14-08-2012, 12:25 PM
Wireless Sensor Network
1wireless sensor.docx (Size: 1.29 MB / Downloads: 88)
INTRODUCTION
Applying wireless sensor network (WSN) represents a relatively new approach in the field of wireless communication. The network consists of many low-cost sensor nodes and one or more base stations (BS) which gather the sensed data for further processing. WSN is a highly distributed network and the number of communicating sensor nodes may vary from units to thousands according to the area of interest and size of the environment. The sensors may monitor many attributes like temperature, humidity, light, leakage of chemical compounds or anything that can be sensed. WSN may be utilized in various fields like environmental protection, guarding of buildings, military purposes, factories.WSN requires new ad-hoc techniques allowing hop-by-hop communication, which requires development of novel communication protocols. Their distributiveness and limited resources both energy and computational have to be considered. Hence new protocols and security approaches are being developed rapidly and there are needs to examine their functionality. Since many sensor nodes are usually employed in the communication, simulations of WSN may facilitate the process of development. Examining of the protocols can be cheaper (the devices are still rather expensive), faster (the time of implementation and installation of the network) and can speed up the evaluation (if we need to examine the time to complete battery depletion and the behaviour of the remaining “alive” nodes after that).
SELECTED SIMULATORS FOR WSN
There is a wide range of simulators which can be used for simulating processes realized ina WSN. Some of them are known more than the others but may be too general and have some limitations for wireless networking with many participants. This is for example a case of ns-2 which has limitation in number of participating sensor nodes [1]. On the other hand, this could not be necessarily considered as the main aspect and crucial limitation for some kind of specific simulations. Adequate extensions may be often found for such simulators. Another issue in the area of WSN is the energy consumption constraint which is essential for proper processes simulations. It would not be correct to summarize over all the simulators and choose the best one because the evaluation strictly depends on the selected criteria and priorities.
MechanismFramework
Mechanismprovides twelve files which contain classes that extend corresponding native classes of ns-2. A class called SensorNode extends a class Mobile Node of ns-2 and specifies several sensor characteristics like power consumption for sensing and processing, the number of instructions executed per seconds by its microcontroller or states of different sensor parts. This class can be used as initial approach for different sensor node types development ABattery class extends the EnergyModel of ns-2. It can be used to implement different existing battery models. It provides methods like turning sensor node on and off, it can put it into a sleep mode and it can decrease energy when the sensor performs sensing, processing or disseminatingDataGenerator simulates the sensing task. Sensing interval may be adjusted and some classes for specialized sensing may extend this class. TemperatureDataGenerator and TemperatureAppData are examples of such extensions and add characteristics proper for temperature data and methods to work with the gathered data
Simulator Castalia
This section deals with the simulator OMNeT++ and especially with its extension Castalia, version 3.0. Castalia is based on OMNeT++ platform and is developed for networks of lowpower embedded devices such as wireless sensor nodes. The main declared advantages are advanced channel model based on empirically measured data and radio model based onreal radios for low-power communication and monitoring of the power consumption
Wireless Channel Model
The authors of Castalia claim that their simulator is the most realistic, among the others, concerning the wireless channel. It takes into account various important features which are discussed bellow. Castalia is a “tunable” simulator where many parameters may be adjustedusing input files to simulate real environment. Several examples of simulation applications are provided. The creation of resulting wireless channel model is rather complex and the description of the process follows.
WSN Modelling Using MiXiM
MiXiM enables to model both 2D and 3D environment. Apart from the communicating devices, several other objects like houses or walls to simulate radio propagation of signals can be placed in the environment. The positions of all nodes are managed by Connection- Manager. Multiple managers can be employed to enable different frequency ranges such as radio waves or ultra sound. Different kinds of nodes such as APs or terminals can be represented by nodes module Three modules are used to represent standard network layers: the application layer, the network layer, and theMAClayer which is actually grouped together with the physical layer into a Network Interface Card (NIC) module. A single node can have several NICs to model,for example, a laptop which contains Bluetooth and IEEE 802.11 (Wi-Fi).
Wireless Channel Model
Three following blocks are used to model the radio environment: a propagation model, a modulation model and an interference model. Furthermore two events are generated by WSNet.acarrier sense occurring at transmission start and a packet transmission occurring at transmission end. The carrier sense provides all the nodes information about frequency, code, modulation, SINR, reception power and BER (Bit Error Rate) of the signal. Depending on the user configuration when the packet reception event occurs, the packet is either dropped according to the function PER (Packet Error Rate) or some errors in the packet can be established according to the BER
Radio Model
The radio model of WSNet Provides following modulation techniques: none, step, BPSK (Binary-Phase Shift Keying), OQPSK (Offset quadrature phase-shift keying) and MQAM (Multi-Level Quadrature Amplitude Modulation). The model of the antenna is “omnidirectional”.There are several parameters to be configured: loss induced to the signal by the R/F antenna (in dB), default value is “0”; gain of the transmitter’s and receiver’s antenna, for both the default value is “0”; the orientation of the antenna – in the xy plane regarding to that xy plane, default value is “0” for both. Regarding the PHY layer, there is one very basic model available called “half1d”. When using this model, the transmission and reception cannot be provided simultaneously.
1wireless sensor.docx (Size: 1.29 MB / Downloads: 88)
INTRODUCTION
Applying wireless sensor network (WSN) represents a relatively new approach in the field of wireless communication. The network consists of many low-cost sensor nodes and one or more base stations (BS) which gather the sensed data for further processing. WSN is a highly distributed network and the number of communicating sensor nodes may vary from units to thousands according to the area of interest and size of the environment. The sensors may monitor many attributes like temperature, humidity, light, leakage of chemical compounds or anything that can be sensed. WSN may be utilized in various fields like environmental protection, guarding of buildings, military purposes, factories.WSN requires new ad-hoc techniques allowing hop-by-hop communication, which requires development of novel communication protocols. Their distributiveness and limited resources both energy and computational have to be considered. Hence new protocols and security approaches are being developed rapidly and there are needs to examine their functionality. Since many sensor nodes are usually employed in the communication, simulations of WSN may facilitate the process of development. Examining of the protocols can be cheaper (the devices are still rather expensive), faster (the time of implementation and installation of the network) and can speed up the evaluation (if we need to examine the time to complete battery depletion and the behaviour of the remaining “alive” nodes after that).
SELECTED SIMULATORS FOR WSN
There is a wide range of simulators which can be used for simulating processes realized ina WSN. Some of them are known more than the others but may be too general and have some limitations for wireless networking with many participants. This is for example a case of ns-2 which has limitation in number of participating sensor nodes [1]. On the other hand, this could not be necessarily considered as the main aspect and crucial limitation for some kind of specific simulations. Adequate extensions may be often found for such simulators. Another issue in the area of WSN is the energy consumption constraint which is essential for proper processes simulations. It would not be correct to summarize over all the simulators and choose the best one because the evaluation strictly depends on the selected criteria and priorities.
MechanismFramework
Mechanismprovides twelve files which contain classes that extend corresponding native classes of ns-2. A class called SensorNode extends a class Mobile Node of ns-2 and specifies several sensor characteristics like power consumption for sensing and processing, the number of instructions executed per seconds by its microcontroller or states of different sensor parts. This class can be used as initial approach for different sensor node types development ABattery class extends the EnergyModel of ns-2. It can be used to implement different existing battery models. It provides methods like turning sensor node on and off, it can put it into a sleep mode and it can decrease energy when the sensor performs sensing, processing or disseminatingDataGenerator simulates the sensing task. Sensing interval may be adjusted and some classes for specialized sensing may extend this class. TemperatureDataGenerator and TemperatureAppData are examples of such extensions and add characteristics proper for temperature data and methods to work with the gathered data
Simulator Castalia
This section deals with the simulator OMNeT++ and especially with its extension Castalia, version 3.0. Castalia is based on OMNeT++ platform and is developed for networks of lowpower embedded devices such as wireless sensor nodes. The main declared advantages are advanced channel model based on empirically measured data and radio model based onreal radios for low-power communication and monitoring of the power consumption
Wireless Channel Model
The authors of Castalia claim that their simulator is the most realistic, among the others, concerning the wireless channel. It takes into account various important features which are discussed bellow. Castalia is a “tunable” simulator where many parameters may be adjustedusing input files to simulate real environment. Several examples of simulation applications are provided. The creation of resulting wireless channel model is rather complex and the description of the process follows.
WSN Modelling Using MiXiM
MiXiM enables to model both 2D and 3D environment. Apart from the communicating devices, several other objects like houses or walls to simulate radio propagation of signals can be placed in the environment. The positions of all nodes are managed by Connection- Manager. Multiple managers can be employed to enable different frequency ranges such as radio waves or ultra sound. Different kinds of nodes such as APs or terminals can be represented by nodes module Three modules are used to represent standard network layers: the application layer, the network layer, and theMAClayer which is actually grouped together with the physical layer into a Network Interface Card (NIC) module. A single node can have several NICs to model,for example, a laptop which contains Bluetooth and IEEE 802.11 (Wi-Fi).
Wireless Channel Model
Three following blocks are used to model the radio environment: a propagation model, a modulation model and an interference model. Furthermore two events are generated by WSNet.acarrier sense occurring at transmission start and a packet transmission occurring at transmission end. The carrier sense provides all the nodes information about frequency, code, modulation, SINR, reception power and BER (Bit Error Rate) of the signal. Depending on the user configuration when the packet reception event occurs, the packet is either dropped according to the function PER (Packet Error Rate) or some errors in the packet can be established according to the BER
Radio Model
The radio model of WSNet Provides following modulation techniques: none, step, BPSK (Binary-Phase Shift Keying), OQPSK (Offset quadrature phase-shift keying) and MQAM (Multi-Level Quadrature Amplitude Modulation). The model of the antenna is “omnidirectional”.There are several parameters to be configured: loss induced to the signal by the R/F antenna (in dB), default value is “0”; gain of the transmitter’s and receiver’s antenna, for both the default value is “0”; the orientation of the antenna – in the xy plane regarding to that xy plane, default value is “0” for both. Regarding the PHY layer, there is one very basic model available called “half1d”. When using this model, the transmission and reception cannot be provided simultaneously.