Design of network fingerprint access control system based on ARM and POE

Abstract: This paper designs a fingerprint access control system based on ML67Q5250 and PoE, and describes the system structure, working principle, software and hardware design. The system has the advantages of high-speed fingerprint recognition, fingerprint data protection, Ethernet power supply, simple structure, low power consumption and low cost. It can meet the needs of fingerprint access control system and has good application prospects.

A fingerprint access control design based on ARM and POE embedded system is proposed. The solution uses ML67Q5250 processor and its peripheral modules as the hardware platform and embedded Linux as the operating system to form the entire system. The system uses Ethernet power supply to realize the functions of the fingerprint access control system without the need for a dedicated power supply line. The implementation of this design has broad market application prospects.

1 System Structure

The whole system consists of four parts: FPS200 sensor module, ML67Q5250 fingerprint recognition module, Ethernet power supply module, and Ethernet data communication module form the hardware circuit. The software includes Linux embedded software and background support software to realize the network operation, fingerprint recognition and control, alarm prompt, power management and other functions of the network fingerprint access control system.

2 System Hardware Design

2.1 ML67Q5250 Fingerprint Recognition Design

The ML67Q5250 microprocessor belongs to the ARM7 series. The ML67Q5250 fingerprint recognition processor has fast Fourier transform function and high-performance fingerprint recognition processing. It adopts the fingerprint recognition algorithm of DFT mode, with high-speed recognition time within 0.8 s, high recognition accuracy FAR<0.001%, FRR<1.0%, 15 registrable indexes in the built-in flash memory, and it is scheduled to be expanded to 45. It is equipped with a security function that prohibits incorrect reading of fingerprint data, and has various interfaces that can be connected to external devices. In addition, the processor can realize external memory control and can store a large amount of fingerprint information. Another feature of the system is that the fingerprint recognition module communicates with the background database through the network. The processor has a large number of general input and output ports, which provides a broad space for future function expansion. The solution of the ML67Q5250 soft-core processor is used to create a buffer storage system with comprehensive functions and configurable, providing a solution for the network fingerprint access control system. Figure 1 is the hardware structure of the system.

ML67Q5250, as the system CPU, processes the fingerprint information collected by the sensor module and compares it with the fingerprint information in the database to achieve the purpose of confirmation and identification. In addition, ML67 Q5250 has rich peripherals and I/O, which provides convenience for future system upgrades. 2.2 FPS200 Fingerprint Sensor Module

FPS200 is a touch CMOS sensor based on the principle of capacitor charging and discharging. Its outer surface is an insulating surface. Each point of the sensor array is a metal electrode. The finger acts as the other pole of the capacitor. The sensing surface between the two forms a dielectric layer between the two poles of the capacitor. Due to the different distances between the ridges and valleys of the fingerprint relative to the other pole, the capacitance values ​​of the silicon surface capacitor array are different. In this way, the capacitance array value describes a fingerprint image. The FPS200 fingerprint sensor has a high-resolution 256×300 array, a 1.28cm×1.50cm sensing area, and a resolution of 500dpi. Image search technology can still recognize with high accuracy even when the finger is too dry or too wet. High-speed image transmission 30 frame/s (MCU mode), 13frame/s (USB mode) and 10frame/s (SPI mode). High-strength surface protection can effectively prevent damage to the chip from knocking, wear, chemical corrosion, etc. The automatic finger detection (AFD) function can wake up the CPU and save power consumption.

3 POE power supply design

3.1 Working process of POE power supply

A complete POE system consists of two parts: the power supply equipment (PSE) and the power receiving equipment (PD). The PSE device is the device that supplies power to the Ethernet client and is also the manager of the entire POE Ethernet power supply process. The PD device is the PSE load that receives power, that is, the client device of the POE system, the network fingerprint access control system. Figure 2 is a schematic diagram of the POE power supply principle.

3.2 Ethernet Power Controller LTC4266

Linear Technology introduces the LTC4266, an IEEE802.3af Ethernet Power Supply (PoE) controller, a 4-channel Power Supply Equipment (PSE) controller designed for use in Ethernet Power Supply systems that comply with IEEE 802.3 Type 1 and Type 2 standards. External power MOSFETs enhance system reliability and minimize channel resistance, thereby reducing power consumption and eliminating the need for additional heat sinks, even at Type 2 power levels. External power components also allow for use at higher power levels while remaining compatible with IEEE standards in other respects. Port pins rated for 80 V provide rugged protection against external faults.

4 System software design

Fingerprint image preprocessing is the first step in the fingerprint recognition process, and its quality directly affects the effect of the fingerprint recognition system. Image preprocessing includes filtering, sharpening, binarization, thinning and denoising. Since the lines in the fingerprint image have the characteristics of consistent directionality, basically equal width and basically the same spacing within a local range, directional image filtering is used to process the fingerprint image. Sharpening indicates strengthening the boundaries between fingerprint lines to highlight edge information, enhance the contrast between ridges and valleys, and facilitate binarization. Laplace single mask algorithm can be used to achieve sharpening during design.

FPS200 adopts the query working mode programming method: first initialize the registers of FPS200, write the control word to the corresponding register, set the parameters of fingerprint collection, mainly set the DCR, DTR, PGC registers; query and wait, when the fingerprint is automatically collected by FPS200 and enters the data register, the fingerprint data is stored in the specified storage space. Figure 3 is the fingerprint image collected through the experiment. The fingerprint image obtained by the fingerprint recognition instrument collection system has the advantages of strong contrast, clarity, and low distortion.

5 Conclusion

This paper introduces the design and implementation of a network fingerprint access control system with ML67Q5250 as the core. The network power supply module LTC4266 complies with the Ethernet power supply (PoE) system of IEEE 802.3 Type1 and Type2 standards. The sensor uses the FPS200 fingerprint sensor. Through the form of Ethernet power supply, the processed data is communicated through the network and the computer, which greatly reduces resources. With the development of network technology today, this network fingerprint access control system has a good application prospect.