ATM Tutorial: data link layer

[Guide] The ATM physical layer generally includes the OSI physical layer and the data link layer. features such as the physical media of the OSI physical layer determine the child layer and the transmission collection (TC) child layer with the same functions as the data link. There are no special physical layer features for ATM. On the contrary, SONET, FDDI, and other transmission systems are used to transport ATM cells. Therefore, we will focus on the data link function of the TC sub-layer.

I. data link layer in ATM

The ATM physical layer generally includes the OSI physical layer and the data link layer. features such as the physical media of the OSI physical layer determine the child layer and the transmission collection (TC) child layer with the same functions as the data link. There are no special physical layer features for ATM. On the contrary, SONET, FDDI, and other transmission systems are used to transport ATM cells. Therefore, we will focus on the data link function of the TC sub-layer.

When an application generates a message to be sent, the message enters the transmission line and passes down to the ATM protocol stack, adds the header and tail, and puts the segment into the ATM Cell. Finally, these cells are transmitted to the TC sub-layer. Let's take a look at what happened on the road.

Ii. Cell Transmission

The first step is to perform the header checksum. Each cell has a 5-byte header, which contains 4-byte virtual circuit and control information and 1-byte checksum. The Checksum only contains the first four header bytes, without occupying the paybyte. It is composed of the remainder of 32 head parts divided by polynomial x ^ 8 + x ^ 2 + x + 1. The Checksum is added with a constant of 01010101.

The decision to verify only the header is made to reduce the possibility of incorrectly passing cells due to header errors, and to avoid verification of a much larger payload field at the beginning of verification. If you really need to verify the payload field, you need to go to a higher layer to complete this function. Because the checksum field is only in the header, the eight-digit checksum field is called the header error control HEC (headererrorcontrol ).

Once HEC is generated and the cell header is inserted, the cell is ready for sending. Transmission methods are divided into two groups: asynchronous and synchronous. When the asynchronous method is used, you only need to prepare to send it, and there is no time limit.

When the synchronization mode is used, the cells must be sent according to the specified time cycle. If no data cells are available when necessary, a TC sub-layer must be invented. Such cells are called idle cells (idlecell ).

Another type without data cells is to operate and maintain OAM (operationandmaitenance) cells. The ATM mechanism also uses OAM cells to exchange control and other necessary information to ensure system operation. Matching the ATM output rate with the rate of the transmission system is an important task of the TC sub-layer.

In the receiver, idle cells are processed in the TC sub-layer, but the OAM cells are handed over to the ATM layer.

Another important task of the TC sub-layer is to generate frame information for the transmission system if any. For example, an ATM camera generates only a series of cells online, but it may also use ATM cells to generate SONET frames and embed them into the SONET payload. In the latter case, the TC sub-layer will generate SONET or frames and package the ATM cells. This is not completely unnecessary, because the SONET payload cannot support an integer multiple of 53-byte cells.

Although telephone companies explicitly use SONET as an ATM Transmission System, it can also be defined as a payload field that maps the ATM to other systems, and this new frame is already working. In particular, it is also possible to map to T1, T3, or FDDI frames.

3. Receive Cells

In the output field, the work of the TC sub-layer is to obtain a series of cells, add a HEC to each cell, convert the result into a bit stream, and add OAM cells, match the bit stream to the speed of the physical transmission system. At the input, the TC layer performs inverse transformation accurately. It obtains the arrived bit stream, sets the cell boundary, determines the Cell Header (discards cells with invalid headers), processes OAM cells, and uploads data cells to the ATM layer.

The most difficult part is to set the cell boundary in the incoming bit stream. In some cases, the physical layer for transmission provides help. However, sometimes the physical layer cannot help frame formation. What should I do?

The trick is to use HEC. As the bit stream reaches the TC sub-layer, a 40-shift register is retained, and the bit stream enters from the left and comes out from the right. The TC child layer then examines the 40 bits to see if a valid Cell Header may exist. If yes, the rightmost 8 bits are valid HEC, And the leftmost 32 bits are not. If this condition does not exist, no valid cell exists in the buffer. In this case, all the bits in the buffer are moved to the right, leaving the backend blank, then a new input bit is added to the leftmost end. Repeat this process until a valid HEC is located. At this point, the cell boundary is clarified because the shift register includes a valid header.