NUMBEREDHEADINGS__
Multilevel principle of building a network
Organization of interaction between devices in the network is a complex task. As you know, a universal technique is used to solve complex problems - decomposition, that is, the division of one complex task into several simpler tasks-modules. The decomposition procedure includes a clear definition of the functions of each module that solves a separate problem, as well as the definition of the functions of the interfaces that connect each module. As a result, a logical simplification of the task is achieved, and, in addition, it becomes possible to correctly modify individual modules without changing the rest of the system, replacing modules.
Decomposition in communication networks uses a multi-level approach. It is as follows:
- the entire set of functional modules is divided into levels;
- levels are organized in the form of a vertical stack, that is, they interact on the basis of a strict hierarchy;
- the set of modules that make up each level is formed in such a way that, in order to perform their tasks, they apply with requests only to the modules of the immediately adjacent level that lies lower in the hierarchy;
- on the other hand, the results of the work of all modules belonging to a certain level can only be transferred to the modules of the neighboring level, which lies higher in this hierarchy.
Such a hierarchical decomposition of the task implies a clear definition of the functions of each level and the interfaces between them. An interface defines a set of functions that the level below in the hierarchy provides to the level above. As a result of hierarchical decomposition, relative independence of levels is achieved, and hence the possibility of their easy replacement.
The number of levels, their names, content and purpose of functional modules vary from network to network. However, in all networks, the purpose of each layer is to provide some services to the upper layers, hiding from them the details of the implementation of the service provided.
Level n network node (one machine), communicates with the level n another network node. The rules and conventions used in this communication are called protocol level n. Essentially, a protocol is an agreement between the communicating parties on how communication should take place.
Protocol is a set of formalized rules, procedures, specifications, a certain format and method of data transmission.
Typically, the protocol provides interaction between processes that are at the same hierarchical level, but at different terminal and transit points of the network. Data elements sent at one hierarchical level are called Protocol Data Unit (PDU) data elements.
On fig. 1. shows a five-layer network. Objects of various network nodes include the corresponding levels. They communicate virtually (logically) using protocols. In reality, no data is sent from the layer n one car per level n another car. Instead, each layer of the sending machine, starting at the top, passes data and control to the layer below it until the lowest layer is reached. Such messages are called service data elements SDU (Service Data Unit). Below the first level is a physical medium, on which information is exchanged. On the receiving side, the transmitted data block sequentially passes through the levels of the receiving machine from bottom to top. Each level performs its own group of functions necessary for receiving data.
Between each pair of adjacent levels interface. These are hardware and software tools, as well as a set of rules that ensure the interaction of adjacent levels.
When network designers decide how many layers to include in a network architecture and what functions each layer should perform, defining clear interfaces between layers is very important. Each level needs to perform a specific set of well-understood functions. Minimization of the amount of service information transmitted between the levels, clearly delimited interfaces greatly simplify the change in the implementation of the level (for example, replacing telephone lines with satellite channels). With a layered approach, all that is required is that the new implementation of a certain layer provides the same set of services to the higher layer as the previous one.
The set of layers and protocols is called network architecture. The architecture specification should contain enough information to write the software or hardware for each layer so that it correctly follows the requirements of the protocol. Neither implementation details nor interface specifications are part of the architecture, as they are hidden inside the machine and not visible from the outside. To make it easier to understand the essence of multi-level communication, you can use the following analogy (Fig. 2).
Imagine that there are two subscribers Bob and Alice (level 3), one of them speaks English and the other speaks French. Since there is no common language in which they can communicate directly, they each use an interpreter (level 2 peer processes). Each of the translators in turn hires a secretary (Tier 1 peer processes). Bob wants to say "I love you" to his interlocutor. To do this, he sends a message in English via interface 2/3 (the interface between the second and third levels) to his translator. The translators agreed to communicate in a neutral language - Russian. Thus, the message is converted to the form "I love you." Language selection is a layer 2 protocol and is handled by layer 2 peer processes. The interpreter then gives the message to a secretary for transmission, eg by fax (layer one protocol). When the message is received by another secretary, it is translated into French and transmitted via interface 2/3 to the subscriber Alice. Note that each protocol is completely independent because the interfaces are the same on each side. Translators can switch from Russian to, say, Finnish, provided that both agree. At the same time, nothing will change in the interfaces of the second level with the first or with the third level. Similarly, secretaries can change from fax to email or telephone without affecting (or even informing) other levels. Each change concerns only the exchange of information at its level. This information will not be passed on to a higher level.
Consider a technical example: how to provide communication for the top layer of a five-layer network (Fig. 3). Message M is created by the application running at layer 5 and passed to layer 4 for transmission. Layer 4 adds a header (34) to the message, for example, to identify the message number, and passes the result to layer 3. In many networks, messages (data) transmitted at layer 4 are not limited in size, but almost always such restrictions are imposed on the protocol of the third level. Accordingly, layer 3 must break the incoming message into smaller units - packets, prefixing each packet with level 3 headers - Z31 (for M1) and Z32 (for M2). In this example, message M is split into two parts M1 and M2. Header 331 and 332 include control information, such as sequence numbers, to allow layer 4 of the receiving machine to deliver messages to its application in the correct order if the lower layers are out of sequence. At some levels, the headers also include the sizes of the data blocks to be sent, the time spent on the network, and other control fields.
Layer 3 decides which of the output lines to use, that is, determines the direction of further transmission, and transfers the packets to layer 2.
Rice. 3. An example of a five-layer network
Here we consider load sharing, when part of the connection M is transmitted on one channel, and the other part on another channel. Level 2 adds not only the Z21 and Z22 headers to each packet, but also the K21 and K22 trailers, which terminate the packet. Level 2 headers and trailers are used to detect packets distorted in the channel and repeat them from the buffer. Layer 2 packets are passed to layer 1 for physical transmission. On the receiving machine, the message is passed up through the layers, with headers removed at each level as the message progresses. Lower level headers are not passed to higher levels.
It is necessary to understand the relationship between virtual and real communication and the difference between a protocol and an interface. Layer 4 peer processes, for example, perceive their communication as horizontal, using a layer 4 protocol. Each of them has a procedure with a name like "Transmit to the opposite side" or "Receive from the opposite side." In fact, these procedures do not communicate with each other, but with lower levels using 3/4 interfaces.
The abstraction of peer processes is key to network design. With its help, the extremely difficult task of developing an entire network can be broken down into several smaller, and quite solvable problems, namely, the development of individual layers.
The above example refers to a reliable service based on connection establishment between users. In the next section, we will look at examples of connection-oriented and connectionless, reliable and unreliable services.
Connectionless and connectionless services, reliable and unreliable connections
Layers can offer two types of services to higher layers: connection-based or connection-free. In communication engineering, the procedure for exchanging messages in the process of establishing or disconnecting connections is called signaling(signaling).
A typical example of a connection-oriented service is a telephone connection: the subscriber first establishes a connection, talks, and then disconnects the connection. The same can be true for data transfer. In some cases, the requesting side negotiates the quality of service parameters, while the other side rejects or accepts.
An example of a connection failure is email advertising. Whether a connection is established or not, a service can be either reliable or unreliable. A reliable service ensures that data is sent without loss.
Reliable service is implemented using acknowledgments sent by the receiver in response to each received message. An example of a reliable service is file transfer, which ensures delivery without distortion. Not all applications are suitable for a reliable service (for example, for the transmission of voice or video information - for them, a large delay due to the retransmission of data received with distortion is unacceptable). The no-acknowledgement, no-connection mode is called datagram mode. By analogy with the telegraph, the sender is not provided with confirmation of receipt of the telegram. Datagram mode is used in addition to voice and video transmission also when reliable data delivery is provided by higher layer protocols.
Packet switching
In accordance with the Law "On Communications" dated June 18, 2003, packet switching is a strategic direction for public communication networks.
When switching packets, the user's message is divided into packets at the terminal communication node - message elements equipped with a header. For example, on the network X.25 maximum length of packet data field
Rice. 4. Data transmission in the CP network
set by agreement (default - 128 bytes). The packet header sets the address information necessary to deliver the packet to the recipient's end device. The X.25 network uses the addressing format defined in the ITU-T X.121 recommendation, containing the area code (there are 7 areas in the world), the code of the specific network in the area, and the ten-digit network terminal number. On fig. 4 shows the transmission of a subscriber-to-subscriber message. Subscriber a i (\displaystyle a_(i)) connected to the switching center A, and the subscriber a j (\displaystyle a_(j)) to the switching center D. Before transmission, the message is divided into three packets, which arrive at the recipient through the transit switching centers B and C.
Packet-switched networks include not only the X.25 network, but also more modern technologies (Frame Relay, ATM networks), as well as the Internet. The channel capacity in the KP network with uneven traffic is significantly higher than in KK networks. The same physical channel is used to serve many subscribers, providing its bandwidth to different connections of subscribers in turn. The greatest effect of the KP is achieved with a high ripple coefficient of the traffic of network users.
The traffic ripple ratio of an individual network user is defined as the ratio of the peak rate over a short period of time to the average data exchange rate over a long period of time and can reach values of 100:1. If you use circuit switching, then most of the time the link will be idle. At the same time, part of the network resources remains assigned to this pair of subscribers and is not available to other network users.
On fig. Figure 5 shows an example of multiplexing packets of different information flows in one physical channel.
Rice. 5. An example of packet multiplexing in one physical channel
The first three axes show the streams of packets generated by subscribers a 1 (\displaystyle a_(1)), a 2 (\displaystyle a_(2)), a 3 (\displaystyle a_(3)). Double numbering of packages in fig. 5 denotes the subscriber number and the packet number in the stream. The channel is used to serve three subscribers - by time division, i.e. alternate provision of the channel to subscribers. One channel can provide the work of many interacting subscribers.
Thus, the packet is transmitted in stages, with a hop, through a number of nodes to the destination. Packets can have a variable length, but within a fairly narrow range: from 50 to 1500 bytes. Packets are transported across the network as independent information units and assembled into a message at the destination node. Packet network switches have an internal buffer memory to temporarily store packets if the switch's output port is busy transmitting another packet.
X.25 packet network protocol stack
Having considered the multi-level principle of building a network, let's move on to the stack of protocols (or levels) of a specific packet switching network of the standard X.25.
 |
The study of the protocol stack of this particular network is explained by the following reasons:
- ITU-T X.25 recommendations and related ones (X.3, X.28, X.75, X.121, etc.) most fully correspond to the standardized ITU-T reference model of open systems interconnection OSI, (Open System Telecommunication), which includes 7 levels. It should be noted that the OSI model does not fully reflect the architecture of building modern communication network technologies. Despite this, the OSI model is an excellent mechanism for analyzing the underlying architecture of these networks.
- many of today's technologies have their roots in the X.25 standard. X.25 networks continue to be in operation (including in Russia - the ROSPAK public network, the ATM protection system of Sberbank of Russia, etc.).
- a presentation of the principles of software in the X.25 network allows you to study the procedures for the operation of more modern network technologies (Frame Relay, ATM, IP networks, SS#7, MPLS).
- various features are used to classify communication networks. Most often, networks are divided by the size of the territory that the network covers. The reason is the difference between the technologies of local and global networks. Wide area networks, which include X.25 networks, are designed to serve a large number of subscribers scattered over a large area - within a region, country, continent or the entire globe. WAN services can be used by local area networks of enterprises or individual computers. Historically, global networks appeared first, although their technology is much more complicated. It was during their construction that the basic concepts of networks were first reflected, such as the multi-level construction of communication protocols, packet switching technology, and quality of service requirements. QoS(Quality of Service) and SLA (Service Level Agreements).
On fig. 6 shows the X.25 PC network protocol stack.
Here is a transport network consisting of three packet switching centers CKP (TsKP1, CKP2, TsKP3) and two end stations - A and B. The DCT includes three lower levels corresponding to the OSI model:
- the physical layer (layer 1) that transmits the bits;
- the data link layer or X.25/2 data link layer (layer 2), which performs error-free transmission over a separate communication channel;
- the X.25/3 network layer (layer 3), which provides routing (switching) of messages over channels connecting the MSC.
At these levels, transport network protocols between MCUs and network access protocols operate. As a rule, the protocols of the upper layers of the OSI model (from 4 to 7) are implemented only at the end devices of the network and are end-to-end protocols.
The fourth layer in the OSI model is the transport layer. The transport layer is located at the terminal stations and provides an interface between the transport network (DCT1, DCT2, DCT3) and the top three data processing layers hosted by the user. The transport layer, in particular, performs segmentation of data transmitted to the network, if necessary.
The data processing layers, sometimes referred to as the application layers, include the application, presentation, and session layers. The application layer provides support for the user's application process and is responsible for the semantics, that is, the semantic content of messages exchanged between the sender and recipient machines. At the application layer are network applications: e-mail, file transfer over the network, etc.
The representation or representation level defines the syntax of the messages to be transmitted, that is, the set of alphabetic characters and how they are represented as binary numbers (primary code). The layer provides a process for negotiating different encodings, and can also perform encryption, decryption, and data compression. The presentation layer provides the application process with independence from differences in syntax.
The session layer manages interaction sessions between user application processes. A session is created at the request of the user process, passed through the application and presentation layers. This level determines which party is currently active and ensures that the dialogue is synchronized. Synchronization tools allow you to organize checkpoints in long transfers so that in the event of a failure you can return to the last checkpoint without starting the entire data transfer from the beginning.
On fig. Figure 7 shows data transfer in PC X.25 networks through all levels of end devices of subscribers A and B, as well as the lower three levels of transport network nodes. Here, the designations D3, D4, D5, D6, D7 are accepted - data blocks of levels, respectively, levels 3, 4, 5, 6, 7. Designations З2, З3, З4, З5, З6 - headers of data blocks, respectively, levels 2, 3, 4, 5, 6. Data transmission is physically carried out vertically: for transmission from the upper level to the lower one and vice versa for reception. To send a message of the fourth level of the terminal device (consisting of the header Z4 and data D5), it is nested ( encapsulated) into a third-level (network) package. This adds a 33 header (including the address) to the packet. Based on the header addresses, switching is performed at the packet switching center. Next, this packet is encapsulated in a second-level frame. As can be seen from the figure, in addition to the header Z2, a trailer K2 is added to the frame, which serves to detect a frame distorted in the channel at the reception.
Encrypting Messages in a Packet Switched Network
Encryption is one way to counter threats to the information security of some restricted networks. When using encryption, it is necessary to decide what exactly should be encrypted and at what level of the OSI reference model the information should be protected. For such networks, information packets are switched based on the routing table, which includes physical addresses. In this respect, they differ from the X.25 packet-switched network (detailed in Chapter 7), and they provide two main encryption options: link and end-to-end encryption. Their use is shown in Fig. eight.
Rice. 8. Encryption in the packet network
At channel encryption each vulnerable channel at the third layer of X.25 is equipped with encryption devices at both ends. Thus, the entire data flow in the channel is protected. Although this would require a significant number of encryption devices (per network link) in a large network, the advantages of this approach are clear. The downside is that the message must be decrypted each time it passes through the packet switch, because the switch must read the address in the packet header in order to route the packet in the right direction. Therefore, the message is vulnerable in each switch.
At end-to-end encryption the encryption process is performed at a level above the third only in two end stations. The source data is encrypted in the end device of the message source. The encrypted data is then transmitted unchanged across the entire network to the recipient. The recipient uses the same key as the sender and can therefore decrypt the received data. This scheme seems to be safe from the point of view of protection against influences in the communication channel or packet switching nodes. However, this approach also has a weak point.
What part of each packet, when end-to-end encrypted, should the source encrypt? Let's say it encrypts the entire packet, including the header. But this cannot be done, since only the recipient will be able to decrypt. An MCU receiving such a packet will not be able to read the header and therefore will not be able to forward the packet according to the address. It follows that the sender should only encrypt the part of the packet that contains the user's data and leave the header intact.
So, with end-to-end encryption, user data is protected, which cannot be said about the data stream itself, since packet headers are transmitted in clear text. The ability to study the flow structure from the addresses of passing packets is called traffic analysis. To achieve a higher level of security, a combination of channel and end-to-end encryption is needed, for example, as shown in Fig. 8, which shows a packet switching network with four DSC switching centers. Terminal devices are connected to three of these DCUs a 1 (\displaystyle ~a_(1)), a 2 (\displaystyle ~a_(2)), a 3 (\displaystyle ~a_(3)). Consider the following situation. The two end devices establish a data connection using encryption. Messages are sent in packets consisting of a header and a data field. What part of the packet should the source endpoint encrypt?
With both forms of encryption, the source node encrypts at a level above the third packet of user data using an end-to-end encryption key. The entire packet is then encrypted with the channel cipher key. As a packet travels across the network, each switch first decrypts the packet using the corresponding channel's encryption key to read the header, and then re-encrypts the entire packet to pass it on to the next channel. Now the entire packet is protected almost all the time - except for the time when it is in the memory of the packet switch, where the packet header is exposed.
Principles of construction and components of the X.25 network
The main feature of the X.25 network is the use of virtual channels to ensure information exchange between network components. Virtual channels are designed to organize a call and direct data transfer between network subscribers. Information exchange in the X.25 network is in many ways similar to the analogous process in ISDN networks and consists of three mandatory phases:
- Call establishment (virtual circuit)
- Information exchange over a virtual channel
- Call release (virtual circuit)
Information interaction in the X.25 network is carried out at the physical, data link and network levels. At the physical layer, any generic or specialized interfaces can be used. Network components are devices of three main categories:
- DTE devices(Data Terminal Equipment)
- DCE devices(Data Circuit-Terminating Equipment)
- PSE devices(Packet Switching Exchange)
PAD device(packet assembler/ disassembler) is a specific device of the X.25 network. PAD is designed to ensure the interaction of non-dedicated terminals with the network, to convert the stream of characters that comes from the non-dedicated terminal into X.25 packets and perform the reverse conversion.
Interworking at the data link layer of the X.25 network
The HDLC/SDLC link layer protocols have been developed in order to solve the following tasks:
- Ensuring the transmission of messages that can contain any number of bits and any possible combination of bits is a code transparency requirement.
- When transmitting a stream of bits, procedures must be performed that allow errors to be detected on the receiving side.
- The occurrence of an error during transmission should not result in the loss or duplication of message components, i.e. to its distortion.
- The link layer protocol had to support both point-to-point and multipoint physical circuits.
- The protocol must provide the connection of duplex and half duplex lines
- The protocol must provide information exchange with significant variations in the signal propagation time
HDLC family protocols
Protocols carry out data transmission in the form of frames of variable length. The beginning and end of a frame is marked with a special bit sequence called flag. To ensure the discipline of managing the data transfer process, one of the stations that provide information exchange can be designated as primary, and the other (or other) stations can be designated as secondary. The frame sent by the primary station is called team(command). The frame that the secondary station generates and transmits is called answer(response).
Modes of organization of interaction at the data link level
The segment secondary station can operate in two modes: normal response or in mode asynchronous response. A secondary station that is in normal answer mode only starts data transmission if it has received an enable command from the primary station. The secondary station, which is in the asynchronous response mode, may initiate the transmission of a frame or group of frames on its own initiative. Stations that combine the functions of primary and secondary stations and are called combined.The symmetrical mode of interaction of combined stations is called balanced mode.
LAPB procedure
Procedure LAPB(Link Access Procedure Balanced) is used in X.25 networks as a link layer protocol.
Flag
The LAPB protocol uses as a flag a combination of 8 bits, which consists of 6 ones and two zeros, which frame this sequence in front and behind (01111110). The frame reception process ends when the next flag is received. If the receiver has received less than 32 bits by the time the termination flag is received, the received frame is considered erroneous and is discarded. A special procedure is used to prevent the appearance of a flag combination in the frame body.
LAPB frame structure
Recommendation X.25 defines two main types of LAPB procedure - basic type (modulo 8, basic) and extended type (modulo 128, extended). These modes differ in the size of the counters that are used to control the flow of frames. The LAPB protocol frame contains 4 fields: ADDRESS, CONROL, Data, FCS. The DATA field in the LAPB frame MAY be absent.
ADDRESS field
The ADRESS field occupies one byte in the frame. This field contains the C/R (Command /Response) flag bit. The ADDRESS field of the control command frame contains the physical address of the receiving station. The ADRESS field of the command response frame contains the physical address of the transmitting station.
CONTROL field
The content of this field field determines the frame type.
- information frames(Information Frames, I-frames). The bits of the CONTROL field contain the 3-bit number of the transmitted frame and the 3-bit number of the frame that is expected to be received to provide flow control.
- Control frames(Supervisory Frames, S-frames). The CONTROL field contains a 3-bit number of the information frame that is expected to be received and two bits that determine the type of control frame being transmitted.
Designation frame type Bit #3 Bit #4 RR Receiver Ready 0 0 RNR Receiver Not Ready 1 0 REJ Refusal/request (Reject) 0 1 Most often in the process of information interaction, control frames of the type RR. Frames of this type are transmitted by the recipient of data in order to indicate readiness to receive the next frame, in the case when he himself does not have information to transmit. Personnel RNR are used by DCE and DTE devices to inform the subscriber about the occurrence of an emergency in which further reception of information frames is impossible. REJ frames are used by DCEs and DTEs to signal to the subscriber that an emergency has been resolved in which information frames could not be received. The REJ frame is transmitted after the RNR frame and confirms that the line has transitioned to normal operation.
- Unnumbered Frames(Unnumbered Frames, U - frames). They are intended for organizing and breaking a logical connection, matching line parameters and generating signals about the occurrence of fatal errors in the process of data transmission by I-frames.
Designation Type of sign SABM(E) Set Asynchronous Balanced Mode Team DISC Disconnect Team DM Disconnect Mode Answer U.A. Unnumbered Acknowledgment Answer FRMR frame reject Answer - Frame FRMR transmitted by the secondary station to indicate the occurrence of an emergency that cannot be resolved by retransmission of the emergency frame.
Network Layer X.25
For transmission over the network, X.25 packets are encapsulated in LAPB frames. The LAPB protocol provides reliable delivery of these packets over a link that links one network component to another. One physical channel in an X.25 network can be used to carry packets that belong to several different network layer processes. Unlike the principle of static time division, which is used in ISDN networks, the X.25 network uses the principle of dynamic division to allocate channel resources.
X.25 virtual circuits
The network layer process acquires a portion of the bandwidth of the physical link as a virtual link. The total bandwidth of the channel is divided equally among the virtual channels that are currently active. There are two types of virtual circuits in an X.25 network: switched(SVC) and permanent(PVC).
X.25 Packet Format
An X.25 packet consists of at least three bytes that define the header of the packet. The first byte contains 4 bits common format identifier and 4 bits logical channel group number. The second byte contains logical channel number, and the third package type identifier. Packets in the network are of two types − control packets and data packets. The packet type is determined by the value of the least significant bit of the packet type identifier.
General format identifier
The generic format identifier field contains a flag that specifies the type of packet flow control procedure (modulo 8 or modulo 128).
Logical channel number
The logical channel number is specified by the contents of two fields - the logical channel group number from 0 to 15 and the channel number in the group from 0 to 255. Thus, the maximum number of logical channels can be up to 4095. The logical channel number determines the virtual port with which a particular user interface is associated. process.
Packet type identifier
| DCE "width="11" height="9"> DTE | DTE "width="11" height="9"> DCE | Code (16) |
| incoming call | call request | 0B |
| Call Connected | Call Accepted | 0F |
| clear indication | Clear Request | 13 |
| Clear Confirmation | Clear Confirmation | 17 |
| Interrupt | Interrupt | 23 |
| Interrupt Confirmation | Interrupt Confirmation | 27 |
| Receiver Ready (RR) | Receiver Ready (RR) | X1 |
| Receiver Not Ready (RNR) | Receiver Not Ready (RNR) | X5 |
| | Reject (REJ) | X9 |
The network addresses of the recipient and sender of the packet are placed in the "data" field, and are intended for call control.
X.25 network address format
The network address consists of two parts
- Data Network ID Code (DNIC)
- Network Terminal Number
The DNIC field contains 4 decimal digits and specifies the country code and provider number. The content of the Network Terminal Number field contains 10 or 11 decimal digits that are defined by the provider and are intended to identify a specific user.
Frame flow management
Packet flow control at the X.25 network layer uses the same procedures and mechanisms that are used for frame flow control at the X.25 network link layer.
In order to provide the ability to connect various types of terminals to the X.25 network, special algorithms and parameters are used that control the process of assembling and disassembling packets.
This recommendation defines the names and purposes of the main parameters with which the PAD is configured. The X.3 parameters are denoted by symbols P1 - P32. The P1 parameter determines whether it is possible to exit from the transmission mode to the command mode at the initiative of the terminal operator.
For flow control, special code combinations XON and XOFF are used. In the event that the terminal is temporarily unable to receive characters from the PAD for some reason, it sends the XOFF (^S) character. The PAD must stop transmitting data to this terminal before receiving an XON(^Q) enable character from it. The meaning of these symbols can be overridden using parameters P28 and P29.
This recommendation defines the procedures by which a user can read or change the current X.3 PAD parameter values. To change the set X.3 PAD parameters, the user must use the SET command. In order to read the current X.3 PAD parameter values, the user must use the PAR command.
WANs with packet switching
Lecture number 11.
X.25 networks are the very first packet-switched networks used to interconnect corporate networks. Initially, networks were designed for low-speed data transmission over communication lines with a high level of interference, and were used to connect ATMs, cash terminals that accept credit cards, and to interconnect enterprise networks.
For a long time, the X.25 network was the only widespread commercial network (the Internet, as a commercial network, began to be exploited quite recently), so there was no choice for corporate users.
To date, the X.25 network continues to operate successfully, using high-speed digital links to connect its switches. For example, most Western banks and industrial enterprises use the X.25 network to organize remote access to their networks.
An X.25 network consists of switches connected to each other in a point-to-point manner and operating with the establishment of a virtual circuit. To connect the switches, digital PDH/SDH lines or analog modems operating on a leased line can be used.
Computers (routers) that support the X.25 interface can be connected to the switch directly, and less intelligent terminals (ATMs, cash registers) can be connected using a special PAD (Packet Assembler Disassembler) device. The PAD can be built into the switch or remote. Terminals access the embedded PAD over the telephone network using modems (the embedded PAD is also connected to the telephone network using multiple modems). The Remote PAD is a small stand-alone device located at the client's premises and connected to the switch via a dedicated line. The terminals are connected to the remote PAD via a COM port (RS-232C interface).
One PAD usually provides access for 8, 16 and 24 terminals.
Terminals do not have end addresses on the network X.25 - address is assigned only to the PAD port.
Addressing in X.25 networks according to the following principle: Decimal digits are used in the address, the length of the address cannot exceed 16 digits. If the X.25 network is not connected to the outside world, then it can use any address. If the X.25 network plans to communicate with other networks, then it is necessary to adhere to the international addressing standard (X.121 standard - International Data Numbers, IDN).
The address format in an X.25 network is as follows:
4 digits - network identification code (Data Network Identification Code, DNIC), 3 digits - identify the country in which the X.25 network is located, 1 digit - the X.25 network number in this country, the remaining digits - the number of the National Terminal (National Terminal Number, NTN) (corresponding to the computer address on the network).
From the above format, it is clear that there can only be 10 X.25 networks in one country. If it is required to number more than 10 networks, then several codes are given to one country. For example, until 1995 Russia had one code - 250, and in 1995 another code was allocated to it - 251.
The address can use not only numbers, but also arbitrary characters (to do this, you need to add a special prefix to the address), which allows universal switches, such as ISDN network switches, to work with X.25 network packets.
The main disadvantage X.25 network is that it does not guarantee network bandwidth. The most it can do is set priorities for individual virtual channels. Therefore, the X.25 network is only used to carry traffic that is sensitive to delays (such as voice). Frame Relay and ATM networks deal with this problem.
The phrase in the title of this article, in a nutshell, reflects the current situation with X.25 technology. In the Western press it is now very difficult to find consideration of the problems associated with the use of the X.25 protocol; hot topics today in area networks are, for example, frame relay and ATM technologies. Despite this, even in Western countries, the most advanced companies that previously produced only high-speed equipment are supplementing their range of devices with X.25 equipment. An example of this is the appearance this year of X.25 nodes in the range of StrataCom equipment.
Consideration of X.25 technology solutions for Russia and neighboring countries with a similar channel infrastructure is especially relevant.
In this article, we will discuss the X.25 protocol and its associated protocol stack, as well as networks based on this technology. Our goal is to show what X.25 networks are and why it is beneficial for a wide range of users to use already functioning X.25 backbone networks, and for some of them, representing large organizations, even to build their own networks.
We will refer to networks as X.25, or packet switched networks, as networks accessed in accordance with CCITT X.25 recommendations (in accordance with X.3/X.28 in the case of asynchronous access).
So why X.25 networks? The fact is that today, despite the emergence of new, integrated technologies for data transmission networks / communication networks designed for high-speed communication channels, X.25 networks are still the most common.
If we consider all the public data transmission networks available today, it turns out that X.25 networks can most justifiably be likened to telephone networks. In the same way that by picking up the handset of a telephone set connected to the nearest PBX, you can reach a subscriber almost anywhere in the world, so by establishing a connection from your computer to the nearest X.25 network node, you can communicate with any of the millions of network users X.25 worldwide. To do this, you only need to know its network address.
What are X.25 networks? What are they needed for? What equipment and what theory are they based on?
X.25 NETWORK PROTOCOLS
X.25 networks are named after the "X.25" recommendation issued by the CCITT (International Telephony and Telegraph Advisory Committee). This recommendation describes the interface for user access to the data network and the interface for interacting with a remote user through the data network.
Within the network itself, data transmission can occur in accordance with other rules. The core of the network can also be built on faster frame relay protocols. However, when considering the issues of building X.25 networks within the framework of this article, we will keep in mind networks within which data transfer is also carried out using the protocols described in the X.25 recommendation. It is in this way that the majority of corporate X.25 networks in Russia are currently being built.
Today, a sufficiently high level of compatibility of equipment manufactured by different companies has been achieved, both within the same network and various X.25 networks. The greatest compatibility problems arise when it is necessary to manage from one center network nodes built on the basis of equipment from different companies. However, thanks to the installation of SNMP agents on X.25 equipment, this problem will apparently be solved in the near future. At the same time, work is underway to expand the capabilities of the SNMP protocol in terms of its compliance with the tasks of managing large geographically distributed networks.
The first describes signal levels and interaction logic in terms of a physical interface. (Those of the readers who, for example, had to connect a modem to a serial port of a personal computer via an RS-232/V.24 interface, have an idea about this level.)
The second (Channel Access Protocol/Balanced Channel Access Procedure, LAP/LAPB), with various modifications, is now quite widely represented in mass-market equipment - for example, in modems - by protocols such as Microcom's MNP network protocol, which are responsible for error correction when transmission of information over a communication channel, as well as in local networks at the LLC logical channel control level.
This protocol layer is responsible for the efficient and reliable transfer of data over a point-to-point connection, i.e. between neighbors in an X.25 network. This protocol provides error correction during transmission between neighboring nodes and data flow control (if the receiving side is not ready to receive data, it notifies the transmitting side about this, and it suspends transmission). In addition, it determines the parameters, changing the values of which, the transmission mode can be optimized in terms of speed depending on the length of the channel between two points (delay time in the channel) and its quality (probability of information distortion during transmission).
To implement all the above functions in the protocols of the second level, the concept of "frame" (frame) is introduced. A frame is a piece of information (bits) organized in a certain way. Starts a flag frame, i.e. a sequence of bits of a strictly defined type, which is a separator between frames. Then comes the address field, which in the case of a point-to-point connection is address A or address B. This is followed by a frame type field indicating whether the frame carries information or is purely service (for example, slows down the flow of information or notifies the transmitting side of the reception / previous frame was not received). The frame also has a frame number field. Frames are numbered cyclically. This means that when the specified threshold value is reached, the numbering starts again from zero. And finally, the frame ends with a control sequence, calculated during the transmission of the frame according to certain rules. According to this sequence, at the reception, a check is made for information distortion during frame transmission.
The frame length can be changed when setting the protocol parameters to the physical characteristics of the line. The shorter the frame, the less likely it is to be corrupted during transmission. However, if the line is of good quality, then it is better to work with longer information frames, because the percentage of redundant information transmitted over the channel (flag, frame service fields) decreases. In addition, the number of frames sent to the transmitting side without acknowledgment from the receiving side can also be changed. This parameter is related to the so-called "numbering module", i.e. with the value of the threshold, upon reaching which the numbering starts again from zero. This field can be set to between 8 (for those channels where the information transmission delay is not too high) to 128 (for satellite channels, for example, when the delay in information transmission over the channel is large).
And, finally, the third level of protocols is network. It is most interesting in the context of discussing X.25 networks, since it is he who determines their specifics in the first place.
Functionally, this protocol is primarily responsible for routing in the X.25 data network, that is, for bringing information from the “entry point” into the network to the “exit point” from it. For its part, the protocol of the third level also structures the information, in other words, breaks it into "portions". At the third level, a piece of information is called a "package" (packet). The structure of a packet is similar in many respects to the structure of a frame. The package has its own numbering module, its own address fields, package type, control sequence. During transmission, the packet is placed in the data field of information frames (frames of the second level). The packet fields are functionally different from the corresponding frame fields. This mainly concerns the address field, which in the packet consists of 15 digits; the packet field must provide identification of subscribers within all packet-switched networks around the world. The network address structure is defined by the X.121 recommendation.
Having introduced the term "packet", we can move on to the next question, namely: how is the delivery of information from one subscriber to another through the X.25 network? For this, the so-called "packet switching" method is used, in connection with which X.25 networks are also called packet switching networks. This method is implemented by establishing virtual between subscribers, i.e. logical (as opposed to physical) connections (virtual circuits). In order to transfer information from subscriber A to subscriber B, a virtual connection is first established between them, otherwise, packets are exchanged "call request" - "call accepted" ("call accept"). Only then can information be exchanged between the two subscribers.
Virtual connections can be either permanent or switched. A switched connection, unlike a permanent virtual connection, is established in each information exchange session. Here you can give direct analogies from the field of telephony. Indeed, if you have a dedicated ("permanent") telephone channel between two subscribers, then you do not need to dial your subscriber's number every time - you just need to pick up the phone. The number of virtual connections supported simultaneously on the basis of one physical link depends on the specific type of equipment used to provide such connections. Which is quite understandable, because. to support each connection on this equipment, a certain resource (for example, RAM) must be reserved.
ADVANTAGES OF X.25 NETWORKS
The packet switching method underlying X.25 networks determines the main advantages of such networks, or, in other words, their scope. What is this advantage? The networks under consideration make it possible to share the same physical channel to several subscribers in real time, unlike, for example, the case of using a pair of modems connected through a channel of one type or another. In fact, if you and your subscriber have modems installed on their computers, you can exchange information with him. However, no one else will be able to use the telephone line you are using at the same time.
Thanks to the channel sharing mechanism implemented in X.25 networks between several users at once, in many cases it turns out to be more cost-effective to pay for each byte of transmitted or received information, rather than pay for the time the telephone line is used when transmitting data over the X.25 network. This advantage is especially noticeable in the case of international connections.
The method of dividing a physical channel between subscribers in X.25 networks is also called channel multiplexing, more precisely, "logical" or "statistical" multiplexing (Fig. 1). The term "logical multiplexing" is introduced to distinguish this technique from, for example, channel time division. With the time division of the channel, each of the subscribers separating it is allocated a strictly defined number of milliseconds per second for transmitting information. With statistical channel separation, there is no strictly regulated degree of loading by each of the channel subscribers at a given time.
Picture 1.
Channel multiplexing in X.25 networks.
The efficiency of using statistical multiplexing depends on the statistical or probabilistic characteristics of the multiplexed information flow. Does this mean that before you connect to an already existing X.25 network or start building your own network, you need to conduct a detailed analysis of the probabilistic characteristics of the information flows circulating in your system? Of course not. Such calculations have already been made. A lot of experience has been accumulated in using X.25 networks. It is known that the use of the X.25 network is effective for a wide range of data transmission tasks. Among them are the exchange of messages between users, and the appeal of a large number of users to a remote database, as well as to a remote e-mail host, local area network communications (at exchange rates not exceeding 512 Kbps), the integration of remote cash registers and ATMs. In other words, all applications where network traffic is not uniform over time.
What other benefits does the X.25 network provide? Perhaps one of the most important advantages of networks built on the protocols described in Recommendation X.25 is that they allow data to be transmitted over public telephone network circuits (dedicated and switched) in an optimal way. By "optimality" is meant the achievement of the maximum possible speed and reliability of data transmission on the indicated channels.
An effective mechanism for optimizing the process of transmitting information over X.25 networks is the alternate routing mechanism. The ability to set alternative routes in addition to the main route, i.e. redundant, available in X.25 equipment manufactured by almost all companies. Different types of equipment differ in the algorithms for switching to an alternative route, as well as the allowable number of such routes. In some types of equipment, for example, the transition to an alternative route occurs only in the event of a complete failure of one of the links of the main route. In others, the transition from one route to another occurs dynamically depending on the congestion of the routes, and the decision is made on the basis of a multi-parameter formula (Motorola ISG equipment, for example). Through alternative routing, the reliability of the network can be significantly increased, which means that there must be at least two different routes between any two points of a user's connection to the network. In this regard, the construction of a network according to a star-shaped scheme can be considered a degenerate case. True, where there is only one X.25 network node installed within a given public network, this network topology is still used quite often.
USER ACCESS TO X.25 NETWORKS. PACKAGE COLLECTORS
Let us now consider how access of different types of users to the X.25 network is implemented in practice. First of all, it is possible to organize access in batch mode (recommendations X.25). To access the network from a computer in packet mode, for example, you can install a special board in your computer that provides data exchange in accordance with the X.25 standard.
Boards from Microdyne, Newport Systems Solutions, etc. are also used to connect a local network via an X.25 network. In addition, access from a local network to an X.25 network can also be organized using remote access bridges / routers that support the X protocol. 25 and made in the form of stand-alone devices. The advantages of such devices over boards embedded in a computer, in addition to greater performance, are that they do not require the installation of special software, but are interfaced with a local network using a standard interface, which allows for more flexible and universal solutions.
In general, connecting a user equipment to a network in batch mode is very convenient when multiple users need to access this equipment via a network.
If you need to connect the computer to the network in exclusive mode, then the connection is made according to other standards. These are the X.3, X.28, X.29 standards that define the functioning of special network access devices - packet assemblers / disassemblers - PDA (packet assembler / dissasembler-PAD). In practice, the term "PSA" is rarely used, so we will also use the term "PAD" as a Russian language.
PADs are used to access the network of subscribers in an asynchronous mode of information exchange, i.e. through, for example, the serial port of a computer (directly or using modems). The PAD usually has several asynchronous ports and one synchronous (X.25 port). The PAD collects the data coming through the asynchronous ports, packs them into packets and transmits them through the X.25 port (Fig. 2).
(1x1)
Figure 2.
An example of a complex X.25 network connecting devices of various types: from computers to bank terminal equipment.
The configurable PDA parameters are determined by the tasks being performed. These parameters are described by the X.3 standard. The set of parameters is called "profile" (profile); the standard set consists of 22 parameters. The functionality of these parameters is the same for all PADs. The profile includes parameters that set the exchange rate over the asynchronous port, parameters specific to text editors (sign and line deletion character, display of the previous line on the screen, etc.), parameters that include the mode of automatic line completion with insignificant characters (for synchronization with slow terminals), as well as a parameter that determines the condition under which the packet formation ends.
NETWORK NODES X.25. PACKET SWITCHING CENTERS
The parameters describing the X.25 channel are also important for the nodal elements of the X.25 network itself, called Packet Switching Centers - PSC (or packet switch, packet switch), however, the list of PSC parameters, of course, is not exhausted by them. In the process of configuring the MCU, it is necessary to fill in the routing table (routing table), which allows you to determine to which of the ports of the MCU the packets received by them are directed, depending on the addresses contained in these packets. The table contains both main and alternative routes. In addition, an important function of some DCUs is the function of network docking (gateway between networks).
Indeed, there are a great many X.25 networks in the world, both public and private, or in other words - corporate, departmental. Naturally, in different networks, different values of transmission parameters over X.25 channels (frame and packet lengths, packet sizes, addressing system, etc.) can be set. In order for all these networks to be able to interface with each other, the X.75 recommendation was developed, which defines the rules for negotiating parameters when moving from network to network. It is recommended to pair your and neighboring networks through the MCU, which fully implements support for gateway functions - such a MCU, for example, should be able to "translate" addresses when moving from one network to another. This function is usually implemented by configuring a special address translation table in the gateway MSC. For MSCs that do not interface with nodes of another packet switching network, the presence of gateway functions is not mandatory.
X.25 networks are the first packet-switched networks and are by far the most common packet-switched networks used to build corporate networks. The X.25 network protocol is designed to transfer data between computers over telephone networks. X.25 networks are designed for low quality, high noise lines (analog telephone lines) and provide data rates up to 64Kbps. X.25 performs well on poor quality links due to the use of handshaking and error correction protocols at the data link and network layers.
Principles of construction and components of the X.25 network
The main feature of the X.25 network is the use of virtual channels to provide information exchange between network components. Virtual channels are designed to organize a call and direct data transfer between network subscribers. Information exchange in the X.25 network is in many ways similar to the analogous process in ISDN networks and consists of three mandatory phases:
Call establishment (virtual circuit)
Information exchange over a virtual channel
Call release (virtual circuit)
Information interaction in the X.25 network is carried out at the physical, data link and network levels. At the physical layer, any universal or specialized interfaces can be used. The figure shows the block diagram of the X.25 network, which shows the main elements:
DTE (Data Terminal Equipment) devices
Data Circuit Terminating Equipment (DCE) devices
PSE (Packet Switching Exchange) devices
The PAD device (packet assembler/ disassembler) is a specific device of the X.25 network. PAD is designed to ensure the interaction of non-dedicated terminals with the network, to convert the stream of characters that comes from the non-dedicated terminal into X.25 packets and perform the reverse conversion.
The X.25 interface provides:
1) remote user access to the host computer;
2) remote PC access to the local network;
3) communication of a remote network with another remote network.
X.25 interface
The X.25 interface contains the three lower layers of the OSI model: physical, data link, and network. A feature of this network is the use of switched virtual channels for data transmission between network components. Establishment of a switched virtual circuit is performed by service protocols that act as a signaling protocol.
Physical layer The X.25 physical layer uses analog leased lines to provide a point-to-point connection. Analog telephone lines can be used, as well as digital leased lines. There is no validation or flow control at the network layer. The X.25 physical layer implements one of the X.21 or X.21bis protocols.
Link layer At the link layer, the X.25 network provides guaranteed delivery, data integrity, and flow control. At the data link layer, the data stream is structured into frames. Error control is performed in all nodes of the network. If an error is detected, the data is retransmitted. The link layer is implemented by the LAP-B protocol, which only works with point-to-point links, so no addressing is required.
The HDLC/SDLC link layer protocols have been developed in order to solve the following tasks:
Ensuring the transmission of messages that can contain any number of bits and any possible combination of bits is a code transparency requirement.
When transmitting a stream of bits, procedures must be performed that allow errors to be detected on the receiving side.
The occurrence of an error during transmission should not result in the loss or duplication of message components, i.e. to its distortion.
The link layer protocol had to support both point-to-point and multipoint physical circuits.
The protocol must provide the connection of duplex and half duplex lines
The protocol must provide information exchange with significant variations in the signal propagation time
HDLC family protocols Protocols carry out data transmission in the form of frames of variable length. The beginning and end of a frame is marked with a special sequence of bits called a flag. To ensure discipline in the control of the data transmission process, one of the stations that provide information exchange may be designated as primary, and the other (or other) stations may be designated as secondary. The frame that the primary station sends is called a command. The frame that the secondary station generates and transmits is called a response.
Modes of organization of interaction at the data link level
The segment's secondary station can operate in two modes: normal answer mode or asynchronous answer mode. A secondary station that is in normal answer mode only starts data transmission if it has received an enable command from the primary station. The secondary station, which is in the asynchronous response mode, may initiate the transmission of a frame or group of frames on its own initiative. Stations that combine the functions of primary and secondary stations are called combined stations. The symmetrical mode of interaction of combined stations is called balanced mode.
LAPB procedure
The LAPB (Link Access Procedure Balanced) procedure is used in X.25 networks as a link layer protocol.
The LAPB protocol uses as a flag a combination of 8 bits, which consists of 6 ones and two zeros, which frame this sequence in front and behind (01111110). The frame reception process ends when the next flag is received. If the receiver has received less than 32 bits by the time the termination flag is received, the received frame is considered erroneous and is discarded. A special procedure is used to prevent the appearance of a flag combination in the frame body.
LAPB frame structure
Recommendation X.25 defines two main types of LAPB procedure - basic type (modulo 8, basic) and extended type (modulo 128, extended). These modes differ in the size of the counters that are used to control the flow of frames. The LAPB protocol frame contains 4 fields: ADRESS, CONROL, Data, FCS. The DATA field in the LAPB frame MAY be absent.
The ADRESS field occupies one byte in the frame. This field contains the C/R (Command /Response) flag bit. The ADDRESS field of the control command frame contains the physical address of the receiving station. The ADRESS field of the command response frame contains the physical address of the transmitting station.
CONTROL field
The content of this field field determines the frame type.
information frames(Information Frames, I-frames). The bits of the CONTROL field contain the 3-bit number of the transmitted frame and the 3-bit number of the frame that is expected to be received to provide flow control.
Control frames(Supervisory Frames, S-frames). The CONTROL field contains a 3-bit number of the information frame that is expected to be received and two bits that determine the type of control frame being transmitted.
Most often in the process of information exchange, control frames of the RR type are used. Frames of this type are transmitted by the recipient of data in order to indicate readiness to receive the next frame, in the case when he himself does not have information to transmit. RNR frames are used by DCEs and DTEs to notify the subscriber of an emergency situation in which further reception of information frames is not possible. REJ frames are used by DCEs and DTEs to signal to the subscriber that an emergency has been resolved in which information frames could not be received. The REJ frame is transmitted after the RNR frame and confirms that the line has transitioned to normal operation.
Unnumbered Frames(Unnumbered Frames, U - frames). They are intended for organizing and breaking a logical connection, matching line parameters and generating signals about the occurrence of fatal errors in the process of data transmission by I-frames.
Frame FRMR transmitted by the secondary station to indicate the occurrence of an emergency that cannot be resolved by retransmission of the emergency frame.
network layer The X.25 network layer is implemented by the Packet-Layer Protocol (PLP). At the network level, frames are combined into one stream, and the overall stream is divided into packets. The PLP protocol manages the exchange of packets through virtual circuits. A communication session is established between two DTE devices upon request from one of them. Once a switched virtual circuit is established, these devices can communicate in full duplex. A session may be terminated at the initiative of any DTE, after which a subsequent exchange will need to establish a connection again.
The PLP protocol defines the following modes: Connection establishment is used to establish a switched virtual circuit between DTEs. The connection is established as follows. The calling party DTE sends a request to its local DCE, which includes in the request the calling party address and the unused logical channel address for use by its connection. The DCE determines the PSE that can be used for this transmission. A packet transmitted along the PSE chain reaches the final remote DCE, where the DTE of the destination node is determined, to which the packet is delivered. The calling DTE responds to its DCE, which sends the response to the remote DCE for the remote DTE. Thus, a switched virtual circuit is created. The data transfer mode that is used when exchanging data via virtual circuits. This mode performs error checking and flow control. Idle mode is used when a switched virtual circuit is installed but no data is being exchanged. Connection reset is used to end the session, a particular virtual connection is terminated.
X.25 virtual circuits
The network layer process acquires a portion of the bandwidth of the physical link as a virtual link. The total bandwidth of the channel is divided equally among the virtual channels that are currently active. There are two types of virtual circuits in an X.25 network: switched (SVC) and permanent (PVC).
X.25 Packet Format
An X.25 packet consists of at least three bytes that define the header of the packet. The first byte contains 4 bits of the common format identifier and 4 bits of the logical channel group number. The second byte contains the logical channel number, and the third byte contains the packet type identifier. There are two types of packets in the network - control packets and data packets. The packet type is determined by the value of the least significant bit of the packet type identifier.
General format identifier
The generic format identifier field contains a flag that specifies the type of packet flow control procedure (modulo 8 or modulo 128).
Logical channel number
The logical channel number is specified by the contents of two fields - the logical channel group number from 0 to 15 and the channel number in the group from 0 to 255. Thus, the maximum number of logical channels can be up to 4095. The logical channel number determines the virtual port with which a particular user interface is associated. process.
Packet type identifier The network addresses of the recipient and sender of the packet are placed in the "data" field, and are intended for call control.
X.25 network address format
The network address consists of two parts Data Network ID Code (DNIC) Network Terminal Number
The DNIC field contains 4 decimal digits and specifies the country code and provider number. The content of the Network Terminal Number field contains 10 or 11 decimal digits that are defined by the provider and are intended to identify a specific user.
Frame flow management
Packet flow control at the X.25 network layer uses the same procedures and mechanisms that are used for frame flow control at the X.25 network link layer.
In order to provide the ability to connect various types of terminals to the X.25 network, special algorithms and parameters are used that control the process of assembling and disassembling packets.
This recommendation defines the names and purposes of the main parameters with which the PAD is configured. The X.3 parameters are denoted by symbols P1 - P32. The P1 parameter determines whether it is possible to exit from the transmission mode to the command mode at the initiative of the terminal operator.
For flow control, special code combinations XON and XOFF are used. In the event that the terminal is temporarily unable to receive characters from the PAD for some reason, it sends the XOFF (^S) character. The PAD must stop transmitting data to this terminal before receiving an XON(^Q) enable character from it. The meaning of these symbols can be overridden using parameters P28 and P29.
This recommendation defines the procedures by which a user can read or change the current X.3 PAD parameter values. To change the set X.3 PAD parameters, the user must use the SET command. In order to read the current X.3 PAD parameter values, the user must use the PAR command.
Advantages and disadvantages.
Advantages of the X.25 network:
high reliability, network with guaranteed information delivery;
both analog and digital data transmission channels (leased and switched communication lines) can be used.
Network Disadvantages:
significant packet transmission delays, so it cannot be used to transmit voice and video information.
Literature.
Novikov Yu.V., Kondratenko S.V. Fundamentals of local networks, 2005
