WIFI Trying to improve

WIFI has tightly entered our lives. Most laptop users that I know have already installed, or plan to soon install an access point at home. There are many reasons for this, and the main one is that WI-FI gives users of mobile devices true mobility, but so far only within the limited coverage area. However, this technology is far from flaws, because we are developing a firmware that ensures the absence of some flaws within the cottage village (launch in June), and subsequently free access to downloading this firmware.


In this article, you will not find source codes or descriptions of the interaction of devices on the network, because the firmware is still in a closed test on a limited number of devices. Here you can read only the basic principles of work and problems to be solved. If after reading you have a desire to get acquainted with the firmware, or to contribute to the development of the project, you will find the contacts below.

So, in our opinion, the main problems of WI-FI are as follows:
• Limited coverage area.
• Attenuation of the signal with increasing distance of the receiver from the access point.
• Principle - "The lower the speed, the greater the distance."
• The rule that applies to most access points is “If two or more devices with different standards and access speeds work in the network, then the access point automatically selects an operation mode equal to the minimum device mode”.
• A similar rule for the connection speed is “The access point selects the exchange rate equal to the speed of the minimum of devices”.
• Lack of seamless roaming when using more than one access point.
• Channel congestion in networked areas, such as multi-story buildings.

I can anticipate comments - this does not apply to private users using WI-FI within their home, because for an apartment one access point is enough. And such comments will be fully justified. However, we are considering a network for a cottage village, in which the areas differ from the areas of urban apartments, and the coverage area is more global and amounts to several tens of hectares.

There is MESH or Mesh technology - a distributed construction topology, in which each access point performs, in addition to the main, the functions of a relay and router for other network participants. In our opinion, this scheme is the most promising for covering large areas with a stable link and the absence of “Dead zones”. (In more detail about this topology, you can find a lot of information on the Internet, or, if you have the proper amount of your comments, I’m ready to write a separate article on this issue.) However, with a closer acquaintance it turned out that in this network all devices should work on one channel and with one SSID, which is extremely undesirable in our version. Therefore, the first urges arose for the creation of “My” MESH network, taking into account these changes.

In addition, we investigated several access points from MTS in a vandal way, additionally providing access to other subscribers to the MTS-FON network, at your expense and without prior agreement with you. As one of the Soviet movie heroes said, “Shurik! This is not our method! ”, But it was also possible to borrow some of the best practices from there.

So - we proceed directly to the description of the network for which our firmware is being developed:
As in the MESH network, each of the access points acts as a repeater and router for other points of network participants, however, a special SSID is used to transfer service traffic to other networks. Also, when slower devices appear on the network than currently used by the main SSID, the point clones the parameters of the main SSID and creates a similar, but with the parameters of a slower device, switching the device by MAC address to the created SSID.

The following - all points belonging to the network exchange information on the service SSID with the server. This server contains information about the MAC addresses of devices connected to the network. Based on this information, a client device registered in the client device database can connect to any access point that belongs to the network.

At the moment, the number of SSIDs broadcast by one device is, in most cases, limited to 4, but this is mainly dictated by the weak hardware stuffing of modern budget devices. In this network, we have limited the maximum number of SSIDs to the same number. The list of devices for which the software is developed is also limited. Basically, we settled on D-Link devices (a total of 6 devices were tested, -4 were considered suitable, some with hardware improvements), Mikrotik (All), Zyxel (4 devices are in the test phase).

I would like to know your opinion about the firmware being developed and the prospect of such networks. You can also send your questions to mappix@valkiry.ru, I will try to answer everyone. We also invite WI-FI professionals and enthusiasts to interact.

Also popular now: