in a sensor node as illustrated in (b) of Figure 2.10. Mapping such an execution model of concurrent processes to a sensor node shows that there are some granularity mismatches. This problem is severe for smaller tasks to be executed when compared to overhead. 56 Fig 10 / Programming Models for WSN 57 3. Event-based Programming The system waits for any event to happen, where an event can be the availability of data from a sensor, or arrival of a packet. Such an event is then handled by a short sequence of instructions that stores the occurrence of event and necessary information. This is called event based programming model as shown in Figure 2.11. This programming model distinguishes between two different “contexts”: - time-critical event handlers (execution cannot be interrupted) and for the processing of normal code (only triggered by the event handlers). 58 Fig 11 / Event Based Programming Model 59 4.Interfaces to Operating System In WSNs, the interfaces should be accessible from protocol implementations. This interface is closely tied with the structure of protocol stacks. For example Application Programming Interface (API) comprises, a “functional interface, object abstractions, and detailed behavioral semantics”. Abstractions are wireless links, nodes and so on. The inquiry,possible functions include state manipulation, transmitting of data, access to hardware and setting of policies. 60  Operating System & Protocols Stack In communication protocol structuring, the individual protocols are stacked on top of each other, each layer only using functions of the layer directly below. This layered approach has multiple benefits in keeping the entire protocol stack manageable. As an example, consider the use of information about the strength of the signal received from a communication partner. This physical layer information can be used to assist in networking protocols to decide about routing changes. Hence, one single source of information can be used by many other protocols not directly associated with the source of this information. Such cross-layer information exchange is one way to loosen the strict confinements of the layered approach. 61  Dynamic Energy & Power Management 1.Probabilistic State Transition Policies These policies regulate the transition between various sleep states. They start out by considering sensors randomly distributed over a fixed area and events arrive with certain temporal distributions and spatial distributions. This allows them to compute probabilities for the time to the next event, once an event has been processed. 62 2. Controlling Dynamic Voltage Scaling For example, only a single task has to be run in an operating system. Hence, a clever scheduler is required to decide exact clock rate to use in that situation to meet all deadlines. This can require feedback from applications for example, video playback in reference. 3. Trading off fidelity against energy consumption There are certain tasks that can be computed with a higher or lower level of accuracy. The fidelity achieved by such tasks is a candidate for trading off against other resources. In a WSN, the natural trade-off is against energy required to compute a task. 63