Sensors & Transducers, Vol. 156, Issue 9, September 2013, pp. 89-94

Sensors & Transducers © 2013 by IFSA http://www.sensorsportal.com

An Embedded-technology-based Security Monitoring System Design of Intelligent Building Yuanyi ZHANG, Ying ZHANG School of Architecture, Fuzhou University, Xueyuan Road No. 2, Fuzhou, 350108, China Tel.: (86)-591-22866392, fax: (86)-591-22866380 E-mail: [email protected] Received: 18 June 2013 /Accepted: 25 August 2013 /Published: 30 September 2013 Abstract: Security monitoring system is the important content of safety management in intelligent building. This paper aimed at the defects of security monitoring system in intelligent building, and used embedded technology, Internet technology, video compression/decompression technology, sensors technology to design a security monitoring system based on Internet. Users can detect the illegal intrusion and fire hazards indoors in real-time, through computers or mobile phones. The validity and efficiency of the system was validated via testexperiment. The results show that this system has several merits such as good safety, good generality, simple operation, easy maintenance, low cost and so on. As an indispensable part of digital home, this system has extremely important realistic meanings to the future’s construction and development for intelligent building. Copyright © 2013 IFSA. Keywords: Intelligent building, Internet technology, Security monitoring, Sensor technology.

1. Introduction Intelligent building is a new discipline. It is a product which combined with information technology, automation technology, modern communications technology, microelectronics, building technology and other advanced technology. Meaning from the engineering, the intelligent building is to set building automation systems, communication network systems, office automation systems and security systems in the buildings, and to make a uniform management. It is a system engineering which combined with monitoring, service and management, and to achieve optimization. As the building intelligence is gradual increase with the development of science and technology, and its contents have been enriched. So, there is no uniform definition of the “intelligent building” in the international arena to now. With the development of social economy and promote high-tech applications,

Article number P_1331

due to negligence caused by various types of fire accidents have occurred, a variety of criminal means to become more subtle, more specialized. Therefore, to enhance monitoring efforts inside and outside of buildings is particularly important. At present, the security monitoring program of intelligent building is mainly achieved through the two subsystems: video surveillance system and fire detection and alarm system. Video monitoring system, also known as closed circuit monitoring system, is mainly acquiring the signals through the front of the camera, monitor, intrusion detectors and other equipment, and sending all kinds of information to the matrix host computer of central control room through the decoder that field configuration and the transmission lines, then the host switch and display all kinds of signals to different output devices. Fire detection and alarm system is a controller which mainly consisted by the primary backup power

89

Sensors & Transducers, Vol. 156, Issue 9, September 2013, pp. 89-94 supply, fire alarm, and fault alarm. It collected the fire signal, and displayed the controller state via LED [1, 2].Therefore, either from the video surveillance system or fire detection and alarm system, their structure is relatively complicated and the installation costs are high, it is not suitable for the needs of ordinary families. Thus greatly restricts the spread and development of intelligent building. In addition, the traditional video surveillance system used analog signal transmission, and it requires a lot of video recorders, video tapes to store image information, which is bound to bring a lot of wiring work, a large number of images stored in the work, a lot of image retrieval [3]. Moreover, in the operation, video surveillance system and fire detection and alarm system are two independent systems, when the fire hazards and illegal intrusion occurred simultaneously, they can not be controlled unity. Therefore, this paper plans to use embedded technology, Internet technology, video compression/decompression technology and so on, integrated video surveillance system with fire detection and alarm system. Then through the Internet to transmit video signals and fire signals, use of computer hard disk system to store images, so that to save the system installation costs and reduce the workload of image storage and retrieval effort.

2. System Design Scheme Security monitoring system consisted of the main controller AT91SAM9260, Internet interface circuit, memory circuit, video capture circuit, and sensors module (body sensors and smoke sensors). Among them, the video capture circuit and the body sensor mainly achieved the video monitoring of system. When the body sensors detected there were any illegal intrusions happened, the system will alarm and record the live video through video capture circuit. Smoke sensors main achieved fire detection and alarm function, when the fire hazards happened, the system will issue a warning signal and displays the location of the fire. Fig. 1 illustrated the structure diagram of system.

3. Hardware Design of System Considered the developing cost and the need of this system is to use Java language to develop embedded Web Server, we selected AT91SAM9260, the 32 bit micro-controller with ARM9EJ-S kernel [4], as the main controller. It has abundant system resources, and support Java language (ARM Jazelle Technology for Java Acceleration). It includes a 10/100M Base T Ethernet MAC, which can process TCP/IP protocol easily, and transmit IP packet. The system can access Internet via LAN, router or switch. And users can use Internet to detect the illegal intrusion and fire hazards indoors in real-time.

3.1. Internet Interface Circuit As the video signal transmission is achieved through the Internet, so the design of Internet interface circuit is very important. Internet interface is composed of MAC controller and physical interface from the view of hardware. Since AT91SAM9260 has integrated Ethernet controller (EMAC), it only needs to realize the physical interface which can reduce the complexity of system design and improve its efficiency [5, 6]. In this system, it took DM9161 as Ethernet’s physical interface. DM9161 connect with Internet through a network insulate transformer HR911105. It is very simple to connect Ethernet controller DM9161 with the network insulate transformer HR911105, so long as connected DM9161 receiving signal wire RX+, RX- and sending signal wire TX+ and TX- with HR911105 RD+, RD- and TD+, TD- respectively. Network insulate transformer HR911105 has integrated the RJ45 joint. Therefore it may access Internet directly through HR911105. Since HR911105 takes the work status indicator light itself, it can indicate the system’s work status. The signal’s sending end and receiving terminal access Internet through HR911105.

3.2. Memory Circuit Taken into account that the system needs to take up a lot of memory resources in dealing with video images, the system extended a K9F1208UOM chip which is a 64 M NAND Flash memory produced by Samsung, and a HY57V561620CT-H chip which is a 32 M SDRAM memory produced by Hynix Semiconductor Inc.

3.3. Video Capture Circuit

Fig. 1. The structure diagram of system.

90

Video capture circuit mainly constituted of the image sensor. The AT91SAM9260 has an Image Sensor Interface (ISI). So it can connect a CMOS Image Sensor to capture all kinds of images via ISI.

Sensors & Transducers, Vol. 156, Issue 9, September 2013, pp. 89-94 And the processor can transform data before it saving images to the memory through DMA channel. The principle of video capture circuit is illustrated in Fig. 2. PB[0..31]

PA[0..31]

PB10 PB11 PB18 PB19 PB20 PB21 PB22 PB23 PB24 PB25 PB26 PB27 PB28 PB29 PB30 PB31 PA23 PA24

1.8V

C68 Cap Semi 104

3.3V C69 Cap Pol3 10V/10uF

PIO_CNTRL1 TWCK ISI_HSYNC ISI_PCK ISI_D8 ISI_D6 ISI_D4 ISI_D2 ISI_D0

PB18 PA24 PB30 PB28 PB10 PB26 PB24 PB22 PB20

C70 Cap Pol3 10V/10uF

P5 1 3 5 7 9 11 13 15 17 19 21

2 4 6 8 10 12 14 16 18 20 22

PB19 PA23 PB31 PB29 PB11 PB27 PB25 PB23 PB21

C71 Cap Semi 104

PIO_CNTRL2 TWD ISI_MCK ISI_VSYNC ISI_D9 ISI_D7 ISI_D5 ISI_D3 ISI_D1

CMOS Sensor

Fig. 2. The schematic diagram of video capture circuit.

3.4. Sensors Module

as the smoke sensor in this system. SS-168A has two dry contacts (output pins), when it detects there are some fire risks happen, its buzzer will give alarm and the indicator light will display, then outputs signals to the host controller from the two dry contacts. AT91SAM9260 will call Embedded Web Server to display the location of the fire on the home page of system after it received and processed the signals.

4. Software Design of System Security monitoring system used embedded Linux 2.6.19 as the software platform, and adopting modularity design. So that it has good portability and good expansibility. The kernel and root file system of Linux 2.6.19 was programmed in the Flash memory inside, so the boot process of embedded Linux 2.6.19 must be based on the target platform to prepare, and to plan the address space of memory.

4.1. Planning of Memory Address Space

The sensors module consists of two parts: body sensors and smoke sensors. Fig. 3 illustrated the schematic diagram of sensors module. In this system, we designed two channels body sensors and two channels smoke sensors. The body induction module is mainly for the function of guard against theft. In this system we chose TDL718 which is a body induction module based on infrared technology. The module has many merits such as high sensitivity, reliable, and it can induce automatic. The second pin of TDL718 will output a high level when there were some people entering its inductive scope, and stop the high level then output a low level when people left the inductive scope. Therefore, the system can know whether there were any people enter the user’s house illegal via detect the level signal of PC25 and PC26. The system will automatically alarm and notify the image sensor to record live video when there are some abnormal conditions happened.

In this system, the address space of flash memory was from 0x0200A000 to 0x0800A000, so we can divide the flash memory. The address space from 0x0200A000 to 0x0280A000 was the storage space of Boot-loader, from 0x0280A001 to 0x0400A000 was used to store the kernel image file, from 0x0400A001 to 0x0800A000 was used to store the root file system, and as a device file was loaded into the system. Fig. 4 illustrated the address space planning of flash memory.

PC［ 0..31］ P6 5V PC25

1 2 3

P8 PC27 PC28

TDL718_1 P7 5V PC26

1 2 3 TDL718_2

1 2 3 SS-168A_1 P9

PC29 PC30

1 2 3 SS-168A_2

Fig. 3. The schematic diagram of sensors module.

The function of smoke sensors is to realize fire alarm. The system will detect the ions of smoke through smoke sensor. When the concentration of smoke indoor is exceeded the limit, it will automatically alarm and display the location of the fire on the home page of system. We chose SS-168A

Fig. 4. The address space planning of flash memory.

After the system reset, according to the system’s setting, system program counter (PC) points to the SDRAM address 0 at first, to complete the necessary system initialization. Then, the system program counter jump to flash address 0, that is 0x0200A000, to complete the system SDRAM initialization, and the Boot-loader code will be copied to the SDRAM address 0, which is 0x00001FFF. Subsequently, PC pointer will jump to the flash address 0x00001FFF again, to executive the Boot-loader kernel and load code in the internal of SDRAM memory. The reasons of executive Boot-loader within the SDRAM without

91

Sensors & Transducers, Vol. 156, Issue 9, September 2013, pp. 89-94 in flash memory, is because the speed of executive code in SDRAM is much faster than in the flash inside. So it can help to speed up the boot speed of system. Then the work of Boot-loader was copied the compressed kernel image to SDRAM address 0x0000A004, and transfer the address parameters of the root file system to the kernel. After the kernel had copied to SDRAM, PC pointer will jump to 0x0000A004, to extract the kernel, and executive the kernel boot. The system will find the root file according to the address parameter which transferred from Boot-loader, and load to security monitoring system. In accordance with the above arrangement, we download Boot-loader, kernel and root file system to the Flash in succession through the software SAMBA1.11 and tftp, then we can start the system, embedded Linux can be boot up into the normal state.

4.2. Design of Video Compression Algorithm Embedded Web Server has unified interface and communication protocol, so it can provide unified operation mode and control interface based on Internet Explorer. In this system, there were four modules on the Embedded Web Server: Files module, HTTP request and process module, Java Applet data request and process module, and Monitoring module. Monitoring module is responsible for monitoring ports, and establishing a connection between the system and the Explorer of the client through three times handshake process of TCP protocol. When there is a request from the Internet Explorer, the monitoring module will judge whether it is a HTTP request or a Java Applet request according to the port number. Then call HTTP request and process module or Java Applet data request and process module to start the corresponding process. Files module was realized by the files system of Embedded Operating System. The Files module including GGI processing files (Common Gateway Interface), Static web page files and Applet code files. And the GGI processing files have two functions: one is to realize user authentication mechanism; another is to parse and process the parameters which come from HTTP message, when the users set up the parameters for control through Internet Explorer.

5. System Testing In order to confirm the correctness of system design, we took a comprehensive test to the system functions (Including the communication interface, data transfer and overall performance.) and network performance. Before system testing, we should set up related parameters, such as IP address, gateway address, server address, the video frame size, number of frames per second acquisition, network bandwidth and so on.

92

5.1. System Functional Testing First of all, we establish two hyper terminals in one personal computer, one is connected with the serial port another is connected with Socket. When we input some data on the first hyper terminal, the second hyper terminal was response immediately. It means the communication interface of system is well [8]. Secondly, we connected the system with the serial port of personal computer through electrical level transfer circuit, and verified the data transmission via hyper terminal. Thirdly, we establish a Web Server through Java Script language and PWS Server. As the embedded Web Server has a standard interface format and communication protocol, the web servers embedded in equipment can provide a unified browser-based operation and control interface, for any legitimate users who accessed the network [9]. Finally, used a net wire to connect this system with a router (or switch), and communicated with the Web Server through Internet Explorer. After several experiments, the results show that it is easy to establish connection between security monitoring system and Internet, and the users can detect the illegal intrusion and fire hazards indoors in realtime, through computers or mobile phones. Fig. 5 illustrated the operation interface of security monitoring system.

Fig. 5. The operation interface of security monitoring system.

The whole safety monitoring system can be set by monitor settings, including the storage path of video and the save format. At the same time, you can set that system to automatically record video and save it when someone illegally broken into the monitor area. And you can download the video which temporarily stored in the experimental system through the option of video download. Report print is mainly for intrusion detection and fire detection. It can print the monitoring data within a period of time, to print. As the intelligent building security monitoring system is based on B/S model, so the function of background management was to management and set the embedded Web Server. When the body sensors detected that someone had

Sensors & Transducers, Vol. 156, Issue 9, September 2013, pp. 89-94 entered monitor region, the system will display “1” in the corresponding position of intrusion detection. Similarly, when the corresponding location of the smoke sensor was displayed “0”, it means the system detected there was no fire hazards, displayed “1” to means detected fire hazards.

5.2. Network Performance Testing Network performance testing was mainly for the video monitoring of system. It was necessary to install the WildPackets Etherpeek Nx software in the host before testing. Through the software, we can capture the network performance data of system. Here, we set to sample the network data packet every 10 seconds, the main network parameters which captured by the software are: the packet number and byte count of sending every 10 seconds, the utilization of network bandwidth and so on. We tested the system through setting different bandwidth. Table 1 illustrated the network performance situation of system in different bandwidth.

From Fig. 6 we can see clearly that before 9:23:39 and after 9:29:19, there were still image-based in monitor area, and the average rate of video transmission was to 960.6 Kbps. There were sharp moving objects in monitor area between 9:23:39 and 9:29:19, lead to a greater jitter of video transmission rate, the maximum rate was 1189 Kbps and the minimum rate was 806 Kbps. The tests found that, for still images or motion do not change much, the overall compression ratio was high, and using smaller bandwidth can obtain clear graphics. And motion pictures are required for higher bandwidth, the system can be based network bandwidths, automatically adjust the amount of data the size of the single frame to ensure picture fluency. When the network bandwidth is a serious shortage, the display screen will appear mosaic phenomenon. After several experiments, the results show that it is easy to establish connection between Security Monitoring System and Internet, and the users can detect the illegal intrusion and fire hazards indoors in real-time, through computers or mobile phones.

6. Conclusion Table 1. Network Performance in Different Bandwidth. Setting Bandwidth (Kbps)

The average bandwidth (Kbps)

100

131.5

200

211.4

600

605.8

1200

960.6

Display Effect More blurred, with Mosaic Slightly blurred, no mosaic phenomena Still images clear, motion images slightly blurred Still images clear, motion images is also clear

When the system was running at a steady state, and the setting bandwidth was 1200 Kbps, the resolution of video stream was 640 × 480, and the frame rate was 50. In this case, the network performance of the system was shown in Fig. 6. Network performance Rate(Kbps)

1400 1200 1000 800 600 400 200 0 9:32:59

9:31:49

9:30:39

9:29:29

9:28:19

9:27:09

9:25:59

9:24:49

9:23:39

9:22:29

9:21:19

9:20:09

Rate Kbps

Time(Hours:Minutes:Seconds)

Fig. 6. The network performance of video monitoring system.

The information technology whose core is computer technology, control technology and measurement and control technique, is rapid development since the 1990’s. It makes people to set a higher request to security, comfort and efficiency of building environment. As the important content of safety management in intelligent building, the design of security monitoring system will affects the performance of the entire intelligent building directly. There were many merits of this system, such as good safety, good generality, simple operation, easy maintenance, low cost and so on. As an indispensable part of digital home, this system has extremely important realistic meanings to the future’s construction and development for intelligent building, it can meet the needs of security monitoring for an ordinary home.

Acknowledgements This work is supported by 2012 National Science and Technology Support Program of China (No.2012BAJ14B05), the National Natural Science Foundation of China (No.51278123), Natural Science Foundation of Fujian Province (No.2011J01303), and Education Department Foundation of Guangxi Province (No. 200808LX273).

References [1]. R. R. Yager, Some measures relating partitions useful for computational intelligence, in International Journal of Computational Intelligence Systems, Vol. 1, No. 1, 2008, pp. 1-18.

93

Sensors & Transducers, Vol. 156, Issue 9, September 2013, pp. 89-94 [2]. Fang Qiansheng, Experimental Guide to Intelligent Buildings, China Communication Press, 2003. [3]. Geng Jianjun, Design and application of video surveillance system in intelligent building, China, Shandong University Master’s Thesis, 2007, 5. [4]. AT91 ARM Thumb Microcontrollers AT91SAM9260 Preliminary (http://www.atmel.com/ dyn/resources/prod_documents/6221s.pdf). [5]. Xiang Yang, Yuanyi Zhang, Rongyang Zhao, Study and design of home intelligent system based on embedded internet, in Proceedings of the International Conference on Embedded Software and Systems Symposia (IEEE CS PRESS), 2008, pp. 344-349. [6]. M. V. C. Da Silva, N. Nedjah, M. de Macedo Mourelle, Optimal IP assignment for efficient NoCbased system implementation using NSGA-II and MicroGA, International Journal of Computational

[7]. [8].

[9].

[10].

Intelligence Systems, Vol. 2, No. 2, 2009, pp. 115-123. Mei Dacheng, Yang Daqian, Zhou Yong, Embedded wireless video capture system design, Microcomputer Information, Vol. 23, No. 7, 2007, pp. 19-20. Yuanyi Zhang, Xiang Yang, Design of phone remote measurement and control system in intelligent building, in Proceedings of the 9th International Conference on Electronic Measurement & Instruments (IEEE PRESS), 2009, pp. 392-395. M. Can Filibeli, O. Ozkasap, M. Reha Civanlar, Embedded web server-based home appliance networks, Journal of Network and Computer Applications, Vol. 30, Issue 2, 2007, pp. 499-514. Bu Chunyuan, Xu Dahua, Development of intelligent laboratory-monitoring system, Experimental Technology and Management, Vol. 24, Issue 7, 2007, pp. 145-147.

___________________ 2013 Copyright ©, International Frequency Sensor Association (IFSA). All rights reserved. (http://www.sensorsportal.com)

94