KR20040016921A - Control System for Home Appliance Network - Google Patents

Abstract

PURPOSE: A system for controlling a network of household appliances is provided to embody an optimized household appliance network in a low cost by embodying the network using a serial communication function of a low function microcomputer used in the household appliance. CONSTITUTION: A plurality of household appliances have a serial communication function, respectively. A communication medium connects a plurality of the household appliances. A communication layer having an application layer, a data link layer and a physical layer is defined for communication among a plurality of the household appliances. If a predetermined communication event is generated, the communication is performed among a plurality of the household appliances through the communication layer.

Description

Home Appliance Network Control System

The present invention relates to network control apparatus and method, and more particularly, to a home appliance network control system.

At present, home automation for automatically controlling home appliances in each home or remote is almost commercially available. In the early stages of home automation, each device was controlled separately using a telephone or an infrared ray, and there was no linkage between the devices, but now, a communication network is used to establish a network between home appliances and control the network. We are using a method that allows integrated management.

Many microcomputers currently applied to home appliances have a built-in serial communication function and are configured to communicate with other microcomputers or devices. Such microcomputers have various sizes of resources that can be used for communication, such as memory, depending on the characteristics of the product. In the case of multimedia products such as PC (Personal Computer) and TV or audio, a high-performance hardware specification is adopted to operate various basic functions, so a specification for a large amount of data and high speed communication is required.

On the other hand, refrigerators, washing machines, microwave ovens, lamps, gas alarms, stands or boilers are generally simpler than the functions of the PC or multimedia products described above. . In the case of home appliances employing such low-function microcomputers, basic remote control and operation status monitoring are the main communication targets. Therefore, a standard for communication using a small microcomputer resource is required.

However, in the case of communication protocols used or ongoing for the purpose of communication between home appliances, a separate communication module such as a modem is additionally installed in each device or used to use the high specification communication standards used in the PC or multimedia devices. It is attempting to modify some of the high specification communication standards.

However, the home appliance according to the prior art has to install a separate hardware communication module such as a modem for each device by applying a high specification communication standard used in a PC or a multimedia device to apply unnecessary communication specifications beyond the actual function. There is a problem of inefficiency and cost increase.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide a home appliance network control system capable of satisfying low cost / high efficiency characteristics in home appliances employing low-function microcomputers used at home. .

1 is a view showing a network of home appliances

2 is a diagram illustrating a communication structure between a master / slave combined device according to the present invention;

3 is a diagram showing a detailed configuration of a master-slave communication structure according to the present invention;

4 illustrates a half duplex communication structure according to the present invention.

5 shows a 1 Request-1 Response communication cycle.

6 is a diagram illustrating a communication cycle in case of a packet error

7 illustrates a 1 request-multiple response communication cycle.

8 shows a 1 request communication cycle.

9 is a diagram illustrating a division form of a communication layer.

10 illustrates an inter-layer packet communication structure.

11 shows the overall packet structure

12 illustrates the structure of a request / notification packet.

13 illustrates the structure of a response packet.

14 shows an address structure

15 is a diagram illustrating a network code classification method.

16 is a view showing a group address according to product type

Fig. 17 shows the group address according to the installation place

18 is a view showing group addresses according to installation place and product type

19 is a diagram illustrating a header structure of an event file.

20 is a view showing the body structure of an event file

21 is a diagram illustrating the configuration of a response packet when an error occurs.

In order to achieve the above object, the present invention comprises a plurality of home appliances having a serial communication function and a communication medium connecting the home appliances, the application layer, data link layer for communication between the home appliances And a communication layer comprising a physical layer, and when a predetermined communication event occurs, communication is performed at the home appliance using the communication layer.

Therefore, according to the present invention, it is possible to implement a simple and efficient home appliance network control system at low cost.

Hereinafter, a preferred embodiment of a home appliance network control system according to the present invention with reference to the accompanying drawings in detail as follows.

First, referring to the network of the present invention, as shown in FIG. 1, multimedia device products such as TV and audio connected to an A / V network through a gateway connected to an external Internet network, a printer, a scanner, a PC camera connected to a PC network, etc. Its sub-products include products such as refrigerators, air conditioners, washing machines, vacuum cleaners, microwave ovens, humidifiers, lights, stands, and gas alarms, connected via a network manager to a living network. In this case, the network manager may be a separate management device or a user's PC.

Looking at such a home appliance network control system of the present invention.

First, the present invention is applied to the master (slave) method. In other words, all communication cycles are initiated by the master and terminated by the master device. Any device can be a master, but in order to do so, it must have a function to control the flow of data on a communication line and have information and control codes for products connected to a network. Therefore, only a device with a user interface function such as a PC is equipped with all the functions of the master, and other devices can only serve as masters of limited functions such as communication with a predetermined slave or communication by a simple control code.

In addition, while maintaining the basic master-slave communication method as described above, a device in which the master and the slave logically coexist so as to enable direct communication between devices, that is, peer-to-peer communication. That is, as shown in FIG. 2, a device (hereinafter, referred to as a P2P device) that is physically one device but logically independent of a master and a slave is defined.

All products connected to the network are based on P2P devices, but, as shown in FIG. 3, may be defined as master, slave, transmit only, or receive only due to the hardware characteristics of the product.

That is, the master has a function of starting and terminating communication with the corresponding slave when a communication start event occurs due to a change of state in the end user or device in order to start a new communication cycle.

In case of slave, it is always in standby state and cannot request communication from other device by itself.

In the case of P2P device, the master and slave logically coexist inside, and when communication start event occurs due to the change of end user or device internal state, it acts as master and leads the communication cycle. Receive standby.

In the case of a transmission-only device, only a transmission is possible due to hardware characteristics. An infrared remote controller corresponds to this.

Receiving-only device is a device capable of receiving only due to the hardware characteristics, and this includes a product that is equipped with an infrared receiver and is operated by a battery.

Next, the home appliance network control system according to the present invention performs signal transmission in a half-duplex method using one bus, as shown in FIG. 4.

In other words, data is not received from another device at the time of transmission, and data is not transmitted to the other device at the time of reception. This is to minimize the use of memory for communication, and at the same time to correspond to a network composed of one bus using a serial communication function, such as the home appliance network of the present invention.

Therefore, both the master and slave can share the memory for transmission and reception. In the case of P2P devices, since the master and slave do not operate at the same time, the memory for transmission and reception can be shared as one. This eliminates the need for Interupt, which increases programming freedom on the product side.

Next, the home appliance network system according to the present invention performs communication in a one-cycle manner. The 1-cycle method may be divided into a 1 request-1 response method, a 1 request-multi response method, and a 1 request method.

In this case, the 1 request-1 response method is a method in which one master transmits only 1 packet to one slave and the slave receives 1 packet in response thereto, and the master receives and terminates it. And, as shown in FIG. 7, in the one request-multi-response scheme, one master transmits one packet to multiple slaves, and each slave transmits one packet in response thereto, and the master continues to wait for a response and then sets a predetermined maximum. The communication is terminated when the reception time elapses. As shown in FIG. 8, the one request method is a method in which one master transmits one packet to one or more slaves and terminates communication without waiting for a response. At this time, if there is data transmission composed of several packets, the master divides the packet into the size of the corresponding slave and transmits the data in one packet unit. Meanwhile, FIG. 6 illustrates a method in which a response is resent when a response error occurs in a slave and communication is terminated by receiving a response accordingly.

Next, the home appliance network control system according to the present invention has a protocol consisting of a physical layer, a data link layer, and an application layer.

In the case of the TCP / IP protocol currently used as an Internet protocol, the communication layer is divided into an application layer, a transport layer, a network layer, a data link layer, and a physical layer. Other protocols for home appliances or factory automation also include an application layer, a data link layer, and a physical layer, and additionally, a transport layer or a network layer. However, since the present invention is intended to correspond to the above-described communication method for home appliances of low specification, it has a communication layer consisting of only a physical layer, a data link layer, and an application layer, and loads of a microcomputer to fit into a master-slave method and a half duplex method. Minimize the physical and data link layers as much as possible and allocate a large portion of the product's operation to the application layer.

Referring to the detailed configuration of the above-described communication layer, as shown in FIG. 9, first, in the case of a slave, an application layer consisting of application software, message configuration, message execution, message combination, and message duplication check area, packet configuration, and packet transmission And a data link layer consisting of a CSMA / CD, a packet check, an address check, and a packet data receiving area, and a physical layer consisting of a UART. In this case, the physical layer may include an adapter selected when power line communication is used.

In the case of the master, the application layer consists of application software, message composition, message partitioning, and message combination areas, and data link consisting of packet configuration, packet transmission, packet transmission confirmation, CSMA / CD, packet check, address check, and packet data reception area. It consists of a layer and a physical layer consisting of a UART. The physical layer of the master may include an adapter that is selected when using power line communication like the slave.

At this time, the physical layer is configured to receive the bit signals on the communication line to form a packet or to send the packet received from the data link layer as a bit signal on the communication line.

The data link link layer configures a packet using data received from the application layer and transmits the packet to the physical layer, or processes the packet received from the physical layer and hands it over to the application layer. The data link layer has a slightly different role between the master and the slave. The master includes a process for ensuring packet transmission to the physical layer. The physical layer and the data connection layer have the same structure as the product acting as a slave.

The application layer consists of a message set and is responsible for interpreting and processing messages. In the product acting as a slave, the message includes methods such as load control and memory control, and in the master, the slave manages the slave or controls the entire network by using the result of processing the message. The application layer therefore contains different content for different products. The application layer of the master also includes a role of a transport layer that divides and transmits a packet when there is more data to transmit than one packet range, or combines and processes a packet when the divided packet is received. The distinction between transport layers is included in the application layer because the splitting and combining of packets is done only for a specific message and not all messages.

Such a packet communication structure between communication layers is shown in FIG. In other words, the application layer and the data connection layer are interfaced in a message unit, and the data connection layer and physical layer are interfaced in a full packet unit. When communicating between the data connection layer and the physical layer, communication is performed in full packet units so that the packet can be used as it is without having to configure packets for each layer. However, since the length of the header to be added in the data connection layer and the physical layer is not known at the application layer, it is difficult to interface in the full packet unit. Instead, data to be transferred from the application layer to the data connection layer is transmitted in message units. The packet should contain all the information you want to communicate with, and it should be structured to cope with future extended communication.

As shown in FIG. 11, a packet basically includes a header area consisting of a packet header field and a field (field) for future packet function addition, a body consisting of a message header field, a future message function addition field, and a message field. It consists of a zone and a trailer zone.

The detailed configuration of the request / notification packet used in the master of the packet is made up of 8 bits, as shown in FIG. 12, for a home code (HC) for identifying a home on which a network is formed. A receiver address (RA) consisting of 16 bits to indicate the receiver, a sender address (SA) consisting of 16 bits to indicate the sender, and a packet length to indicate the length of the packet consisting of 8 bits. Length (PL), Access Priority (AP) consisting of 3 bits to indicate transmission priority, Packet Header Length (PHL) consisting of 5 bits to indicate the length of the packet header, Protocol Version (PV) for indicating the version of the protocol, consisting of 8 bits, Packet Type (packe) for indicating the packet type, consisting of 4 bits t Type) (PT), a Retransmission Counter (RC) for indicating the number of retransmissions consisting of 2 bits, and a Packet Number (PN) for indicating a new packet transmission, consisting of 2 bits, 8 bits Home appliance network control system comprising a Message Length (ML) of, Message Header Length (MHL) of 8 bits, Port (PO) of 8 bits. An 8-bit Command Code (CC) and Argument (ARG) with variable number of bits, 16-bit error check (CRC), and 8-bit ETIX (ETX) to mark the end of the packet. It consists of minimum 17-byte and maximum 255-Byte.

In addition, the response packet used by the slave is the same as the request / notification packet except that an 8-bit ACK / NAK is included in the body part as shown in FIG. 13.

In this case, the home code (HC) is a code for logically classifying households configured with a network, and is used within the range of nucleus values 0x03 to 0xFE to distinguish each household, especially when a transmission line between households such as a power line is connected without distinction. do.

The receiver address RA is located in front of the sander address SA in order to early determine whether the receiver will continue to receive or ignore the packet when it begins to receive the packet. In order from the upper bit, 2Bit classifies the network type, 6Bit classifies products with independent functions such as washing machine and refrigerator, and the lower 8Bit class is allocated to distinguish each other when there are many products of the same type.

The packet length (PL) is composed of 1-byte that stores the number of bytes calculated from the home code to ETX. After receiving the packet length (PL) data value, the receiver receives only this much data and processes subsequent steps. Therefore, the packet length (PL) value is used to know in advance the size of the buffer required for reception, and is used to detect errors in the received packet data. That is, after reading the last byte of the packet, if the value is not ETX, it can be determined as a packet error.

Access Priority (AP) prioritizes messages that need to be prioritized, such as packets that need to send urgent messages at the application layer, packets that have failed to be transmitted, or need to be retransmitted, or are less important than general communication. Field indicating the priority of transmission to perform / CD function. This field is meaningful when the adapter performing the CSMA / CD function can transmit according to priority. Otherwise, this field is ignored. The access priority values according to each communication case are as follows.

0: Retransmission due to collision or emergency

1: When sending bulk data through message segmentation

2: normal communication

3: Report network connection status (in case of conflict, Priority is kept at 3)

The packet header length (PHL) is a field for scalability of the packet header. When the extension field is added to the current packet header, the packet header length (PHL) is changed accordingly. If there is no change, it is 9 bytes and can be extended up to 32 bytes.

The protocol version (PV) is a 1 byte field indicating the version of the adopted protocol. Versions and subversions take values from 0 to 15 in order in which they are updated.

The packet type (PT) is a 4Bit field value specified when transmitting a packet, and is divided into a request packet, a response packet, and a notification packet. The response packet is divided into a successful response and a failed response. The master should be set to the request packet and the slave to the response packet. When the device operates only as a slave, it should only process request packets. The reason for splitting the response packet into two is to see only the packet type (PT) field in the packet header even if the data link layer does not know the message contents, and to retransmit the packet immediately without sending the packet to the application layer when the failed response is failed. Notification packets indicate packets that do not require a response. An arrayed packet is used to send a large amount of data at high speed when transferring all the data without sending a response to each packet. The nucleus value is as follows.

0: request packet

1 to 3: Reserved

4: Successful response packet

5: Failed response packet

6 to 7: Reserved

8: Broadcast Notification packet

9: Arrayed packet

10: End packet of arrayed data

11-15: Reserved

The retransmission count RC is a 2Bit field value for preventing duplication of the same message when a communication error occurs. The master may retransmit up to three times if the received response packet contains a CRC error code or if the received packet is a CRC error or a received byte time over. The slave is always one transmission.

The packet number PN is also a 2Bit field value for preventing duplication of the same message when a communication error occurs. The master increments the packet number by 1 each time a new packet is transmitted, and maintains the same packet number when retransmitting the same packet. Therefore, the slave memorizes the packet number and the transmission address of the previous message, and if the same message is received again, the slave ignores the duplicate message and performs the difference if it is different. When the slave responds to the received message, the slave constructs a response packet by copying the packet number of the received message.

The message length (ML) is information for knowing this because the length of the message field is variable. Therefore, the application layer handles the length of the message field by looking at the message length.

The message header length (MHL) is a field for future expansion of the message field and may add a message header in the case of encryption of the message field, change of an application protocol, and the like.

The port number (PO) is a field for the extension of the message set, which can classify the message set by port. Message sets can be placed per port for version-up of message sets or compatibility with other application protocols.

The message consists of command code to request the master to perform a function, input parameters required to execute this command, and parameters sent to the master after the slave executes the command. In addition, it should be configured and defined to facilitate program work in 8-bit microcomputer. In other words, modular programming should be possible to easily reflect the message even if it is corrected. To do this, every message must have an independent function for each. This means that every message cannot contain a subordinate message, and there is no causal relationship between the routines in the S / W implementation. If the messages have functions that are independent of each other, the messages can be combined to extend the functions for controlling and monitoring the product. If the slave executes the command normally, the arguments sent to the master are {ACK + Return arguments}, otherwise it is {NAK + error code}. Each device can have up to 256 Commands and is included. Input Arguments and Return Arguments are determined according to the command code and the number of bytes.

The data format of Arguments is as follows.

boolean: 1Byte

char, unsigned char: 1Byte

int, unsigned int, short int, unsigned short int: 2 Byte

long, undigned long: 4Byte

string: Send and receive including NULL.

In addition, the following basic concepts are used to classify the Command code.

All products use 256 command codes (0x00 ~ 0xFF) independently, but the commands commonly used in all products use common code. Incorporate the product's functions into a generalized structure to facilitate item addition. All Command codes are classified as mandatory and optional. Essential are the basic information of the device and the commands necessary for communication. Functionally, it is divided into instantaneous command (indicated by I) and program command (indicated by P). Instant Command can be executed immediately by Slave and Program Command needs Sequence to execute. Command codes in the Algorithm Area are not specified by standard codes for all products. Since the same type of product can communicate using different algorithms depending on the model name or product manufacturer, different functions can be performed using the same command code. Therefore, all Comand codes in this area should be given as unique numbers of product models and included as arguments. In addition, the message is a protocol for processing at the application layer. The structure of the message is different from that transmitted from the master and when the slave responds to the master. When transmitting from the master, it consists of a command code and an input argument (ARG) to perform it. The number of arguments and their data types depend on the Command code. When a slave receives one packet from the master, the message structure is divided into two types, one with an error and the other with no error. If there is no error in the packet received from the master and the command code is normally executed, the message structure consists of the command code, the ACK, and the result of executing the command code (ARG). The number and data types of the result arguments also depend on the Command code. If an error occurs in a packet received from the master, it consists of a command code, a NAK, and a packet error code. If the packet is normal but an error occurred during the execution of the command code, it consists of a command code, a NAK, and an error code.

CRC is a value that detects an error in a received packet or allows a receiver to detect a packet error in transmission. CRC is processed by 16-Bit, and its value is generated by using data from ETX to byte just before CRC field or error is detected.

ETX (0x03) is a communication character indicating the end of a packet, and provides a method of detecting a packet error without using a CRC with a packet length field at the time of reception. That is, when receiving the byte length of the packet length, if the last byte is not ETX can be determined as a packet error. In this case, the packet error check using the CRC may be omitted.

The address in the configuration of the above-described packet will be described in detail as follows.

Each master and slave on the network are recognized as one address. Each system has an assigned address of 2 bytes and can transmit a packet to the corresponding party through the assigned address. As shown in FIG. 14, an address is composed of a 3-bit network code and a 5-bit product code, which distinguishes a physical address and a plurality of products for each product that are not determined and cannot be changed at the time of product shipment, and are used as a group address or a communication. It is divided into 8-bit logical addresses that can be changed by

Networks in the home can be divided into PC series, AV series, Living, that is, household appliances, etc. A 3-bit network code is used to distinguish these networks. Although the communication standard is different from that of the network, a network classification field is required for communication with the PC series and AV series products. An example of classification of network codes is shown in FIG.

Allocate 5-Bit for the product code, namely product name (washer, refrigerator, health appliance, lamp, security, etc.) and 8-Bit to distinguish multiple products for each product. This takes into consideration conditions such as inns and hotels.

The field for distinguishing multiple 8-bit products is also used as a group address divided according to the installation location of the product. The installation location of the product is entered when the user registers the product on the network in the network manager.

At this time, there are two ways to set the group of products. The first group address can be set to all targets corresponding to the lower fields by filling each field with '1'. At this time, the group means the same product or a product belonging to the same network. For example, if the value of the network code is '111', it indicates all the networks in the house. If the product code value is '11111', it means all the products in the network. As shown in FIG. 16, a predetermined product group can be selected according to the group address. If the value is '11111111', it means all products corresponding to the corresponding network and product code. As the second Group address, as shown in FIG. 17, a product group at a predetermined installation location may be selected according to the group address. At this time, the product code is '11111' to designate all products and the logical address value is designated according to the installation location. If the network code is '111' and the product code is '11111', the Logical Address field is a field indicating a place code. In addition, as shown in Fig. 18, a group address may be designated to select a predetermined product group at a predetermined installation location.

Next, a plug and play process, that is, a process of assigning an initial home code and address by connecting a device to a network, and setting a communication environment will be described.

In order for devices connected to the network to communicate, all devices must be set to the same communication speed, assigned a unique address within the physical network, and the master must have a database of names and addresses of all devices. In addition, each transmission line should be distinguished when the transmission lines between the homes are not separated, such as a power line. To do this, the network manager in charge of Plug and Play first sets a home code for distinguishing each home at the initial power-on. After setting the home code, the network manager goes through the Discovery & Addressing step of receiving information about the device from the user and assigning an address to the device when the device is first connected to the network. After this process, a pre-request step is required to change product information such as the model name or controller version of the device to be communicated, the size of the buffer for packet construction, or the communication speed. The pre-request step is not always necessary and is performed as needed. Product information is needed to know the name of the connected product, and a buffer size request is used to determine the length of a packet to send a large amount of data from the master to the slave. The rate change request is requested from the master to the slave when sending or receiving a large amount of data. When the pre-request phase ends, it enters the normal communication mode. The plug-and-play concept from the point of view of the user should be readily available when the device is powered up without any additional installation or input from the user. However, after connecting the device to the network, the network manager prompts the user for information about the device. The reason for this is to consider the power line as the media of the network. In the case of the power line, when the device is connected to the network, the transmission line between neighboring homes is not distinguished. There is a problem that can be assigned. Therefore, first, device information is input to the network manager so that the network manager requests an address registration from the device.

Referring to the above-described home code setting process, the network manager transmits a home code confirmation message using the reception address as all devices to set a unique home code that can be distinguished from other homes at the initial power-on. At this time, the argument is a value generated as an arbitrary home code within the range of 0x03 to 0xFF. If there is no response, the home code is set to its own home code in the physical network to which the network manager is connected. If there is a response, the home code is not the only value. . The home code set in this way is set to each product at the same time when the network manager confirms each product's address, so that the products in the same home can use the same home code to distinguish between homes. However, address-related information, including the home code, is stored in the nonvolatile memory of each product, which can cause problems if you move after assigning each product's address. In other words, if another home connected to the same physical network as the moving home uses the same home code, there is a problem in that the home code is not divided by the home code because the home code is collided between homes. Therefore, at the time of moving, all the products are turned off, and only the network manager is inputted to repeat the process of checking the home code conflict when setting the initial home code. If there is no conflict, you can use it without changing the home code. If there is a conflict, reset all the product's home code and reset the home code. In this case, the user ID is used as a parameter to distinguish the home product from the conflicting home code. The user ID is entered through the network manager when the device is first connected to the network.

Next, in the discovery and addressing process, a user inputs a product name, number of products, installation location, and user ID of a device connected to a network manager capable of managing the network when the device is first connected to the network. The network manager then requests registration from the input device, that is, the device connected to the network for the first time, and assigns an address when receiving a temporary address registration message from the device. At this time, even if there are plural products of the same type, a new address is designated so that the address does not conflict. The reason for receiving the number of products is to know whether registration messages are received from all connected products when several products of the same type are connected to the network at the same time. In addition, the reason for receiving the installation place is to allow the user to easily identify the product through the location information of the product appearing on the screen of the network manager. The user ID is to identify the product at the time of home code reset as described above. The address of the network manager is fixed to 0x00 regardless of power on / off. However, other devices have the product address assigned to the factory when the power is off, regardless of the master and slave. The network manager sets the area of free address that can be selected by the products by using 'Join request' command and registers as a temporary address to unaddressed devices (device designated as the representative address for each product) entered by the user. To ask. In case of air conditioner, select 0x20 as receiver address and spare address as 0x21 ~ 0x2E. 'Join request' Command can recognize only those products whose address is not confirmed. Called products operate random number generator and randomly select among the set range of address, set it as own address (temporary address) and inform the network manager of the value. If 0x25 is selected among 0x21 ~ 0x2E, 0x25 is sent to the network manager. The temporary address remains as the address of the product until the network manager or master changes it using the 'Address Change' command or turns off the product. Network manager uses 'Address Change' command to reset the duplicated address to the representative address, and the address of the products that have selected the temporary address that does not overlap is determined in order from the free address and recalls the relevant products. At this time, the network manager's own home code and the user ID entered by the user are also transmitted to set the home code and user ID of the corresponding products. If non-overlapping temporary addresses are 0x2A, 0x25, 0x23, and free addresses are 0x21 ~ 0x2E, the product with temporary address 0x23 is 0x21, 0x25 is 0x252, 0x2A is 0x23. If there are duplicate temporary addresses, repeat the above process. The above discovery and addressing process is repeated 14 times for all product types (0x01 to 0x0F).

As described above, after the discovery and addressing process, the names, addresses, and installation locations of the devices are stored in a database of the network manager. Devices connected to the network manager start to operate as masters or slaves after addressing by the network manager is completed. At this time, the device acting as the master reads the device name and address database from the network manager, stores it in its memory, and starts communication with the slaves. All devices must act as slaves before address assignment. Devices assigned an address periodically broadcast a message indicating their presence. This is because masters that can control each device must know that each device is connected to the network. If there is no distinction between the connected device and the non-connected device, it may be difficult to distinguish whether the communication failure is due to a power off or a device failure. In addition, devices that are not connected to the network manager screen are displayed by being deactivated so that the user can know them. Every device assigned an address will periodically broadcast its presence in an Alive Message, which needs to be adjusted. It doesn't matter if the number of devices is small, but if the period is too short for too many devices, too many alive messages are sent in the network, which degrades the performance of the network. Set the initial period long and set it in proportion to the number of devices. In other words, when the number of devices is small, the period is small, and when the number is large, the network performance is maintained by holding it large. At this time, it is the network manager that sets the period, and each device broadcasts its own period as a factor of the live message. The network manager receives the alive message of each device and sets the period of the device as its own period if it is different from the period set by the network manager.

Next, an event that generates a communication condition of the device will be described.

An event refers to a case where a state change occurs in a device, and may be classified into the following five types according to the cause of occurrence. First, user events that occur when a user commands a device directly through a key, etc., Periodic events that occur automatically at regular intervals (for example, Alive Notification messages are sent from the network manager at regular intervals), and the status of the device is monitored. Status events (e.g. changes in temperature, humidity, laundry administration, etc.) due to voluntary status changes, error events that occur when there are errors related to operation in the system, or external events that occur when requested from outside the system, such as a web server. There is a request for communication from a remote location when the network manager acts as a home server.

Devices equipped with user interfaces such as keyboards, mice, and monitors can be peer-to-peer communication with all devices by all five events, but devices lacking a user interface can be peer-to-peer communication. In order to perform the function, it is possible to set the conditions for communication in advance or only by the events (User event, Status event, Error event) generated by internal factors of the device.

In the present invention, when an event occurs, it performs a communication function by notifying all devices of its own state change by a notification packet. The reason why an event is important is that when a user monitors the device's status, when the device's status value changes, rather than requesting a status value each time it wants to know the device's status, the device will notify itself of the status change. Because it is much more efficient. In addition, in the event of a device failure or error, it should be notified immediately when it occurs, so it is necessary to notify the status change immediately when an event occurs.

Events implemented in each product are represented by 1 byte event code, and there are two types of events, common events and private events implemented by products, according to the implementation method common to all devices. The classification of codes uses area classification rather than field classification because the number of items that can be added in the future may vary.

Common events can be further divided into event-related events and error event areas common to all devices. Events related to user operation are events such as key and dial input, door opening and closing, load input, etc., and have event code values of 0x00 to 0x2F. The common event is an event such as a door open during operation and has an event code value of 0x30 to 0x4F. Private events can be classified into error events by product and status events by product. The error event for each product is the event code value of 0x70 ~ 0xAF, which indicates the error state or failure state inherent to each product, and the operation status event for each product is the event that occurs whenever the state changes during product operation. It has an event code value.

Event codes can be categorized into mandatory event codes (e.g., fault event codes) that must be implemented by all devices and event codes that can be optionally implemented. The required event code basically broadcasts the status change in a notification packet when an event occurs so that other devices can monitor the status change. At this time, the command code used is an event notification command and has an event code (1 byte) and a state value (4 bytes) of the event code. When an event occurs in one device, it can simply broadcast a notification packet to notify the fact, but can also command the operation of another device. For example, when the washing process of the washing machine is finished, the porch light may be turned on or a text message may be displayed on the air conditioner. To do this, each device should store the information of the target address to communicate with and the operation command code and arguments of the communication target whenever an event code occurs. However, since such information varies according to the user's taste, the user must set communication conditions when an event occurs in each device through the network manager. In this case, in case of broadcasting not only for commanding the operation of another device but also for notifying the event, the user should set the case. Another thing to consider when setting the event communication conditions. This is the time interval between event notifications. When the temperature state or the sensor state, etc., which set up event communication change rapidly, the network has too many such event communication packets, which may degrade the network performance. Therefore, a minimum event interval is required. Therefore, when a user sets a communication condition when an event occurs through a network manager, an event code, a communication counterpart device, a minimum event occurrence interval time, and a communication message should be specified.

The event communication condition is stored in the nonvolatile memory of each device as an event file consisting of the header of FIG. 19 and the body of FIG. 20. In case of resetting, user can set it again in the network manager.In the initial setting, first check whether the size of the current device's non-volatile memory is enough (Event Code Buffer Size Read Command). If there is room, set it. In addition, since the size of the nonvolatile memory is limited, information on unnecessary event communication conditions should be deleted. There is also a message for this purpose and the user can use it to delete unnecessary event communication conditions (event code Delete Command code).

To execute the event code, first read the total number of events (total_event_no) and event codes that need to be executed from the header of the event file stored in nonvolatile memory. In the system main program, when the state of the defined variable changes during operation, it is saved in the relevant memory. The event code execution routine determines whether to execute by comparing the system state value with the event code read from the event file. When the state of various systems changes at the same time, only one event code is executed at a time to prevent the microcomputer from occupying the resources of the microcomputer.

Next, a method of controlling various errors will be described.

Communication errors include data bit errors due to noise on communication lines, errors when communication frequencies are different, data bit errors due to data collisions, errors due to attenuation of transmission signals when impedances between lines and devices are not matched, and data bits. There is no error, but it is divided into an error due to transmission and reception of data that the receiving device cannot process. The noise in the communication line may cause a receiving physical layer, that is, a UART frame error, or change a data value. If the communication frequencies of the transmitting and receiving devices are different, most of them generate a UART frame error on the receiving side. Even when multiple devices transmit simultaneously, most receive UART frame errors on the receiving side. If the impedance between the line and the device is not correct, the receiver will not receive any signal.

When the master sends a request packet to the slave, the slave receives the packet and detects the defined errors. When the slave detects an error in the received data bits, it sends a response packet as shown in FIG. 21 including the code value for the detected error to the sender, and the master functions for retransmission or error processing according to the error code. Do this.

At this time, the error code is composed of 1 byte, and divided into common error code assigned to the area (0x00 ~ 0x9F) commonly used by all devices and fault code assigned to the area (0xA0 ~ 0xFF) independently used for each device. Lose. Common error codes are values for communication error, and fault codes are values for diagnosing the inherent functions of devices such as sensors apart from communication functions. They can have 96 items per device, including packet error, receiver error, It can be divided into Bad command, Illegal arguments, Illegal access, and fault code. Details are as follows.

First, a packet error is subdivided into a CRC error of a received packet, a 1 byte reception time over, and a response waiting time over.

The CRC error of the received packet occurs when the calculated CRC value and the CRC value included in the received packet are different in both the master and the slave. If a CRC error occurs in the response packet received from the slave, the response packet including the CRC error value is transmitted to the master. When the master receives this response packet, it resends up to three times. When a CRC error occurs in a packet received from the master, it is retransmitted up to three times.

1 Byte Receive Time Over error occurs when the time interval between Byte Receive is out of 2BTU (minimum time 3mSec: 9600bps) due to noise or other causes on communication line. In this case, however, the receiver side must receive up to the Packet Length field. If the Byte time interval exceeds the specified value before receiving the Packet Length field, the received data is ignored. If a one-byte receive time-out error occurs, the receiver stops receiving and fills the rest of the receive packet buffer with zeros and forwards it to the upper layer. Eventually, the receiver will receive a CRC error. When the master sends a request packet to the slave, if an error occurs in the slave due to noise on the line, a response packet including a CRC error is transmitted to the master. If the master has successfully received the response packet, it will resend up to three times. Slave normally receives request packet and transmits normal response packet. However, even if 1 byte receive timer error occurs when master receives, the master retransmits up to 3 times. Byte Receive Time Over error may occur in both master and slave, but it is an error code required between each communication layer and does not give information of 1 Byte Receive Time Over error to sender. In other words, the Byte Receive Time Over error is for internal processing and is not transmitted between master and slave. The timeout of the receive time over is related to the busy check. Since the present invention transmits and receives data in packet units, the time interval between byte transmissions should be shortened to inform that the bus is occupied at the time of busy check by another device. If the time-out of the receive time over is large, a time delay may appear during the transmission of one packet.In this case, when another device attempting to transmit has a busy check, the bus determines that it is idle and starts transmitting immediately. This happens.

A Response Waiting Time Over error occurs when the master sends a request packet and no data is received. That is, it occurs when there is no communication target slave. The master's physical layer waits for up to 5 seconds and if no data is received, it creates a packet containing a Response Waiting Time Over code and sends it to the data link layer. In the data link layer, the message is sent to the application layer, and it can be seen that the device does not exist in the application layer.

Next, the receiver error is subdivided into insufficient memory, communication reject, remote control rejection, protocol version unmatch, and message port unmatch.

An out of memory error occurs when the master does not have enough memory to write the specified data when sending a command code such as memory write, LCD write, or EEPROM write.

A communication reject error occurs when the slave receives a request packet from the master normally but tries to perform other functions other than communication. Master receiving communication reject error can retry after minimum 5 seconds (master's maximum waiting time).

The remote control rejection error occurs when the slave receives a command message for control while the slave is not in the state of remote control.

Next, the bad commands are subdivided into commands that can't be executed and arguments that can't.

Unable to execute a command code error occurs when the slave normally receives a request packet from the master but contains a command code that the slave cannot execute.

Unable to perform an argument value error occurs when an executable Comamnd code has been received and the scope of the argument has also been set within the defined range but the value cannot be performed on the device. For example, in a load On / Off? Command of a microwave oven, for example, a hood fan is an input value, but in a microwave oven without a hood fan, a parameter value error that cannot be performed occurs.

Next, Illegal Arguments are subdivided into defined and different number argument errors and over range errors.

A number argument that is different from the defined number error occurs when the slave normally receives a request packet from the master, but the number of input arguments to execute the command code is different from the number specified in the message set. The number of arguments is the number of bytes. If the input argument is a variable defined as an unsigned int, the number of input arguments is 2 because it consists of 2-bytes.

An Over Range error occurs when the slave normally receives a request packet from the master but the input parameter value for executing the command code is out of the range specified in the message set.

Next, the lllegal Access error due to the prohibited action detection occurs when the slave normally receives the request packet from the master, but the input argument value for executing the command code specifies the prohibited memory area or the control prohibited load.

Next, the fault code will be described.

All devices have their own functions other than communication and can remotely diagnose if there are any problems with those functions. At this time, when there is an error in the function of the device, the value sent in the response packet is the fault code. For example, if a request packet containing a command code for reading a temperature sensor value is received from a slave, and the sensor is determined to be faulty, the fault code value is included in the response packet along with 'NAK' and sent to the master. The fault code is assigned by all devices using a common area.

The present invention treats all cases of abnormality in the bits of the data constituting the packet, namely, receiver address error / sender address error / transmit / receive address error / packet length error, as CRC errors. Is as follows.

First, the receiver address error will be described.

If an error occurs in a bit in the receiver address field, the device that is not calling will receive it. At this time, the received device detects a CRC error due to an error in a bit of a receiver address field. In the first case, the error packet was sent by the master (A) as a request packet to the slave (A), and if it was received by another slave (B), the slave (B) mastered the response packet containing the CRC error value. Send to The master (A) receiving the response packet from the slave (B) responded by a device other than the slave (A) originally intended to call, but the slave (A) attempting to call the original response was ignored while ignoring the sender address. Count as one. That is, one packet received after transmitting one packet is regarded as a response packet from the device to be called. Receiving the response packet, the master (A) retransmits to the slave (A) up to three times. In the second case, if the other slave (B) receives a response packet sent by the slave (A) to the master (A) for a request packet sent by the master (A) to the slave (A), the slave is received. (B) transmits a response packet including the CRC error value to the slave (A). At this time, if there is no packet error, the slave (A) knows that it is a response packet in the packet type field value and ignores the received packet. Master (A) continues to wait for a response packet from slave (A) for up to 10 seconds. After 10 seconds, the communication started from the request packet sent by the master (A) to the slave (A) is terminated without executing any command code on the slave side.

Next, the sender address error will be described.

When an error occurs in a bit of the sender's address field, the device the sender wants to call receives. However, a CRC error is detected due to an error in the bit of the sender's address field. In the first case, if an error occurs in the sender address field when the master (A) sends a request packet to the slave (A), the slave (A) receiving the packet is the device of the address value in the sender address field (master or slave). The response packet is sent. If the response packet is transmitted without error and the other slave (B) receives it, the slave (B) ignores the received packet because it knows that it is a response packet in the packet type field value. If the response packet is received by another master B, the master B ignores the received packet according to the principle of one packet transmission-one packet reception because the master B has not sent a request packet. Master (A) waits for up to 10 seconds for a response packet from slave (A). After 10 seconds, communication is terminated without executing any command code on the slave side. Second, the master A normally transmits a request packet to the slave A, but the slave A receives an error in the response packet sender address field to the master A. If the response packet is transmitted without error and received by another slave (B), the slave (B) ignores the received packet because it knows that it is a response packet in the packet type field value. If the response packet is received by another master B, the master B ignores the received packet according to the principle of one packet transmission-one packet reception because the master B has not sent a request packet. Master (A) waits for up to 10 seconds for a response packet from slave (A). After 10 seconds, communication is terminated without executing any command code on the slave side.

Next, transmission / reception address errors will be described.

If an error occurs in the bits in the receiver address field and the sender address field, another device that the sender does not want to call will receive a CRC error. In this case, communication between the devices is completed in the same sequence as the recipient address error and the sender address error, and then terminates.

Finally, the packet length error will be described.

The receiver constructs the received packet buffer by using the number of bytes corresponding to the value of the packet length field. First, if the value of the packet length field is larger than the actual value, the receiver has received up to the last byte of the actual packe but continues to wait for data. When no more bytes are received and the time limit between bytes passes, a Time Over error occurs and a CRC error occurs because the receiver fills the rest of the received packet buffer with random data. Therefore, the master retransmits up to three times. Second, even if the value of the packet length field is smaller than the actual value, the receiver detects a CRC error.

Home appliance network control system according to the present invention has the following effects.

First, the present invention is a communication system of the master-slave, 1-cycle and half-duplex method, and serial communication of low-performance microcomputers used in home appliances by applying a simplified and standardized protocol. Implementing a network using features enables a low cost optimized home appliance network.

Second, the present invention is to maximize the user's convenience, so that the user can be informed of the change in the operation state of any one of the home appliances connected to the network through the other devices on the network and the user can select the communication conditions and target devices, etc. Can be.

Third, since the present invention can use a power line as a communication medium, a network connection is possible by connecting a power plug of a home appliance to an outlet without a separate operation.

Claims (1)

A plurality of home appliances having a serial communication function; It comprises a communication medium for connecting the home appliances, A communication layer consisting of an application layer, a data link layer, and a physical layer is defined for communication between the home appliances, and when a predetermined communication event occurs, communication is performed at the home appliance using the communication layer. Home appliance network control system.