WIRELESS SENSOR NETWORKS BASED ON EMBEDDED SYSTEMS

WIRELESS SENSOR NETWORKS BASED ON EMBEDDED SYSTEMS

Peter Safir

Bachelor of Science, The Azrieli College of Engineering in Jerusalem (JCE),

Israel, Jerusalem

ABSTRACT

This paper describes the concept of wireless sensor networks. Applications of wireless sensor networks. The topology of a wireless sensor network such as a wireless ad hoc network. What protocols and operating systems are used in wireless sensor networks for transmitting data between the sensor and the embedded system. Description of the ZigBee standard and the TinyOS and LiteOS operating systems.

Keywords: wireless ad hoc network, wireless sensor networks, TinyOS, LiteOS, ZigBee, Zeroconf.

INTRODUCTION

With the development of new wireless technology standards such as Bluetooth and Wi-fi [1] and progress in microprocessor manufacturing and availability, many embedded system developers have begun to develop and integrate wireless sensor networks under the control of various embedded systems into their designs. Sensor networks are widely used in agriculture, medicine, winetillation system control, security systems, and many other industries requiring high mobility or to collect data from large or difficult to access areas.

The concept of wireless sensor networks

Wireless sensor networks consist of a control board and the sensors (motes) which remotely transmit data to the control board. The embedded system makes the first processing of received data and transmits the data through the Internet to the cloud store, where the data is analyzed and visualized in graphs or any other form convenient for the user. The board with the sensor is pretty miniaturized and contains a microcontroller, memory, ADC, DAC, a battery or external power supply, a RF receiver and transmitter, the sensor, and a number of connectors for contact control and correction of the sensor. The choice of the sensor depends on the functions to be carried out. It can be temperature change or pollution level of the room or any other sensor. There are a gigantic number of sensors used in wireless sensor networks.

Figure 1. Wireless sensor network topology

The most common topology for building a wireless sensor network is the wireless ad hoc network. The topology of this network has no permanent structure and no centralized control. Sensors are randomly connected to each other and thus form a wireless network. Each client of the wireless sensor network transmits data intended for the other clients of the network. With this topology, the sensor dynamically determines who to forward the data to, based on the connectivity of the network. This is in general different from wired networks or other centrally managed wireless networks. Centralized networks require expensive hardware such as routers or access points. A wireless ad hoc network does not require any network administration since it is operated by its own sensors. In a wireless ad hoc network, it is enough to connect a few access points to access the network and all the traffic will be sent by the sensors closest to the sensor which sends the data. The sensors connected to the network have their own range of coverage depending on the transmitter installed on their board. If the sensor is located at one end of the network and has to transmit data to the middle of the network, but its transmitter is not powerful enough to do so, it will pass through several sensors, following a prearranged path to the required device. In case of a failure of some sensors, the sensor network will continue to work after reconfiguration, but the number of sensors will be reduced. This topology of the network makes it possible to increase the range of the network by attaching new sensors. The capacity of each transmitter in this case should not be large and this reduces the cost of the sensor and the electromagnetic radiation and this improves the indicators of the impact of electromagnetic pollution on the environment.

Figure 2. Wireless ad hoc network

There are no standards for the use of software in wireless sensor networks. There are many different data processing protocols and communication protocols. It is possible to install a special operating system on each sensor for the correct operation of the network. Many engineers use the most popular operating systems TinyOS or LiteOS [2] for their sensor networks. TinyOS is an open source operating system developed at Berkeley University. This operating system allows real-time data processing and very limited computing resources. One of the most important functions of TinyOS is the formation of a sensor network of the required topology and the control and reconfiguration of the whole network in case of sensor failure. LiteOS is Huawei's operating system for the Internet of Things segment, which takes up only 10 KB of memory. LiteOS supports Zeroconf, a technology that allows you to update available networks and automatically configure access to network connections. This makes it possible to use LiteOS for any kind of sensor. LiteOS is also categorized as open source.  Because of the limited battery life usually supplied by wireless sensors, only primary data processing is done on the sensors, to reduce the data being transmitted. There are also specific communication protocols for the wireless sensor network. One of the best known is the ZigBee [3] communication protocol. The ZigBee [4] alliance was created in 2022. The alliance aims to standardize data transmission in wireless sensor networks. The ZigBee alliance includes the major electronics developers such as Samsung, Texas Instruments, IBM, Motorola, Philips, Ember and many others. ZigBee has developed a standard that is supported by a production of fully compatible hardware and software products. IEEE 802.15.4 [5] is the main standard for data transmission in wireless sensor networks which describes the physical layer and the level of network access for low power, low transmitter power but high fault tolerance devices. ZigBee [6] took this standard as the basis for its protocol stack. Let's list some of the features of ZigBee:

• ZigBee has a security policy as well as encryption of transmitted data in the network.
• ZegBee has a mechanism of self-recovery and guaranteed data delivery in case of failure of a single communication node or breakdown of communication between sensors.
• ZegBee standard sensors have low power consumption, the final devices can be set to sleep mode, this standard allows such devices to work for up to three years on AA batteries.
• ZigBee standard sensors are small and inexpensive.
• ZigBee indication of all sensors connected to the network and then generates a data routing scheme.
• ZigBee is a self-organizing network and is formed automatically by attaching new sensors to the network.
• The ZigBee standard transmits data in several frequency bands, but the maximum data transfer rate can be achieved in the range from 2.4 GHz to 2.48 GHz, where there are 16 channels at 5 MHz each [7].
• The disadvantages of ZigBee is a low data transfer rate of 250 kbit/s including service information, while the average data rate is from 5 to 40 kbit/s [8].

Devices based on ZigBee standards can be categorized into three classes:

• Coordinator - it controls the operation of the entire network, stores the architecture data of the entire wireless network and acts as a gateway to transmit data collected by the entire network.
• Router - its function is to receive and transmit data as well as route it.
• Sensor - it collects the data and transmits it to the nearest sensor.

CONCLUSION

Wireless sensor network allows solving many problems in many areas where data monitoring is required and it is difficult or impossible to collect the data using traditional cable networks. Let us highlight some of the advantages of the wireless sensor network:

• Easy addition of the sensors to an existing network without any cabling.
• Low cost of the sensors.
• High fault tolerance of the wireless sensor network.
• There are many off-the-shelf solutions with open source code.
• Low cost of installation and maintenance of the wireless sensor network.
• Data transfer to the cloud for visualization and processing.

References:

1. Shahin Farahani PhD.ZigBee Wireless Networks and Transceivers. Newnes 2008. P. 83 - 110.
2. Ata Elahi, Adam Gschwender. ZigBee Wireless Sensor and Control Network. 2009. P. 39 – 70.
3. Robert Faludi. Building Wireless Sensor Networks: with ZigBee, XBee, Arduino, and Processing. O'Reilly Media 2011. P. 88 – 105.
4. Houda Labiod, Hossam Afifi, Costantino de Santis. Wi-Fi™, Bluetooth™, Zigbee™ and WiMax™. Springer 2010. P. 45 – 56.
5. Gaotao Shi, Keqiu Li. Signal Interference in WiFi and ZigBee Networks (Wireless Networks). Springer 2016. P. 78 – 90.
6. Ibrahiem M. M. El Emary, S. Ramakrishnan. Wireless Sensor Networks: From Theory to Applications. CRC Press 2013. P. 113 – 133.
7. Ian F. Akyildiz , Mehmet Can Vuran. Wireless Sensor Networks. Wiley 2010. P. 78 – 97.
8. Anna Hac. Wireless Sensor Network Designs. Wiley 2003. P. 123 – 125.