Banner

Sponsors

A+ R A-

Industrial Networks: access methods and error control

E-mail Print

The interface and bus access method, handle the digital signal can be transported safely between a transmitter and receiver. These tasks are called access control (MAC: Media Access Control) and are part of network administration. This is partially or fully installed in each bus device. The network management controls the exchange of data between different levels of processing within a field device. Some of its most important tasks are:

• Maintain an active list and message system: at least one device should be checked at regular intervals other devices are present on the bus and pass this information to their peers. All devices must know when a message is sent.

• Processing of the signal to be transmitted: this includes receiving the signal from the application program, addition of the transmitter and receiver addresses, the determination of the parity bits and error checking (checksum), packaged in a pattern or frame (adding data management), transfer to the interface and sending the message to the application program that the signal is coming.

• Interpretation of the signals received: this includes the reception on the interface, unpack it to the plot, verification of sender and receiver addresses, identification and comparison of parity bits and error checking, dispatch of ia message confirming proper reception of the message or the message transfer to the application program.

• Interpretation of the application program commands: the command transmitter por.ejemplo ASCII programming language or higher level is coded in a way they understand the application program.

Figure: Architecture of Control with MAC computers only

And then ... As the signal is transmitted?

The way these functions are performed depends on the type and shape of the network and the type of devices that are connected to it. For example, a computer and printer connected to each point to point in a star-like structure through an RS-232-C, they communicate their availability for the transmit and receive control signals that pass through a printer stranded wire. The data are invariably transmitted in block and control characters tell one if you have to wait for more data or close the connection.

For a network with only two bus lines, some of these tasks are performed by the medium access control (MAC). This ensures that only one message travels on the bus at any given time. The message itself contains the control characters or a bit pattern that signals the receive, the directions of transmitter and receiver as well as current information to be transferred. This, the so-called data link layer performs the task of signal processing, interpretation and recognition. The application layer initiates all transfer activities with the appropriate commands.

No interest if using control lines or control characters, it should be a check on whether the information arrives intact. The result of this test is packaged together with confirmation that the signal has arrived. If the test fails, or has sent a negative acknowledgment, the message must be transmitted again. It is also the same case if there is no confirmation within a given time period (time-out). After two or three failed attempts to establish communication, the process is broken and the routing device, marked as out of service.

The error control: analysis

During data transmission noises often appear in the transmission line, which distort the transmitted signal. These noises can be generated by electrical interference, thermal noise, etc.. The capacity of the physical environment to allow the bit stream without causing alterations in the message is known as noise immunity. There are different methods for detection and correction of errors, the ability of a method for detecting and correcting errors in a message is measured by their Hamming distance. As the grouping of bits, error recognition also has two levels: level one characters and a plot level. For the first case we have the parity bit (check character level), parity bit and longitudinal and transverse the second so-called cyclic redundancy.

PARITY BIT

It is a simple method of error detection level characters. Each character consists of a start bit, 5 to 8 data bits, a parity bit and one or two bits of completion. The parity bit serves as a check of the transmitted character. Its value is allocated by the issuer so that the number of ones in the character plus the parity bit is even (even parity) or odd (odd parity). The recipient receives the character, calculates its parity bit and compares the parity bit transmitted with the calculated, verifying the correctness of the received character.

CROSS PARITY BIT

It evolved from the former in which the bits are grouped into a block. It calculates the parity bit different columns and rows, adding the block resulting row and column. Then the entire block is transmitted. The receiver can identify errors received from the block.

Cyclic redundancy code or CRC

It is a method widely available, used in many protocols. From an algorithm that uses a generator polynomial and ios bits that form the fabric, it calculates a number called cyclic redundancy check (CRC) which is added to the end of the frame and transmitted to it. The receiver calculates the CRC using the same generator polynomial and the received bits. Then compare the CRC received with the plot and calculated. For any of the above methods, may render an action in case of detection of an error in the message. This will depend on the specific implementation of the protocol and the type of error. However, while the protocols implemented in general screening algorithms that allow correction of some errors, such as the CRC, it is usual not to use this possibility.

The CRC code contains redundant data with the plot, so that errors can not only detect but also can be solved.

A final concept: OPEN AND CLOSED SYSTEMS

When the communication system is homogeneous, ie the instruments involved are from a single manufacturer, the problems can usually be resolved. For applications with equipment from different backgrounds, the problems multiply. Proprietary systems that work with their own protocols, character sets, sequencing of transmission, etc.. Networks are called "closed". Most were developed at a time when the integration of the instrumentation from different manufacturers was not important and not performed. In contrast, the networks "open" meets specific guidelines are available to all. This allows other manufacturers to produce both compatible equipment, which can be easily and directly integrated into the system without problems, as well as interfaces for integration indirectly.

In the next article we will focus on the most widely used model in the industry, the OSI model applied to the industrial world.

Banner

Enlaces Externos

EMPIEZA EL CURSO AHORA GRATIS!!!

Curso Automatizacíon en Videos GRATIS: Más de 14 horas de VIDEO!!

Curso Casi GRATIS!!

Login

Registro

*
*
*
*
*

* Campo Requerido