Easy-to-use WLAN switch for fast WLAN networking

When we build a wireless network, WLAN switches are like a mainstay and cannot be ignored. Therefore, setting a vswitch quickly means that a WLAN Network is successfully created.

A wlan switch can connect to a WLAN Access Point (AP) through a wired connection (with a switch port ). They can also connect to the enterprise network through their other WLAN switch ports. These switches are the "gateways" connecting to the enterprise's wired network-all data frames from the WLAN client must be sent to the enterprise network through the WLAN switch.

To understand the functions of a WLAN switch and Its Application in the network, you must first understand the network architecture of the WLAN and the functions of the Access Point. We can regard a WLAN switch as a control device and an AP as a wireless terminal. This article analyzes the WLAN Network Architecture and AP and controller functions in detail to clarify the functions of the WLAN switch and controller. This article also introduces different functions of interfaces between controllers and APs. Later, this article will explain the variables related to the second/third-layer movement in the centralized architecture, and finally point out some common incorrect ideas and actual situations about these architectures. This article uses the term wireless terminal (WTP) to refer to the AP, and the term Access Controller (AC) to refer to the WLAN control function (whether deployed on a WLAN switch or an independent Controller ).

Main architecture of WLAN

There are three common WLAN network architectures:

1. autonomous architecture

2. centralized architecture

3. distributed architecture

The following sections describe these three architectures in depth.

Autonomous Architecture

In an autonomous architecture, WTP is fully deployed and supports client-side 802.11 features. Therefore, all the data frames on the wired LAN are 802.3 frames. Each WTP can be managed independently as a separate network entity on the network. Access Points in such a network are usually called "Fat AP ".

In the early stages of WLAN deployment, most APs are autonomous AP and can be managed as independent network entities. Over the past few years, the centralized architecture using AC and WTP (see below) has attracted more and more attention. The main advantage of centralized architecture is that it provides a structured and hierarchical control mode for network administrators for multiple WTP in enterprises.

Centralized Architecture

A centralized architecture is a hierarchical architecture that includes a WLAN controller responsible for configuring, controlling, and managing multiple WTP. A wlan controller is also called an Access Controller (AC ). 802.11 functions are shared by WTP and AC. Compared with the autonomous architecture, the WTP features in this model are weakened, so they are also called "thin AP ". Some functions on the AP are variable. For details, see the introduction below.

Distributed architecture

In a distributed architecture, different WTP establish a distributed network with other WTP through wired or wireless connections. A network composed of WTP is a typical example of this architecture. The WTP in the mesh can be connected to a 802.11 link or a wired 802.3 link. This architecture is usually used in the deployment of urban networks and other "Outdoor" components. The distributed architecture is not covered in this article.

WTP functions-fat, thin, and moderate AP

To understand the autonomous and centralized architecture, you must first analyze the functions implemented by the AP. Let's start with the fat AP, which forms the core of the autonomous architecture. Next we will introduce the thin AP, which is an important part of the centralized architecture based on WLAN network switches or controllers. This article will introduce the functions of a new component named "moderate AP. It is an AP specially optimized for a centralized architecture.

Fat Access Point

An AP is an addressable node in the network and has its own IP address on its interface. It can forward traffic between wired and wireless interfaces. It can also have multiple wired interfaces to forward traffic between different wired interfaces-similar to a layer-2 or layer-3 switch. The connection to the enterprise's wired network can be achieved through a layer 2 or Layer 3 network. It is worth noting that the fat AP does not "return" traffic to other devices through tunnel. This feature is very important and will be mentioned in this article when introducing other AP types. In addition, fat AP can provide "similar to vro" functions, such as Dynamic Host Configuration Protocol (DHCP) server functions.

AP management is implemented through a protocol (for example, Simple Network Management Protocol [SNMP], or Hypertext Transfer Protocol [HTTP] for Web management) and a command line interface. To manage multiple APs, the network administrator must use one of these management mechanisms to connect each AP. Each AP is displayed as a separate node in the network topology. Any node aggregation for management and control must be completed at the network management system (NMS) level. This includes developing An NMS application.

Fat AP also enhances a variety of features, such as the access control list (ACL) that allows filtering traffic on a specific WLAN client ). Another important feature of these devices is to configure and implement features related to QoS. For example, the traffic from a specific mobile base station may be higher than that of other traffic. Alternatively, you may need to insert and implement the IEEE 802.1p priority for the traffic from the mobile base station, or the differential service code point (DSCP ). All in all, because these aps provide many features of a WLAN switch or router, they can act as a WLAN switch or router to a certain extent.

The disadvantage of this AP lies in its complexity. Fat AP is usually built on powerful hardware and requires complex software. Because of the complexity, the installation and maintenance costs of these devices are high. However, these devices can also play a role in small networks. Some fat APS use a controller for the control and management functions at the backend. These controllers form a slightly simplified version of the fat AP-the so-called "moderate AP", which will be described in detail below.

Thin Access Point

As the name suggests, a thin AP aims to reduce the complexity of the AP. An important reason for simplifying AP is its location. Many enterprises adopt high-density installation for APs (because they are distributed in difficult areas) to provide the best RF connection for each base station. In special environments such as warehouses, this phenomenon is more obvious. For these reasons, network administrators want to install the AP only once without complicated maintenance. A thin AP is also known as a smart antenna. Its main function is to receive and send wireless traffic. They send wireless data frames back to a controller, process these data frames, and then switch to wired WLAN.

This AP uses a (usually encrypted) tunnel to send wireless traffic back to the Controller. The most basic thin AP does not even perform WLAN Network Encryption, such as Wired Equivalent encryption (WEP) or WiFi protected access (WPA/WPA2 ). This encryption is completed by the Controller-the AP is only responsible for sending or receiving encrypted wireless data frames, so as to maintain the simplicity of the AP and avoid the necessity of upgrading its hardware or software.

The advent of WPA2 makes encryption on the controller a very urgent task. Although WPA is compatible with WEP in hardware and only requires firmware upgrade, WPA2 is not backward compatible. Network administrators do not need to change the AP of the entire enterprise. Instead, they only need to send wireless traffic to the controller capable of WPA2 decryption, and then the data frame will be sent to the wired LAN.

The Protocol for transmission control and data traffic between the AP and the controller is dedicated. In addition, the AP cannot be managed as a unified entity on the second/third layer-it may be managed through the Controller, NMS can communicate with the Controller through HTTP, SNMP, or CLI/Telnet. One controller can manage and control multiple APs, which means that the Controller should be able to perform switching and routing functions based on powerful hardware. Another important requirement is that the connection and tunnel between AP and AC should ensure that the packet delay between the two entities remains at a very low level.

For a thin AP, QoS execution and ACL-Based Filtering in a WLAN are all handled by the Controller, which does not cause problems, because all data frames from the AP must be transmitted through the Controller in any situation. Centralized Control of ACL and QoS is not uncommon-this method is also used for networks using fat AP. This installation method uses the Controller as the gateway to manage traffic from the AP to the wired network. However, the controller function of the thin AP adopts a new method, especially in terms of data layer and forwarding function. The controller function is integrated into Ethernet switches connected to wireless networks and wired LAN networks-This gives birth to a series of devices called "WLAN switches. In this case, the wireless MAC Architecture is called the remote MAC Architecture. All 802.11 MAC functions are transferred to the WLAN controller, including latency-sensitive MAC functions.