9

I am studying for CCNA. In my book I read that wireless access points belong to OSI layer 2. I did some research and I found that is correct. This raised lots of questions in my head so I hope I don't get on your nerves :P

I know that these devices are connected to switches or routers, I also know that AP's use CSMA-CA (at layer 2).

  1. Is this the only reason why they are Layer 2?

  2. Is there some kind of switching table inside the AP where it associates MAC's with output interfaces? What logical stuff does an AP have inside it? I ask this because if it has to forward packets it has to choose whether to send to the air or to the switch or router direclty connected to it.

jwbensley
  • 5,290
  • 7
  • 37
  • 79
BrunoMCBraga
  • 811
  • 3
  • 9
  • 13
  • 1
    Others are providing more details answers, a simple view though is that they are wireless switches (which as you know are layer 2 devices). ##Disclaimer - There are many kinds of AP and switch, this is a one size fits all comment, its not always accurate## – jwbensley Aug 07 '13 at 10:13
  • 1
    802.11 is said to operate at layer 2, but it is one hell of a layer 2, it even provides routing via 802.11s. – BatchyX Aug 10 '13 at 14:43

4 Answers4

8

Strictly speaking, access points are a L2 device. Their primary function is to bridge 802.11 WLAN traffic to 802.3 Ethernet traffic.

However, in the real world, enterprise wireless vendors often push more functionality to either the AP itself and/or tie them into a controller, with the end result that they often incorporate functionality from higher layers as well.

I don't entirely follow #1, but neither the connection to a switch or router nor the use of CSMA-CA has anything to do with their function as a L2 device. It is a L2 device because that is the layer of the network where it is designed to function. It doesn't care nor need L3 or above to operate and allow devices to communicate (again strictly speaking).

As for #2, yes, an AP (or any bridge) needs to keep track of which interface any individual device is connected. In general (and simply), they work on the principle of frames destined to an associated station gets forwarded out the wireless interface and any other frames get forwarded out the wired interface (or sent to the controller).

YLearn
  • 27,141
  • 5
  • 59
  • 128
  • As an article (http://blogs.aerohive.com/blog/the-wireless-lan-architecture-blog-3/enterprise-wireless-networking) states, pure wireless access points are mixed L1/L2 devices: L1 hub on wireless side and L2 bridge on wired side. – Paul May 29 '15 at 14:47
  • I think you may have linked the wrong blog post. While I would agree the wireless side of the access point operates like a L1 hub because of the medium, I would still contend that an access point is a L2 device. In reality, an AP is a L2 bridge that is bridging 802.3 Ethernet traffic to 802.11 traffic. When bridges were sold as a product to separate broadcast domains, they often operated like half-duplex hubs on both sides, but they are still L2 devices. Similarly, just because a switch operates like a hub on "one side" when connected to a hub, it is not considered a L1/L2 device. – YLearn May 29 '15 at 16:40
  • 1
    Looks like you probably meant [this blog post](http://blogs.aerohive.com/blog/the-wireless-lan-architecture-blog-3/what-is-a-wireless-access-point). After reading it, my assessment still stands. An AP is a single device with different behaviors on each side, not two devices in one "chassis." – YLearn May 29 '15 at 22:05
  • Yes, you are right. Sorry for link mistyping. – Paul May 30 '15 at 16:08
4

1) Access points normally operate in L2. There are some exceptions to this, but let's stick to the rule. They are L2 devices cause they are forwarding frames based on decisions made by looking at the mac-address table they have. When some device connects on the wired or wireless interfaces, the AP learns its mac address from the broadcasts and stores the mac address in a list, depending from which interface the address was learnt. So when a frame arrives at the AP it will look up the mac table and decide on which interface it will be sent.

2) Regular access points shared the same logic as the switches. As I mentioned above, they have a mac address table were they store mac address associations depending on the interface the mac was learnt.

trendy
  • 446
  • 2
  • 6
2

AP serves as a connectivity provider for clients on wireless, it maintains MAC entries against each client. It provides IP using DHCP server on LAN side. Forwards traffic on WAN side (can be on static or dynamic IP). Works on Layer 2 and mostly can be said as connectivity provider between wired and wireless.

engineerbaz
  • 165
  • 1
  • 10
-1

I am just learning about L2, but I think whether "Ethernet" AP hardware uses MAC tables or IP addresses to determining routing is a moot question. The point is that an Ethernet network establishes a unique correspondence between the MAC and IP address, so in principle an AP could route based on either. Hardware that routes on MAC however could be more flexible, for example in setting up "transparent" bridges connected to LAN endpoints.

Kenneth Murata
  • 1
  • 5
    Of course, Wi-Fi (IEEE 802.11) is _not_ ethernet (IEEE 802.3). The frames are incompatible. WAPs are translating bridges, not transparent bridges. "_The point is that an Ethernet network establishes a unique correspondence between the MAC and IP address_" Ethernet knows nothing about IP. Ethernet can carry any number of layer-3 protocols, and it simply doesn't care. – Ron Maupin Feb 03 '17 at 18:27