“With the progress of my country’s science and technology and economy, the automobile industry has entered a stage of rapid development in recent years. According to statistics, in 2013, the number of cars in my country reached 137 million, 5.7 times the number of cars in 2003, accounting for 54.9% of all motor vehicles, an increase of 29.9% over 10 years ago. The number of cars has grown rapidly, but the corresponding supporting facilities and supervision are relatively lagging behind, and parking space resources are scarce. It can be said that “parking difficulty” has become a major problem that needs to be solved urgently in China and the world. To solve this problem, on the one hand, it is necessary to increase the number of urban parking spaces, and on the other hand, it is necessary to improve the efficiency of parking.
“
introduction
With the progress of my country’s science and technology and economy, the automobile industry has entered a stage of rapid development in recent years. According to statistics, in 2013, the number of cars in my country reached 137 million, 5.7 times the number of cars in 2003, accounting for 54.9% of all motor vehicles, an increase of 29.9% over 10 years ago. The number of cars has grown rapidly, but the corresponding supporting facilities and supervision are relatively lagging behind, and parking space resources are scarce. It can be said that “parking difficulty” has become a major problem that needs to be solved urgently in China and the world. To solve this problem, on the one hand, it is necessary to increase the number of urban parking spaces, and on the other hand, it is necessary to improve the efficiency of parking.
The intelligent parking system designed in this paper adopts the geomagnetic detection mechanism combined with the wireless sensor network technology. It has the characteristics of low power consumption of the parking space detection node, convenient system deployment and maintenance, and low construction cost. 433 MHz transmission is used between the detection node and the routing node. No special requirements for the environment, strong anti-interference ability. The car owner can obtain the vacant information of the parking space at the first time through the system, thereby improving the parking efficiency and effectively alleviating the parking pressure.
1 System design scheme
The framework of the intelligent parking system is shown in Figure 1. The geomagnetic detection node located below the parking space collects the parking space occupancy information in real time, and then transmits the collected information to the conversion node after processing. The conversion node packages the received data into Socket data packets and transmits it to the sink node built by the ARM+Android platform. A sink node is responsible for sending the parking situation of a single parking lot to a remote data center, and generates parking guidance information and transmits it to the parking guidance subsystem of the parking lot. Android mobile client users can query the vacant information of parking spaces through the Internet.
2 System hardware design
2.1 Geomagnetic parking space detection node
The geomagnetic parking space detection node adopts a low-power design, with an average current consumption of tens of μA, and can be powered by a lithium battery. A 2 000 mAh lithium battery can be used for more than 3 years. The node does not need to replace the lithium battery frequently or charge the lithium battery, so that the node with waterproof industrial packaging can be buried under the parking space, which greatly facilitates construction, installation and later maintenance. The physical map of the parking space detection node is shown in Figure 2.
2. 1.1 Detection node main control chip
TI’s MSP430 series is a 16-bit hybrid microcontroller with a reduced instruction set and ultra-low power consumption. It has extremely low power consumption, rich on-chip peripherals, and convenient and flexible development methods. Its highly flexible timing system, multiple low-power modes, instant wake-up, and intelligent autonomous peripherals can not only achieve real ultra-high Optimized for low power consumption, but also greatly extends battery life.
2.1.2 Selection of detection methods
Traditional parking space detection methods include radio frequency identification, ultrasonic detection, infrared detection, induction coil, etc. These detection methods have great limitations, some have high power consumption, and some have higher requirements for the installed environment. Others are very susceptible to interference. Some new parking systems take advantage of the fact that cars will disturb the surrounding geomagnetic field, and use high-sensitivity magnetoresistive sensors to detect changes in the geomagnetic field around the parking space, as a basis for judging the presence or absence of vehicles in the parking space.
Current magnetic sensor technologies include Hall, anisotropic magnetoresistance (AMR), giant magnetoresistance (GMR), and tunnel magnetoresistance (TMR). The geomagnetic sensor used in this system is MAG3110 from Freeseale. The MAG3110 magnetometer offers superior performance due to its combination of TMR technology, high-resolution analog design, and dedicated embedded logic.
The parameter characteristics of MAG3110 are listed in Table 1.
The connection between MAG3110 and MSP430 microcontroller is shown in Figure 3.
2.1.3 Detection Node Wireless Transmission Module
Since the geomagnetic parking space detection node is installed below the parking space, when a vehicle is parked in the parking space, it will affect the transmission of wireless signals, and there are many walls or pillars with reinforced concrete structures in the parking lot, which will cause refraction interference to the wireless signal, so It is necessary to select a wireless transmission module below 1 GHz with obstacle penetration capability and transmission distance better than 2.4 GHz.
The nRF905 monolithic wireless transceiver mainly operates in the ISM frequency bands of 433 MHz, 868 MHz and 915 MHz. The nRF905 is programmed and configured by the MCU through the SPI interface. The current consumption of the nRF905 is very low. When the transmit power is -10 dBm, the transmit current is 11 mA and the receive current is 12.5 mA. The current consumption when entering the POWERDOWN mode is the smallest, and the typical value is lower than 2.5μA. , which is very suitable for the low-power and low-cost system design of this sensor node.
The connection between nRF905 and MSP430 microcontroller is shown in Figure 4.
2.2 Relay Node
The only difference in hardware between the relay node and the detection node is the lack of the MAG3110 geomagnetic detection module.
2.3 Converting Nodes
The conversion node, the relay node and the parking space detection node form a simple self-organizing Mesh network, and are responsible for unpacking the received data of the parking space detection node and then re-encapsulating them into Socket data packets, which are sent to the sink node and the subsequent remote data. center.
The conversion node is composed of low-power Wi-Fi module CC3200 and external nRF905. The Simple Link CC3200 device is a wireless MCU that integrates an ARM Cortex-M4 core running at 80 MHz. The device contains a variety of peripherals including a fast parallel camera interface, I2S, SD/MMC, UART, SPI, I2C and quad analog-to-digital converters (ADCs). CC3200 supports base station, access point and Wi-Fi direct mode, also supports WPA2 personal and enterprise security and WPS2.0. Using SmartConfig technology, AP mode and WPS2, a simple and flexible Wi-Fi service can be realized.
The transmit power and receive sensitivity of CC3200 are listed in Table 2.
The main hardware connection of the conversion node is shown in Figure 5.
3 System software design
3.1 Detection and processing algorithm
The working environment of the parking space detection node is relatively complex, and there are many interferences, such as: temperature, surrounding vehicles, etc., and the strength of the geomagnetic field at the same location will change with time, so the data collected by the parking space detection node cannot be simply used directly. These raw data need to be processed with the help of certain algorithms.
The algorithm mainly consists of three parts: smooth filtering algorithm, time-based threshold determination algorithm, and self-correction reference field algorithm. Considering the performance of the single-chip microcomputer of the parking space detection node, the design requirements of low power consumption and the effect of filtering, the smoothing filtering part uses an improved limiting moving window mean filtering algorithm, as shown below:
In the formula, i∈Z+, Ai represents the filtered result of the data collected by the parking space detection node, Si represents the sum of the data whose acquisition window length is W, Ci represents the value of the magnetic field of the parking space collected each time, and B represents the self- Correction of the reference field
value, T represents the judgment threshold, β represents the large noise figure, and △ represents the maximum sampling deviation.
3.2 Implementation of low power consumption
The low power consumption of the parking space detection node is mainly realized by combining the low power consumption mode (LPM) of the MSP430 microcontroller, the normally open circuit control MAG3110 and the event-triggered driving mode. The working process of the node is shown in Figure 6.
3.3 Wireless networking
In this system, the address of the detection node is fixed, the address of the detection node is bound to its actual geographical location, the conversion node, the relay node and the parking space detection node form a simple Mesh network, and the detection node can communicate directly with the relay node and the conversion node. Relay nodes can communicate with each other, and the network topology is shown in Figure 7.
The system data frame structure is shown in Figure 8.
3.4 Socket programming
The conversion node unpacks the received data of the parking space detection node and then repackages it into a socket data packet. In addition to the networking programming, the software part also includes the following three parts: SmartConfig programming of CC3200, parsing the data packet of the detection node, Socket programming.
4 System experimental test
4.1 Network test
Turn on the power of the detection node, relay node, and conversion node in turn, connect the relay node to the serial port, and print the debugging information. Figure 9 records the process that the data packet sent by the detection node FFFF A1 EC is sent to the conversion node 00 00 00 01 through two relay nodes.
4.2 Experimental data of parking space detection
In order to reduce the interference of surrounding vehicles on the detection results, the Z-axis data of MAG3110 is mainly used. The improved limiting moving window mean filtering algorithm is used for the Z-axis data (the window length is 16), and the filtering effect is shown in Figure 10.
The result of combining the self-calibration reference field and the threshold determination is shown in Figure 11.
Epilogue
The parking space detection system designed in this paper combines wireless sensor network technology and geomagnetic detection technology, and has the characteristics of high detection accuracy, convenient networking, strong anti-interference ability and low power consumption. The system can realize real-time monitoring of parking space usage, and has certain practical value and application prospects.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.