The Basic Concept and Architecture of Local Area Network

Here in the local area network, most of the knowledge involved is the knowledge of the data link layer. I also learned a lot when I was learning the knowledge of the data link layer.

The local area network, that is, the computer group directly connected in the same area, the important thing is that the broadcast channel is used. This broadcast is different from the subsequent broadcasts. It is a real broadcast, a public channel, and the content is directly sent to the channel. Go, everyone can receive.

At the same time, we assume that the nature of the link layer will be better.

Let's give an overview first, and then talk about it in detail with the help of Ethernet.

basic concept

Topology

Basically, the basic topology of Ethernet is:

Star, bus, ring, tree

So four.

Media Access Control Mode

CSMA\CD
- This is an important knowledge point, so I will talk about it elsewhere.
token bus
- All workers are arranged in a logical ring in order, and only token holders can control the bus to send information.
token ring

LAN classification

In fact, there are many architectures for LAN, and of course Ethernet is the most used. The advantages of different architectures are not the same, each has its own characteristics.

ethernet
Token Ring
FDDI网 Fiber Distributed Data Interface
ATM network
WLAN

IEEE 802

It is a series of protocol sets, or standard sets.

IEEE 802, also known as LMSC (LAN/MAN Standards Committee, LAN/MAN Standards Committee), is committed to studying the services and protocols defined in the physical layer and MAC layer of LAN and MAN, corresponding to the minimum OSI network reference model Two layers (i.e. physical layer and data link layer).

802.3: Ethernet
802.5: Token Ring
802.11：WLAN

MAC sublayer and LLC sublayer

The IEEE802 series of standards divide the data link layer into two sublayers: LLC (Logical Link Control, Logical Link Control) and MAC (Media Access Control, Media Access Control).

The upper LLC sublayer implements the hardware-independent functions of the data link layer, such as flow control, error recovery, etc.; the lower MAC sublayer provides the interface between LLC and the physical layer.

MAC

The MAC sublayer is responsible for forming the "0" and "1" bit streams of the physical layer into a frame, and performing error checking through the error checking information at the end of the frame; providing access methods to shared media, including Ethernet with conflict detection Carrier Sense Multiple Access (CSMA/CD), Token Ring (Token Ring), Fiber Distributed Data Interface (FDDI), etc.

The MAC sublayer assigns a separate LAN address, which is commonly referred to as a MAC address (physical address). The MAC sublayer adds the physical address of the target computer to the data frame. When the data frame is passed to the peer MAC sublayer, it checks whether the address matches its own address. If the address in the frame does not match its own address, If there is a match, the frame is discarded; if there is a match, it is sent to the upper layer

LLC

The LLC sublayer is defined in the IEEE802.2 standard and is shared by the 802 standard series; while the MAC sublayer protocol depends on its own physical layer.

The MAC layer of IEEE802.15.4 can support multiple LLC standards. The SSCS (Service-Specific Convergence Sublayer) protocol carries the IEEE802.2 Type 1 LLC standard, and also allows other LLC standards to directly use IEEE802. 15.4 MAC layer services.

The LLC sublayer provides a unified logical view to the network layer.

LLC is responsible for identifying network layer protocols and then encapsulating them. The LLC header tells the link layer what to do with the frame when it is received.

The main function:

Transmission reliability assurance and control;
Fragmentation and reassembly of data packets;
Sequential transmission of packets.

Serve:

No acknowledgment for a connectionless service. This is a packet type service.
Connected services. This service is similar to that provided by HDLC.
There is a service that confirms no connection. Delivers a packet with acknowledgment, but does not establish a connection.

The LLC sublayer maintains a list of functions indexed by DSAP. When a data packet is received, the corresponding function is called with the index of DSAP, and the function hangs the data packet to the corresponding receiving queue.

Detailed explanation of Ethernet

Ethernet is the most ubiquitous type of computer network in the real world. There are two types of Ethernet: the first type is classic Ethernet, and the second type is switched Ethernet, which uses a device called a switch to connect different computers.

The standard topology of Ethernet is a bus topology, but Fast Ethernet (100BASE-T, 1000BASE-T standard) uses switches for network connection and organization in order to reduce conflicts, maximize network speed and use efficiency. In this way, the topology of Ethernet becomes a star; but logically, Ethernet still uses bus topology and CSMA/CD (Carrier Sense Multiple Access/Collision Detection, that is, carrier multiple access/collision detection) bus technology.

Ethernet provides connectionless, unreliable service

Ethernet only achieves error-free reception, not reliable transmission.

Unreliable transport service: only discard frames with errors, and do nothing else
Reliable transmission: find a way to realize what the sender sends and what the receiver receives

Error control is controlled by the upper layer.

10BASE-T

basic concept:

10	BASE	T
Transmission rate	The baseband signal is transmitted	Represents the use of twisted pair

Physically, it is a star topology, but logically uses a bus structure; uses Manchester encoding; CSMA\CD control, and collisions will occur.

Adapter and MAC address

In fact, the interaction between our computer and the network is through this communication adapter, that is, our network interface card, network card.

A network card also has its own CPU, RAM and ROM.

Each network card will be equipped with a unique MAC address when it leaves the factory, and it will be recorded in its own ROM, and each network card will be identified by this MAC address.

In a local area network, hardware address = physical address = MAC address. A MAC address is an identifier (in fact, it can be repeated, this MAC address, because MAC addresses generally appear in the local area network, as long as there is no confusion, but for some cloud services, repetition is not very good).

Things like the mouse also have their own network card and MAC address (you can even use the login route to modify the address)

Ethernet MAC frame

Destination address	source address	type	data	FCS
6	6	2	46-1500 contains IP datagrams	For verification, the length is 4

Because the length of our MAC address is 6, the destination address and source address are represented by 6 bytes.

Generally speaking, a preamble of 8 bytes is added in front of the Ethernet MAC frame . The first 7 bytes represent the preamble, and the last byte represents the frame start delimiter.

Unicast
Broadcast, the destination address is all 1
multicast

If Manchester is used, there is no end character, because we use a special level to represent data, which is represented by a transition in the middle. If the transmission ends, there will be no such feature.