基于區(qū)域控制網(wǎng)絡(luò)（can）的智能家居自動化火災(zāi)報警系統(tǒng)外文翻譯（英文）

1、K. C. Lee and H.-H. Lee: Network-based Fire-Detection System via Controller Area Network for Smart Home Automation Contributed Paper Manuscript received July 29, 2004 0098 3063/04/$

2、20.00 © 2004 IEEE 1093Network-based Fire-Detection System via Controller Area Network for Smart Home Automation Kyung Chang Lee, Associate Member, IEEE, Hong-Hee Lee, Member, IEEEAbstract — This paper presents a ne

3、twork-based fire- detection system via the controller area network (CAN) to evaluate the feasibility of using such a home automation protocol in a smart home. In general, a conventional fire- detection system has sever

4、al shortcomings, such as weakness to noise, because it uses an analog transmission with 4 – 20 mA current lines. Hence, as an alternative to a conventional system, this paper describes the structure of a CAN-based fir

5、e- detection system and the design method of a CAN communication network. The performance of the proposed system is evaluated through experimental tests. The CAN has several advantages, such as low cost and ease of i

6、mplementation, as compared to other low layers of the BACNet, such as Ethernet or ARCNET. Therefore, if the CAN is selected for the low layer of the BACNet, a home automation system can be implemented more effectively

7、1. Index Terms — network-based fire-detection system, controller area network (CAN), home automation system, home network system, intelligent building. I. INTORDUCTION Recently constructed buildings tend to be “intelli

8、gent” to enhance the convenience and safety of occupants [1][2]. Thus, the requirements for home networks and automation systems have increased with the demand for smart homes [3]. To satisfy the needs of occupants,

9、home appliances such as refrigerators or microwave ovens, multimedia devices such as televisions or audio systems, and Internet devices such as PCs have been included in intelligent buildings, as shown in Fig. 1. In

10、smart homes, appliances are connected to the home network so that it becomes possible to control and monitor the appliances from inside the house or remotely using a cellular telephone or PDA. To realize a home networ

11、k system, several standards, such as Echonet, Konnex, LnCP, and LonWorks, are being developed by standards organizations and corporations [4]. In addtition, home automation systems, such as HVAC (heating, ventilatio

12、n, and cooling), lighting, anti-crime, and fire-detection systems are being used to improve convenience 1 This work was supported (in part) by the Korea Science and Engineering Foundation (KOSEF) through the Network-ba

13、sed Automation Research Center (NARC) at University of Ulsan. H.H. Lee is with the School of Electric-Electronic Information System Engineering, University of Ulsan, Ulsan 680-749, Korea (corresponding author to prov

14、ide phone: 82-; fax: 82-; e-mail: hhlee@main.ulsan.ac.kr). K.C. Lee is with the Network-based Automation Research Center, University of Ulsan, Ulsan 680-749, Korea (e-mail: gclee@pnu.edu). and safety for occupants. Gen

15、erally, in a conventional home automation system, home equipments, such as switches, valves, or fire detectors, are directly connected to controllers for the HAVC or fire-detection systems. However, because of the an

16、alog transmission method, which uses a 4 – 20 mA current, conventional fire-detection systems detect a fire as occurring when the current received from a fire detector exceeds a predefined threshold. Hence, the system

17、 has disadvantages such as a weakness to noise of various forms, including impulses or short-circuits, and a lack of awareness of the actual location of a fire. Recently, in order to solve these problems, research on

18、 fire-detection systems that use digital or wireless transmission instead of analog transmission has been conducted [5][6]. In parallel with this, research is also being performed on communication protocols, such as t

19、he BACNet or LonWorks [7]. Recently, researchers have proposed applying the controller area network (CAN) [8] to the home network protocol, and attention has increasingly been centered on the CAN in automation (CiA)

20、 [9]-[11]. Especially, the price of a CAN microcontroller, which can be integrated into one-chip by semiconductor manufacturing companies, is low and many CAN development tools are coming onto the market. Also, many

21、application programs had been developed in various fields such as automotive, robotic, and industrial automation systems. Therefore, a home automation system using the CAN be more easily implemented, because realizati

22、on at the sensor or actuator level is relatively straightforward as compared to a system using an Ethernet that is the lower layer of the BACNet. To evaluate the feasibility of the CAN for home automation protocols,

23、 this paper introduces a network-based fire-detection system via the CAN. The paper is organized into five sections, including this introduction. Section II gives the architecture of the network-based fire-detection s

24、ystem. Section III presents the design method of the CAN-based fire-detection system, and verifies that a CAN communication network satisfies the real-time requirements of a home automation system. Section IV describ

25、es an experimental testbed that was implemented to verify the feasibility of the proposed system. Finally, a summary and the conclusions are presented in Section V. II. ARCHITECTURE OF NETWORK-BASED FIRE-DETECTION SYS

26、TEM A. Conventional fire-detection system Fig. 2(a) shows the architecture of a conventional fire- detection system. As shown in the figure, fire detectors (e.g., K. C. Lee and H.-H. Lee: Network-based Fire-Detection S

27、ystem via Controller Area Network for Smart Home Automation 1095measured by each fire detector can be sent to the receiver. It is straightforward to apply such a system to an intelligent fire detection system and to a

28、pply a reasoning algorithm, because the receiver can make use of the digitized detection values of the smoke and heat from multiple fire detectors installed in the same guard area. As changes in the detection signals

29、can be calculated and compared, assessment of risk can be improved [12]. Besides, a PC-based receiver can be used in a network- based fire-detection system for greater convenience. Generally, it is easier to install

30、and maintain a PC-based receiver than a conventional dedicated receiver. Since man-machine interface (MMI) technology can be applied to a PC-based receiver, it is straightforward to create a user interface and integra

31、te the fire- detection system into a home network system. Recently, several protocols, such as BACNet, LonWorks, and Bluetooth, have been developed to implement network-based fire- detection systems in intelligent buil

32、dings [5][6]. III. DESIGN OF A CAN-BASED FIRE-DETECTION SYSTEM A. Overview of the CAN protocol CAN 2.0B is a network protocol that was specifically developed for connecting the sensors, actuators, and ECU’s of a vehicl

33、e. CAN 2.0B supports data rates from 5 kbps to 1 Mbps, which allows the CAN network to be used to share status information and for real-time control. The network topology can be either a linear bus or a star. CAN 2.0

34、B has the following properties: ? distributed medium access control, meaning that each device has the same privilege to use the shared medium. ? contention-based and nondestructive bus access, indicating the acces

35、s to the bus has to be obtained through some type of competition but with no loss of network capacity due to the contention. ? content-based addressing, implying that each message packet has a unique identifier accord

36、ing to its content. ? cyclic redundancy check for error detection, and ? error confinement to block any adverse effects of a network component failure. A device on the network can transmit its message whenever the net

37、work is idle. When the network is busy, the packet transmission has to be delayed until the on-going transmission has finished. Because electronic signals on the network have a finite speed of propagation, there is al

38、ways the possibility that multiple devices will start their transmissions within a short interval. This situation, called message collision, is resolved by comparing the identifiers of the messages involved; that is,

39、the message with the lowest identifier value wins the right to use the network, while the other devices must stop their transmissions immediately. As the identifier is located at the beginning of the packets and the

40、electronic signal for zero is designed to overwrite the signal for one, the message with the lowest identifier value finishes the contention with no damage to the packet. The other devices will try to resend their pac

41、kets when the first transmission has finished. This arbitration procedure is shown in Fig. 4. B. Design of a CAN-based fire-detection system This section presents the design methodology of a CAN for a fire-detection s

42、ystem. Fig. 5(a) shows the frame exchange between the receiver and the fire detector during the normal state. In the figure, the receiver inspects the state of the fire detectors by sending a poll frame to each fire d

43、etector and actuator in the poll list once every TP (polling period). When it receives the poll frame, each fire detector and actuator transmits a status frame that includes its own fire detection value back to the r

44、eceiver. Thus, the receiver can periodically perceive the state of a fire-detection system such as a fire detector failure or a breakdown such as an open circuit. Power onCheck the identifierIs the bus free?Is a mess

45、age transmitted?Receive a transmitted message?Transmit the messageCollision occurred?Continue the transmissionReceive the messageArbitration won?Stop the transmissionyesnoyesnonoyesyesyesnonoFig. 4. Flowchart of the C