Since human beings stepped into the information age, the information of nature comes from sensors. With the development of technology, people are not satisfied with the original single, independent sensor system. In many cases, we need to combine information from different regions to achieve a comprehensive judgment of the site conditions.

In the farm, we need to know the irrigation situation of crops and soil air quality to ensure the healthy growth of crops; in the mining area, we need to know the gas concentration, the location of the miners, the temperature and humidity of the underground mine, and the dust concentration to ensure the personal safety of the workers; In large buildings, we also need to understand the influence of environmental humidity, wind speed and its own aging degree at each location of the building, so as to maintain the structural health of the building in time. Typically, in these cases, the sensors used to collect data are placed thousands of meters apart and require continuous data detection over a period of months or even years, and staff cannot perform frequent maintenance. At this time, long-distance wiring, data aggregation, remote configuration of sensors, long-term system power supply and signal security are all issues that engineers need to consider.

With the development of wireless sensor network (Wireless Sensor Network) technology, these bottlenecks are resolved one by one. In the wireless sensor network, each node can independently collect and summarize the data. Depending on the scale of the application, the number of nodes can reach tens of thousands, monitoring various signals within a range of more than tens of square kilometers.

So, what exactly is a wireless sensor network?

Wireless Sensor Network Technology
　
A wireless sensor network is a sensor network composed of independently distributed nodes and gateways. Sensor nodes placed in different locations continuously collect physical information from the outside world, such as temperature, sound, and vibration. Communication between independent nodes is via wireless network. Each node of the wireless sensor network can realize the collection, simple processing of data, and can also receive data from other nodes, and finally send the data to the gateway. Engineers can get data from the gateway, view historical data records or perform analysis. Generally, the hardware structure of a typical wireless sensor network node includes: sensor interface, ADC, microprocessor, power supply and wireless transceiver.

Wireless sensor networks were born in the 1970s and were first used in US military-funded projects. After nearly 30 years of development, the application of wireless sensor networks has gradually turned to civilian use. It has appeared in the environmental monitoring of forests and rivers, in the intelligent application of the building environment, and in some industrial environments where wired sensors cannot be placed. figure. In 1999 and 2003, it was rated as the 20 most influential technologies of the 21st century and the top 10 new technologies that changed the world by American Business Weekly and MIT Technology Review.

As an application-specific technology, WSNs must be selected in a project with practicality in mind. To build a typical wireless sensor network, the following four important factors must be considered: network selection, topology, power consumption and compatibility.

Choice of wireless network
　
Wireless technology is the most critical part of wireless sensor networks. In typical applications of wireless sensor networks, the acquisition nodes are usually placed far apart, and sometimes the acquisition nodes located outdoors cannot be connected to the power grid. Therefore, when selecting a wireless network, bandwidth, transmission distance and power consumption are three factors. main consideration.

At the beginning of the wireless sensor network technology, the existing wireless protocols were difficult to meet the requirements of low power consumption, low cost and high fault tolerance. At this time, ZigBee technology came into being. Over the past 10 years of development, ZigBee has been proven to be the most suitable wireless technology for wireless sensor networks: it has a bandwidth of 250kbps, and the transmission distance can reach more than 1km. And the power consumption is smaller, the use of ordinary AA batteries can support the device to work continuously for up to several years.

The following figure lists the performance comparison of wireless protocols such as cellular network (Cellular), Bluetooth, Wi-FI and ZigBee. ZigBee protocol wins with its low power consumption and long-distance transmission capability, which is suitable for unattended operation for several years. Monitor applications.

Figure 1. Comparison of ZigBee and cellular network (Cellular), Bluetooth, Wi-FI and other wireless protocols

Network topology
For wireless sensor networks with more than tens of nodes, choosing a suitable network structure can not only expand the transmission distance of the network, but also ensure the stability of signal transmission.

Star is the simplest network topology, each node has a direct channel leading to the gateway, but its transmission distance is limited. Using a tree topology can solve this problem. After adding routing nodes, data on distant nodes can be transmitted to the gateway through routing nodes. However, the tree topology still has reliability problems. Once the routing node has problems, all the paths leading to the gateway from this node will be cut off. Therefore, for a wireless network with high reliability requirements, it is recommended to choose a mesh topology.

Figure 2, Network topology: star, tree and mesh topologies

Take the wireless sensor network of National Instruments (hereinafter referred to as NI) as an example. Every node in NI’s wireless sensor network can be configured as a routing node. Depending on the needs of the application, engineers can choose a tree topology or a mesh topology. As shown in the figure below, in the mesh topology, node 4 has two channels leading to the gateway. Once node 1 fails and is damaged, data can also be transmitted back to the gateway through node 2 to avoid data loss.

Figure 3. NI Wireless Sensor Network in Mesh Topology

System power consumption
Wireless sensor networks are usually placed outdoors and cannot perform long-distance wiring, which involves two problems, one is the transmission of signals, and the other is the power supply of the equipment. Signal transmission problems can be solved by choosing a wireless network; for equipment power problems, external power sources, such as batteries or small power generating devices, must be considered. Due to the limited power that the battery can supply, in order to meet the requirements of long-term use of the device, the energy consumption of wireless sensor network nodes must be strictly controlled. In the hardware structure of wireless sensor network nodes, wireless transceivers and microprocessors are large energy consumers. Therefore, users should choose ZigBee technology to ensure the low power consumption of wireless transceivers, and at the same time, under the premise of ensuring the performance of the processor, they should also choose a processor with sleep function and the lowest possible working energy consumption.

NI selected TI MSP430 MCU as the processor of wireless sensor network node. Its working power consumption is 8mW, and the power consumption during sleep is only 0.2 µW, which is only equivalent to the sleep power of an ordinary ADC chip. Powered by AA batteries, it can be guaranteed to work for three years, and even if it is placed in an inaccessible area, it can still maintain its position for a long time.

compatibility
Finally, we also need to consider the compatibility of the system. Wireless sensor networks can help engineers complete remote data collection, but we also need to consider functions such as data analysis, Display, and distribution through the Internet. For example, connect the wireless sensor network to the remote server to realize the recording and analysis of the massive data obtained by continuous collection; connect to the HMI (human-machine interface) to realize the display of historical data and real-time data; in addition, in some industrial In the application, it is more likely to connect the wireless sensor network to a variety of industrial field devices to work together. In these cases, the wireless sensor network must have good compatibility and realize fast connection with various field devices.

Common wireless sensor networks use special operating systems such as TinyOS or MANTIS, and need to use the nesC (network embedded systems C) development method. The system development engineer must learn a new development method to add wireless sensors to the existing system. network.

In order to help engineers reduce development time, NI uses a unified development platform LabVIEW to realize the development of wireless sensor networks, remote databases and human-computer interaction interfaces. Under the LabVIEW platform, engineers can not only realize remote configuration of wireless sensor networks, and implement algorithms on nodes; at the same time, they can also support all NI industrial platforms, such as automation controller CompactRIO, human-machine interface (HMI). In addition, LabVIEW also supports connection to third-party devices via OPC or various industrial buses. In order to meet the increasing demand for data sharing, LabVIEW also supports the network publishing function of data. Using Web Service, we can quickly publish data to the Internet, and engineers from all over the world can quickly access the data in the network database. Engineers do not need to learn development methods one by one for different equipment and different technologies, and can complete a complete remote monitoring system under a unified platform.

Figure 4. NI Wireless Sensor Network Compatibility

Application Analysis of Wireless Sensor Network
The application potential of wireless sensor networks in various fields is huge: in large buildings, wireless sensor networks can be quickly released to various floors and the inner and outer surfaces of the building, and the sub-item electricity, temperature and humidity, and water consumption can be obtained without considering the wiring problem. and other information, and transmit it back to the control room to realize the monitoring of building energy consumption. On sea-crossing bridges, wireless sensor networks can be placed on both sides of the bridge or at the bottom of the piers to capture data from piezoelectric sensors, acceleration sensors, and humidity sensors to calculate the degree of corrosion of the piers and the degree of deformation of the bridge deck. Ensure the safe use of bridges. In the intensive care unit, wireless sensor networks can be placed in bedrooms, bathrooms, kitchens and living rooms to distinguish the behavior and health status of the elderly through signals from light, temperature, and infrared sensors. In dangerous working environments such as coal mines, oil drilling, nuclear power plants, etc., wireless sensor networks can also replace staff to monitor the scene, eliminate dangerous situations in time, and ensure personnel safety.

The following article will analyze the application of NI wireless sensor network in some hot areas:

Building internal monitoring:
In general, monitoring systems inside buildings need to be erected at the same time as construction. Adding such systems to buildings that are already in use is very complicated. Wireless sensor networks can help engineers simplify the erection process. The University of California, Los Angeles (UCLA) used the NI wireless sensor network to set up an occupancy monitoring system in a parking lot. The monitoring system can monitor whether there is traffic passing by at each entrance and exit, so as to count the vacant parking spaces in different positions of the parking lot. The data is finally transmitted to the gateway through the ZigBee network and aggregated into the NI CompactRIO processor. After processing, the data can also be uploaded to UCLA’s transportation network database for authorized users to view the information.

Figure 5. NI wireless sensor network placed at the entrance of the parking lot

Green energy and environmental monitoring:
Solix is ​​a well-known bioenergy company. In their paddy fields, equipment works around the clock, converting algae into ethanol or diesel fuel, providing clean energy for planes, cars and factory heating systems. In order to ensure that the algae grow in a suitable environment, they deployed the NI wireless sensor network in dozens of acres of paddy fields, and sent back the pH value information of the paddy field in real time to make timely adjustments.

Factory Process Monitoring:
　
In the existing industrial production line, it is usually necessary to realize the monitoring of various information. For example, for a beverage production line, it is necessary to monitor the water flow information of the water inlet and outlet in real time to ensure the effective use of water resources. The feature of NI wireless sensor network without wiring is that engineers can quickly set up monitoring systems on the molding production line, and can obtain various data on the production line in time without interrupting the operation of the production line when installing the system.

summary
As an emerging technology, wireless sensor network has a wide space for development. Using wireless network can realize flexible, reliable and secure data collection. It is believed that with the further improvement of technology, wireless sensor networks will be used in all aspects of life and work to achieve seamless communication between humans and nature.