Software and hardware design of wireless sensor network platform based on embedded network

With the development of technology, there are more and more applications based on distributed, wireless sensor networks. This paper proposes a wireless sensor network experimental platform based on embedded systems, which has good application prospects in the algorithm and protocol verification of wireless sensor networks.

　　The development of integrated circuits, micro-electromechanical systems and communication theory has led to the emergence of wireless sensor networks. This type of wireless sensor network is composed of many self-powered sensor nodes. Each sensor node can collect data about the surrounding environment, perform simple calculations, and communicate with other nodes and the outside world. The multi-node characteristics of the sensor network enable a large number of sensors to work together to perform high-quality sensing and form a fault-tolerant acquisition system. It is precisely because of these advantages that many distributed wireless sensor network applications have emerged in recent years, such as disaster relief, smart homes, and detection and rescue of biological and chemical weapons attacks.

　　However, as an emerging technology, building a well-functioning and robust wireless sensor network still faces many challenges. And due to some of its unique characteristics, the design method of wireless sensor networks is very different from that of existing wireless networks. For example, due to the dense distribution of sensor nodes in sensor networks, a wide range of data management and processing technologies are required. Secondly, wireless sensor network nodes are generally deployed in areas that are difficult for humans to reach and contact, which makes the maintenance of sensor network nodes face great challenges. In addition, power consumption is also a very important issue. As tiny devices, wireless sensor nodes can only be equipped with limited power. In some applications, it is almost impossible to replace the power supply. This makes the life of sensor nodes largely dependent on the life of the battery, so reducing power consumption to extend the life of the system is the primary consideration in the design of wireless sensor networks. Many researchers in the field of wireless sensor networks are focusing on studying new power-saving protocols and algorithms, which require sensor network platforms for experiments and verification. In this article, we will introduce a wireless sensor network platform for experiments and verification of protocols and algorithms.

　　Wireless sensor network platform architecture

　　Generally speaking, a wireless sensor network includes sensor nodes and sensor network gateway nodes, as shown in Figure 1. The gateway node is used to combine the data obtained from each sensor node and is responsible for communicating with the outside world. The node is based on an embedded system.

　　The sensor node first collects environmental data such as sound, light and distance, and transmits the data to the gateway node after simple processing. Wireless sensor networks usually have two application modes: active polling mode and passive mode. The active mode requires the gateway node to actively poll each sensor node to obtain messages, while the passive mode requires the gateway node to respond promptly when an event occurs at a sensor node. The data obtained by each sensor node can also be combined, which greatly improves the efficiency of the sensor network. Of course, this also requires the sensor node to have a certain computing power.

　　System hardware implementation

　　In the system architecture introduced in this article, what needs to be implemented mainly are the hardware platforms of sensor nodes and gateways. The hardware implementation of these two platforms is introduced below.

　　1. Hardware implementation of sensor nodes

　　The function of the sensor node is to collect data that people are interested in and send the data to the gateway of each sensor node group. The sensor node is mainly composed of a power module, a computing module, a storage unit, a communication module and a sensing unit, as shown in Figure 2.

　　The functions of the computing unit of the sensor node have been introduced in the previous section. In our system, TI's 16-bit microcontroller MSP430F149 is used. MSP430 has rich on-chip storage resources. At an operating frequency of 5 MHz, the power consumption of MSP430 is about 1.5mW, and the microcontroller has a variety of power saving modes to choose from. In addition to rich on-chip storage resources and multiple power saving modes, MSP430 also has multiple AD interfaces and I/O data lines, making it easy to program with software. These interfaces can also be used as interfaces with sensor units.

　　The communication module of the sensor node is implemented by the nRF903 RF transceiver. The low power consumption and small size of the transceiver make it very suitable for use in wireless sensor network systems. The transceiver can operate in public frequency bands such as 433MHz, 868MHz, and 915MHz. The RF module communicates with the MSP430 through the serial port. The nRF903 can also determine the transmission power based on the input current. This feature gives it the following advantages:

　　a. A node can automatically adjust the number of neighboring nodes it communicates with, making the scale of the entire network adjustable.

　　b. It allows a node to use less energy when communicating with relatively nearby nodes.

　　c. Can be used to assist in conflict detection of wireless channels.

　　d. Can be used to determine the relative position of a node in the network.

　　Each sensor node is powered by AA batteries.

　　2. Hardware implementation of the gateway

　　The hardware part of the gateway is mainly composed of a central processing unit, a storage unit, a radio frequency transceiver module and a GPRS communication module, as shown in Figure 3.

　　The central processing unit of the gateway is mainly used to process the data collected from the sensor nodes and complete some control functions. The main device of the central processing unit is the AT91RM9200 microprocessor of Atmel. The AT91RM9200 is an ARM processor based on the ARM920T instruction set. The processor has a wealth of peripherals and interfaces, which enables it to complete some feature-rich applications under low-cost and low-power conditions. The AT91RM9200 processor integrates many peripheral interfaces, including USB2.0 interface and Ethernet interface. In addition, the processor also provides multiple communication interfaces that meet industrial standards, including audio, telecommunications, flash memory card, infrared, smart card interface, etc.

　　In order to transmit the collected data to the Internet, the gateway device is also equipped with a GPRS communication unit, which is mainly composed of the GM47 module of Sony Ericsson. This module transmits the data collected by the sensor to the Internet through the existing GPRS network of China Mobile. Users can observe the data collected by the sensor through ordinary PCs and GPRS mobile phone terminals. The gateway is also equipped with the same RF transceiver module as the sensor node, which is used to receive the data sent by the sensor node.

　　System software structure

　　In our wireless sensor network system, the software part is mainly on the gateway and sensor nodes. The main function of the gateway software is to process and manage the data transmitted by the sensor nodes. It is mainly composed of GPRS communication software, RF communication software, command line software and task management software, as shown in Figure 4.

　　Considering various requirements, we use the open source operating system - Linux. Linux is a networked operating system environment, especially suitable for network applications. Linux has a complete TCP/IP protocol stack, and supports a variety of other network protocols, such as the PPP protocol stack, making it easy to implement GPRS dial-up functions. Due to the open source nature of Linux, users can easily develop their own applications based on it.

　　The software on the sensor node is mainly developed using assembly and C languages. Its main function is to receive data from the sensor unit and send the data to the gateway of the sensor node group.

　　Conclusion

　　This paper introduces a wireless sensor network demonstration system for embedded systems. The whole system is based on embedded Linux and ARM processor, with the advantages of low power consumption and easy software development.

　　With the development of society and science and technology, wireless sensor networks will be increasingly widely used. At present, wireless sensor networks still have shortcomings in energy and node scale. As these problems are solved, wireless sensor networks will inevitably be more and more widely used in environmental monitoring, intelligent buildings, military and other fields.