Seamless Wi-Fi roaming: theory in practice. One WiFi network over a large area (seamless WiFi) Seamless roaming wifi d link

Nowadays, various wireless devices for which high-speed network access is possible only via WiFi are gaining popularity. These are Ipad/Iphone, and other mobile gadgets. When you want to organize WiFi access in an area of ​​30 sq. m., then installing a regular Dlink for 1200 rubles will solve all your problems, but if you have an area >500 sq. m. m. and this is only for one floor, this solution will not work. If you use regular access points or routers, then each router will have its own network name (unique SSID) or the routers will need to be spaced far apart so that coverage areas do not overlap, and this will lead to the appearance of zones with very poor reception quality, or none at all signal. About six months ago, I encountered the same problem, a solution was found quite quickly - UniFi.

Example of installing WiFi UniFi in a car service center with several buildings.

UniFi provides wireless coverage to the Arcadia School District in California (translation).

UniFi provides wireless access to upscale hotels in Peru (translation).

Possibilities of UniFi WiFi points:

One network for all WiFi points.

Attractive design.

Easy to install, PoE.

Displays coverage area and access point locations on the administrator display.

Centralized wireless network management.

Guest networks, without access to the local network.

Create temporary passwords for guest users.

Automatic software updates on access points.

High scalability: up to 100 or more points.

Multiple wireless networks with differentiated access rights.

Separation of network user traffic by VLAN.

Fast intranet roaming when switching between access points.

Monitoring user traffic, identifying sources of increased network load.

Large coverage area.

Possibility of generating one-time temporary passwords (relevant for public places: hotels, cafes, etc.)

Connecting points in repeater mode.

An overview of the capabilities of the UniFi Controller is here.

Implementation of WiFi from Ubiquity in hotels in Peru here (translation).

Hardware controller for Ubiquiti UniFi. UniFi Cloud Key.

What it looks like in practice:

A software controller is installed on one of the network computers, on which all wireless network settings are made.

All settings of points and network parameters are subsequently made through this controller. Below are a couple of screenshots of the settings and appearance.

This is a building plan showing the locations of the points.

Setting up a guest network without access to corporate resources.

Monitoring active clients.

Access point monitoring.

View from above.

The installation and configuration process is extremely simple:

1. Place the points and connect them to the local network; UniFi supports PoE so that you only need an ethernet socket to connect them.

2. Install the software controller on any computer on the network, configure the parameters of WiFi networks, initialize the points, after initialization, the settings from the controller will be applied to the point, and the point will be ready for operation. Even when the controller is turned off, the settings on the points are saved.

We understand roaming technologies (Handover, Band steering, IEEE 802.11k, r, v) and conduct a couple of visual experiments demonstrating their work in practice.

Introduction

Wireless networks of the IEEE 802.11 standards group are developing extremely quickly today, new technologies, new approaches and implementations are appearing. However, as the number of standards grows, they become increasingly difficult to understand. Today we will try to describe several of the most common technologies that relate to roaming (the procedure for reconnecting to a wireless network), and also see how seamless roaming works in practice.

Handover or "client migration"

Once connected to a wireless network, the client device (be it a smartphone with Wi-Fi, a tablet, a laptop or a PC equipped with a wireless card) will maintain a wireless connection if the signal parameters remain at an acceptable level. However, when the client device moves, the signal from the access point with which the connection was originally established may weaken, which will sooner or later lead to a complete inability to transmit data. Having lost connection with the access point, the client equipment will select a new access point (of course, if it is within reach) and connect to it. This process is called handover. Formally, handover is a migration procedure between access points, initiated and performed by the client himself (hand over - “transfer, give, concede”). In this case, the SSIDs of the old and new points do not even have to match. Moreover, the client may be on a completely different IP subnet.

To minimize the time spent reconnecting a subscriber to media services, it is necessary to make changes both to the underlying wired infrastructure (make sure that the client's external and internal IP addresses do not change) and to the handover procedure described below.

Handover between access points:

1. Determine a list of potential candidates (access points) for switching.
2. Set the CAC status (Call Admission Control - call availability control, that is, essentially, the degree of device load) of the new access point.
3. Determine the moment to switch.
4. Switch to new access point:

In IEEE 802.11 wireless networks, all handover decisions are made by the client side.

Source: frankandernest.com

Band steering

Band steering technology allows the wireless network infrastructure to move a client from one frequency range to another, usually by forcing the client from the 2.4 GHz band to the 5 GHz band. Although band steering is not directly related to roaming, we decided to mention it here anyway since it is related to client device switching and is supported by all of our dual-band access points.

In what case might it be necessary to switch a client to another frequency range? For example, such a need may be associated with transferring a client from the congested 2.4 GHz band to the freer and higher-speed 5 GHz. But there are other reasons.

It is worth noting that at the moment there is no standard that strictly regulates the operation of the described technology, so each manufacturer implements it in its own way. However, the general idea remains roughly the same: access points do not advertise an SSID in the 2.4 GHz band to a client performing an active scan if that client has been observed to be active on the 5 GHz band for some time. That is, access points, in fact, can simply remain silent about the presence of support for the 2.4 GHz band, if it was possible to establish that the client supports the 5 GHz frequency.

There are several operating modes of band steering:

1. Forced connection. In this mode, the client is basically not informed about the presence of support for the 2.4 GHz band, of course, if the client has support for the 5 GHz frequency.
2. Preferred connection. The client is forced to connect to the 5 GHz band only if the RSSI (Received Signal Strength Indicator) is above a certain threshold, otherwise the client is allowed to connect to the 2.4 GHz band.
3. Load balancing. Some clients that support both frequency bands connect to the 2.4 GHz network, and some connect to the 5 GHz network. This mode will prevent overloading the 5 GHz band if all wireless clients support both frequency bands.

Of course, clients with support for only one frequency range will be able to connect to it without problems.

In the diagram below we tried to graphically depict the essence of band steering technology.

Technologies and standards

Let us now return to the process of switching between access points. In a standard situation, the client will maintain the existing association with the access point for as long as possible (as possible). Exactly as long as the signal level allows it. As soon as the situation arises that the client can no longer support the old association, the switchover procedure described earlier will begin. However, handover does not happen instantly; it usually takes more than 100 ms to complete, and this is already a noticeable amount. There are several IEEE 802.11 working group radio resource management standards aimed at improving wireless network reconnection time: k, r, and v. In our Auranet line, 802.11k support is implemented on the CAP1200 access point, and in the Omada line, 802.11k and 802.11v protocols are implemented on the EAP225 and EAP225-Outdoor access points.

802.11k

This standard allows a wireless network to report to client devices a list of neighboring access points and the channel numbers on which they operate. The generated list of neighboring points allows you to speed up the search for candidates for switching. If the current access point's signal weakens (for example, the client is removed), the device will search for nearby access points from this list.

802.11r

Version r of the standard defines the FT - Fast Transition (Fast Basic Service Set Transition) function, which allows you to speed up the client authentication procedure. FT can be used when switching a wireless client from one access point to another within the same network. Both authentication methods can be supported: PSK (Preshared Key) and IEEE 802.1X. Acceleration is achieved by storing encryption keys on all access points, that is, the client does not need to go through the full authentication procedure using a remote server when roaming.

802.11v

This standard (Wireless Network Management) allows wireless clients to exchange service data to improve the overall performance of a wireless network. One of the most used options is BTM (BSS Transition Management).
Typically, a wireless client measures its connection to an access point to make roaming decisions. This means that the client has no information about what is happening with the access point itself: the number of connected clients, device loading, scheduled reboots, etc. Using BTM, the access point can send a request to the client to switch to another point with better operating conditions , even with a slightly worse signal. Thus, the 802.11v standard is not directly aimed at speeding up the switching process of a client wireless device, but in combination with 802.11k and 802.11r it provides faster program performance and improves the convenience of working with wireless Wi-Fi networks.

IEEE 802.11k in detail

The standard extends RRM (Radio Resource Management) capabilities and allows 11k-capable wireless clients to request a list of neighboring access points that are potential handover candidates from the network. The access point informs clients about 802.11k support using a special flag in the Beacon. The request is sent in the form of a management frame called an action frame. The access point also responds with an action frame containing a list of neighboring points and their wireless channel numbers. The list itself is not stored on the controller, but is generated automatically upon request. It is also worth noting that this list depends on the client’s location and does not contain all possible wireless network access points, but only neighboring ones. That is, two wireless clients located in different places will receive different lists of neighboring devices.

With such a list, the client device does not need to scan (actively or passively) all wireless channels in the 2.4 and 5 GHz bands, which reduces the use of wireless channels, that is, freeing up additional bandwidth. Thus, 802.11k allows you to reduce the time spent by the client on switching, as well as improve the process of selecting an access point to connect to. In addition, eliminating the need for additional scans helps extend the battery life of the wireless client. It is worth noting that access points operating in two bands can provide the client with information about points from an adjacent frequency range.

We decided to clearly demonstrate the operation of IEEE 802.11k in our wireless equipment, for which we used an AC50 controller and CAP1200 access points. The source of traffic was one of the popular instant messengers that supports voice calls, running on an Apple iPhone 8+ smartphone that obviously supports 802.11k. The voice traffic profile is presented below.

As can be seen from the diagram, the codec used generates one voice packet every 10 ms. The noticeable spikes and dips in the graph are explained by the slight variation in latency (jitter), which is always present in Wi-Fi-based wireless networks. We configured traffic mirroring on , to which both access points participating in the experiment are connected. Frames from one access point went to one network card of the traffic collection system, frames from the second - to the second. In the resulting dumps, only voice traffic was selected. The switching delay can be considered the time interval that elapses from the moment traffic disappears through one network interface until it appears on the second interface. Of course, the measurement accuracy cannot exceed 10 ms, which is due to the structure of the traffic itself.

So, without enabling support for the 802.11k standard, wireless client switching occurred on average within 120 ms, while activating 802.11k reduced this delay to 100 ms. Of course, we understand that although switching latency has been reduced by 20%, it is still high. Further reductions in latency will be possible by using the 11k, 11r and 11v standards together, as is already implemented in the home series of wireless equipment.

However, 802.11k has another trick up its sleeve: timing of the switch. This feature is not so obvious, so we would like to mention it separately, demonstrating its operation in real conditions. Typically, the wireless client waits until the last minute, maintaining the existing association with the access point. And only when the characteristics of the wireless channel become completely bad does the switching procedure to a new access point start. Using 802.11k, you can help the client with switching, that is, offer to do it earlier, without waiting for significant signal degradation (of course, we are talking about a mobile client). Our next experiment is devoted to the moment of switching.

Qualitative experiment

Let's move from a sterile laboratory to a real customer site. Two 10 dBm (10 mW) access points, a wireless controller, and the necessary supporting wired infrastructure were installed in the room. A diagram of the premises and installation locations of access points are presented below.

A wireless client moved around the room making a video call. First, we disabled 802.11k support in the controller and set the locations where switching occurred. As can be seen from the picture below, this happened at a considerable distance from the “old” access point, near the “new” one; in these places the signal became very weak, and the speed was barely enough to transmit video content. There were noticeable lags in voice and video when switching.

We then enabled 802.11k support and repeated the experiment. Now the switching occurred earlier, in places where the signal from the “old” access point was still strong enough. There were no lags in the voice or video recorded. The switching point has now moved approximately to the middle between access points.

In this experiment, we did not set ourselves the goal of finding out any numerical characteristics of switching, but only qualitatively demonstrating the essence of the observed differences.

Conclusion

All the described standards and technologies are designed to improve the client’s experience of using wireless networks, make his work more comfortable, reduce the influence of irritating factors, and increase the overall performance of the wireless infrastructure. We hope that we were able to clearly demonstrate the benefits that users will receive after implementing these options in wireless networks.

Is it possible to live in an office without roaming in 2018? In our opinion, this is quite possible. But, having once tried to move between offices and floors without losing the connection, without having to re-establish a voice or video call, without being forced to repeatedly repeat what was said or ask again, it will be impossible to refuse.

P.S. but this is how you can create seamlessness not in the office, but at home, which we will talk about in more detail in another article.

802.11R. Fastswitching between points (handover)

Many Wi-Fi manufacturers promise seamless switching between access points using their “brilliant” proprietary protocol.

Despite the beautiful promises, in practice, delays during switching (handover) can be significantly longer than the stated 50-100 ms (switching can take up to 10 seconds when using the WPA2-Enterprise protocol). The fact is that the decision to switch to another access point is always made by the client equipment. Those. Your smartphone, laptop or tablet decides when to switch and how to do it.

Often the proprietary protocols of well-known Wi-Fi manufacturers are based on forced de-authentication of the device when the signal quality deteriorates. Sometimes in the settings of a Wi-Fi point you can set “roaming aggressiveness” - the minimum signal value at which the device will be “thrown out” of the network. Often client equipment does not react correctly to such a “kick in the ass.” The TCP session is terminated and file downloading stops. The connection to the mail server or virtual machine is lost. Connecting to the SIP server requires re-authentication.

Quite often, the client device, instead of connecting to a nearby point with a better signal ( pushes him to this decisionWiFicontroller) tries to re-establish the connection with the previous point to no avail. It’s even worse if the device tries to connect to another network from the list of saved ones (for example, a guest network).

But even if the switching process goes according to plan, repeated key exchange (EAP) and authorization on the Radius server (WPA-2 Enterprise) take up significant time.

To solve these problems, the Wi-Fi Association developed the 802.11R protocol. Currently, most mobile devices support it (Apple starting from iPhone 4S, Samsung Galaxy S4, Sony Xperia Z5 Compact, BlackBerry Passport Silver Edition,...)

The essence of 802.11R is that the mobile device knows its own and foreign points by the mobile domain membership signal (MDIE). This signal is added to the SSID beacon signal.

If your iPhone sees a point from its mobile domain with the best signal/noise level, before starting the switching procedure along the existing “thread” it carries out a preliminary authorization with another point in the mobile domain.

Secondly, authorization takes place according to a simplified scenario - instead of a long authorization on the Radius server, the client device exchanges the PMK-R1 key with the Wi-Fi controller. (The initial key PMK-R0 is transmitted only during primary authentication and is stored in the memory of the Wi-Fi controller).

At the moment when another point “retrospectively” authorized the device, the actual handover occurs. Reconfiguring the frequency and channel in a smartphone takes no more than 50 milliseconds. In most cases, it goes completely unnoticed for the user.

When choosing a solution for an office Wi-Fi network, pay attention to whether the selected equipment supports the open 802.11R roaming protocol, which is understandable for client devices. For example, Edimax Pro equipment fully supports this protocol, so in most cases there are no problems with roaming. However, if your device is old and does not understand the 802.11R protocol, it is possible to adjust the aggressiveness of roaming based on the signal dropping below a threshold - as other Wi-Fi manufacturers do, touting it as an "innovative solution".

802.11 K.Load Balancing in a Wireless Network

In addition to roaming problems, corporate users often have to deal with congestion at one access point. In a classic Wi-Fi implementation, all devices try to connect to the access point with the best signal. Sometimes, as a result of an incorrect location of the point (radio planning error), all the “office inhabitants” are registered at one point, while the rest are “resting”.

Due to the uneven load, the speed of the local network drops significantly, since the radio broadcast is one large “hub” where devices “speak in turn.”

To smooth out unevenness and optimally distribute users between points operating on different radio channels, the 802.11K protocol was developed.

802.11K works in conjunction with 802.11R (as a rule, devices that support the “R” standard also support the “K” standard).

If a mobile device “sees” a beacon signal from other points in the same mobile domain, the device sends a broadcast request “Radio Measurement Request frame”, which requests information about the current state of other access points within the visibility range:

number of registered users

average channel speed (number of packets transmitted)

how many bytes were transferred in a certain time interval

In the extended specification of the standard, the client's smartphone can query the channel status of other mobile devices connected to a potentially interesting access point that support the 802.11K standard. Devices respond not only with real statistics, but also with signal/noise status.

Thus, if your smartphone sees 2 or more points within one mobile domain, it will choose not the point with the best signal, but the point that will provide higher connection speed to the local network (less congested).

Reception conditions, the number of users and the load at the point can change dynamically, but using the 802.11K and 802.11R protocols, devices will switch seamlessly and the network load will always be distributed evenly.

Many manufacturers using proprietary protocols implement something similar to 802.11K, where an “overloaded” point forcibly disconnects clients with worse reception conditions or limits the maximum number of simultaneously registered devices and disables registration if the number of clients exceeds acceptable limits. These proprietary protocols are not as effective, but still prevent the Wi-Fi network from collapsing completely.

How to save on radio planning thanks to802.11K

The use of equipment that supports the 802.11R and 802.11K protocols partially corrects errors made during radio planning. Dynamic protocols with roaming support make it possible to prevent overloads of individual points and distribute the load between points evenly throughout the network.

The WiFi-solutions team recommends that you always do radio planning, but sometimes in small networks, you can arrange the points chaotically. Dynamic protocols will improve Wi-Fi quality and load distribution between channels of neighboring points.

The use of dynamic protocols for seamless roaming reduces overlap areas. Thus, high-quality coverage can be ensured with a smaller number of points. Savings on equipment - up to 25%.

I need some advice. Contact me.

These questions are often asked by customers who need complex organization of corporate networks over a large enterprise area.

Based on Wifi, a technology has been proposed for providing access to a corporate computer network in warehouses, offices and production facilities.

As you know, Wifi is a wireless communication standard on unlicensed frequencies. The disadvantages of this standard include the limited range of individual access points; this problem is solved by combining individual networks into one multinetwork.

Seamless roaming Users are connected to local, zonal multinetworks from multiple providers. This approach is currently being implemented for heterogeneous networks, for example, to combine WiMAX, Wi-Fi, GSM, CDMA, GPRS, UMTS services.

In case of WiFi seamless roaming there is a combination of different access points and ensuring the subscriber’s transition between Wi-Fi networks is imperceptible to the user.

In general, seamless wifi roaming ensures uninterrupted connection of subscribers when crossing network boundaries.

Seamless access technology has already been proposed by a number of companies. For example, Cisco has developed “seamless roaming” for heterogeneous Wi-Fi networks; Motorola, Microtik and Aruba are actively involved in seamless Wi-Fi roaming in corporate format. These are perhaps the most striking offers on the market today, so we will try to compare them on two application tasks, deploying seamless wifi in a warehouse and wifi for hotels.

Basic elements of seamless wifi technology

The development of “seamless” technologies and networks is the main trend of modern technological development.

In modern networks, manufacturers are trying to combine the computing power of the network into both homogeneous (of the same type) and heterogeneous (of different types) infrastructures. This approach is dictated by a wide variety of network standards, including both packet and circuit switched.

For zone networks, such solutions are usually called multiservice networks. For local corporate networks, there are a number of network applications between which it is necessary to ensure coordination and unconditional access without the user noticing.

The technology is implemented using special software that stores the client’s IP address in a local or zone network, which simultaneously allows for guaranteed data delivery and uninterrupted traffic when moving between networks.

Thus, uninterrupted operation of applications is also implied.

Currently, the paradigm of “seamless WiFi access” is developing, which is implemented within the framework of virtual local networks VLAN - Virtual LAN.

Seamless wifi roaming Motorola, Microtik, Aruba

If we talk about the proposed technological solutions, then it makes sense to pay attention to three companies operating in this segment - Motorola, Microtik, Unifi, these companies actively compete with each other. The technology features and the idea are borrowed from mobile networks, in which a similar seamless roaming function is known as the “handover function.”

As a result of the implementation of seamless roaming, access to the network is provided without a single break when moving between networks. The technology is implemented using special network equipment.

Seamless wifi roaming Motorola, Microtik, Aruba offers similar basic functions: default operation in Bridge/Router modes, DHCP recovery function and the presence of DHCP Relay Option 82.

If you need to deploy wifi for hotels, it still makes sense to choose Motorola wifi, which implements:

• access via HTTP / HTTPS, SSHv2, Telnet, SNMP (v2c, v3)
• Captive Portal feature that manages user accounts and encrypts traffic.

If you use Microtik seamless wifi roaming, there are also quite wide possibilities, especially for network equipment available on the market; user authentication is then implemented based on third-party software.

Seamless wifi roaming Unifi is in many ways a cheap and unstable replica that actively offers its equipment for deploying complex Wi-Fi networks. We do not recommend this equipment.

At the same time, I would still like to highlight the functions of Motorola’s seamless wifi, which, using “native” software and a radius server, routes traffic in wireless multinetworks, supports built-in subscriber billing with high-quality authentication and packet encryption (WEP, WPA, WPA2).

This option is especially recommended for corporate networks when it is necessary to provide network access throughout the company, including when you need to install wifi in a warehouse, production or office in a protected mode with reduced risks of traffic interception.

For full-featured multi-service wifi networks, Motorola, using SMART RF, implements the ability to select channels and power levels on a port in Motorola WiFi routers, which allows you to configure and optimize traffic.

Thus, VLAN (virtual local network) allows you to solve a maximum of corporate problems and implement them on the basis of a wireless network.

Seamless roaming wifi mikrotik allows you to organize inexpensive wireless networks, but is still significantly inferior to Motorola, despite technologically similar offers.

All companies offer a complete set of hardware equipment that allows you to deploy “large” wifi networks, combining individual access points into one network and providing efficient routing.

At the same time, the optimal amount of functions required in modern corporate networks is implemented only in Motorola wifi; of course, this largely concerns network security and access blocking.

At the same time, for example, motorola wifi seamless roaming can be recommended as an inexpensive solution with a full range of functions for organizing wifi for hotels. Again, this is relevant both for hotel packages and for small hotels that are not particularly concerned about the security of intercepting customer traffic on the internal network.

It can be said about the technologies offered by the above companies that they continue to develop, updates and new hardware firmware are released every six months. Moreover, all solutions are offered so that users can constantly update their networks, including using legacy equipment that does not interfere with the implementation of new functions.

If we still consider a wireless local network based on the equipment of the proposed companies, then Motorola has offered a much more developed version - Motorola wifi is recommended by our company by default.

In this article we will learn how to create a single seamless WiFi network on MikroTik / Mikrotik routers. Where can this be useful? For example, in various kinds of cafes or hotels, where one Wi-Fi router is not enough to cover all the premises and access the Internet, and with a large number of access points, various kinds of problems constantly arise: on laptops the connection constantly disappears, and mobile devices do not switch to the nearest access point.

The solution to this situation is seamless WiFi network roaming or handover, which we can get thanks to the CapsMan functionality from several Mikrotik routers, one of which will be a WiFi controller, and the rest will be access points controlled by this controller.

The first thing you need to do is update to the latest software version. The firmware can be downloaded from the official website. Next, going into the MikroTik interface, drag it to the Files section and reboot the router. Along with the firmware, you also need to download the Wireless CAPs MAN package, drag it to the same place and reboot. After completing the steps, you can proceed to configuration.

Let's start with the controller. Open the CAPsMAN section by clicking the corresponding button in the main menu. In the Interfaces tab, click the Manager button (turn on controller mode) and in the window that appears, check the Enable box and save OK. After that, go to the Configurations tab.

The configuration settings will apply to all access points connected to the controller. Click the blue cross and in the Wireless tab indicate the configuration name (3), wireless network mode (4), network name (5), and also turn on all wireless antennas for reception and transmission (6), save (7) and go to the Channel tab .

Here we indicate the frequency (2), wireless network broadcast format (3) and channel (4). Save (5) and go to the Datapath tab.

Here we only need to check the Local Forwarding box - this will transfer traffic control to the access points. All that remains is to fill out the last Security tab.

In the security section, select the authentication type, encryption method and password for the wireless network, click OK.

After we have created the configuration, we move on to the next step - deployment. In the same CAPsMAN section, select the Provisioning tab (1) and click the blue cross. The Radio MAC field (2) allows us to select a specific access point to which our deployment will relate. We leave it as default so that the deployment applies to all access points. In the next Action field (3) select createdynamicenabled, since we have a dynamic interface. In Master Configuration (4) we indicate the name of the configuration created above.

We’re done with the CAPsMAN section, let’s move on to the Wireless section (1). In the Interfaces tab, click the CAP button (3), check the Enabled checkbox (4), select the wlan1 interface and indicate the IP address of our main router, which is also the controller.

If we did everything correctly, then two red lines will appear in the Interfaces tab, which indicate that the Wi-Fi adapter has connected to the controller and adopted all the necessary settings.

This completes the configuration of the main router-controller, and this network can be used to create a telephone network and connect to an office PBX

Setting up access points that will connect to the controller via an Ethernet cable is quite simple. They also need to be updated to the latest version and CAPs MAN installed. Next, we combine all the ports and wi-fi interface into one Bridge in the section of the same name.

The next step in the Wireless section is to do the same as on the controller, except that instead of the IP address in CAPs MAN Addresses, we indicate the Bridge created on the access point in the Discovery Interfaces field. After the manipulations have been completed, the access point will receive settings from the controller and will distribute wi-fi (the same two red lines should appear in the Interfaces tab).