Chapter 8: Wireless LANs

Wireless LAN is the technology that enables computers to communicate without network cables. When laptop computers are getting popular, there comes the need for connecting them to a network. These computers are designed for mobile users and should not be bound to fixed wires. Wireless LAN protocols are defined in a series of standards that are collective called as IEEE 802.11 (802.11a, 802.11b, 802.11d, 802.11f, 802.11g). A street name for the technology is “WiFi” which stands for “Wireless Fidelity”.

    IEEE 802 Protocol Architecture

There are two kinds of wireless networks:

a) Ad-hoc networking (Peer-to-Peer networking): Every computer within certain limited area can communicate each other if they are equipped with wireless network interface cards. They can share files and printers, etc. In this mode, the computers may not be able to access “wired LAN” unless one of the computers is configured as a bridge to the wired LAN. This configuration requires a special software.

Ad-Hoc or Peer-to-Peer networking

b) Access Point (Base station) networking: This is also called “Infrastructure mode”. A device called “Access point” acts like a hub and provides connectivity for the wireless computers. An access point is essentially a repeater which receives a signal and broadcasts to all stations. There are two types of access points: a) Dedicated hardware access points b) Software access points which run on a computer equipped with a wireless network card.

          With Hardware Access Point

          With Software Access Point

8.1 IEEE 802.11

Wireless networking requires technology that deals with radio transmission to transfer data. The most widely adopted standard is IEEE 802.11. This standard defines all aspects of radio frequency Wireless networking.

Let’s look at the IEEE 802.11 protocol architecture.

As shown in the above figure, there are several options in the physical layer but they share the same MAC and LLC sublayers.

8.1.1 IEEE 802.11 Physical layer

IEEE 802.11 uses ISM(Industrial, Scientific, Medical) bandwidth as shown in the following diagram.

There are several choices currently in the physical layer and more may be coming. The details for these physical layer choices would be the subject for communication engineers. Let’s look at the overview of them.

a) Infrared: This uses the infrared beams with wavelength of 0.85 or 0.95 microns. Two speeds are defined, 1 Mbps and 2 Mbps. Infrared is restricted to one open space since it can not penetrate walls. This, due to its lower data rate and the interference from other light sources, is not popular.

b) FHSS (Frequency Hopping Spread Spectrum): This is one of the two techniques called “Spread Spectrum” technology. This uses 79 channels (each 1 MHz bandwidth) starting at 2.4 GHz ISM band. A station generates pseudorandom numbers using a seed (all stations use the same seed) and use the number to choose the next channel to transmit. Therefore, a message is divided into small fragments and each fragment is sent on a channel chosen by pseudorandom number. The amount of time used on each channel is an adjustable parameter called “dwell time”. An intruder will have difficulty since it may not know the current sequence number and the dwell time. The major drawback to FHSS is the limited data rate. Since FCC regulations require that the maximum occupied bandwidth for any single channel is 1MHz. This effectively limits the data rate through this type of systems about 1Mbps.

c) DSSS (Direct Sequence Spread Spectrum): This is the second of the “Spread Spectrum” technology. This is best described with an example: Figure 5.21 Stallings(6^th ed)

d) OFDM (Orthogonal Frequency Division Multiplexing): This is used in IEEE 802.11a that provides up to 54 Mbps in the 5 GHz ISM band. This uses 52 frequency bands (channels) are used.

e) HR-DSS (High Rate Direct Sequence Spread Spectrum): This is used in IEEE 802.11b and provides 11 Mbps using 2.4 GHz ISM band.

f) 802.11g: This is an enhanced version of 802.11a and uses OFDM but operates in 2.4 GHz ISM band. It is designed to achieve up to 54 Mbps.

8.1.2 IEEE 802.11 MAC layer

MAC sublayer handles the mechanism for sharing the medium. In the case of 802.11, the shared medium is “space”. The environment is quite different from Ethernet in two aspects.

The first difference is the “hidden station problem”. It is due to the fact that not all stations are in the range of each other.

Second is the “exposed station problem”. They are illustrated in the following figure.

(a) Hidden station problem (b) Exposed station problem

a) Hidden station problem: A wants to send B, so A listens to the medium and detects no transmission even though C is transmitting to B.

b) Exposed station problem: B wants to send to C, so B listens to the medium and detects A’s transmission (A is transmitting to D) and assumes that the transmission will fail.

 As we can see from the above two problems, Ethernet’s CSMA/CD can not be used in wireless LANs. The 802.11 standard defines two forms of medium access, Distributed Coordination Function (DCF) and Point Coordination Function (PCF). With two methods in DCF, there are three methods in total.

· Access methods

1) DCF CSMA/CA (mandatory)—Physical sensing

In this mode, when a station wants to transmit, it first listens to (physical sensing) the medium.  If the medium is busy, then it defers until the medium is free.

If the medium is free, either after deferral or prior to attempting to transmit again immediately after a successful transmission, the station (in fact all stations) must wait a pre-determined time period (DIFS: Distributed Inter Frame Spacing—several microseconds) and then uses “collision avoidance” by selecting a random backoff interval and waits for that interval before it transmits. However if another station begins transmission within the waiting period, the station goes back to the initial point. The receiving station sends an ACK frame to acknowledge the correct reception of the frame. If ACK is not received, then a collision is assumed and the frame is retransmitted. CSMA/CA is based upon the observation that the chance of collision is highest right after the medium becomes idle—since many stations may be waiting for the medium to go idle. So, make each of them wait for separately generated random amount of time.

2) DCF w/ RTS/CTS (optional)—Virtual sensing

Let’s say A wants to send to B and C is a station within range of A’s signal. D is a station within range of B but not within range of A.

This method requires a station to send a shot frame “RTS (Request To Send)” and the station must receive a “CTS (Clear To Send)” to be able to send. When the data is successfully transmitted, the receiver will send an ACK frame. If an ACK does not arrive within time, then the frame needs to be retransmitted.

Now, what happens to other stations such as C and D?

When A sends RTS, C may receive it and recognize that a station needs to access the medium, so C decides to wait. From RTS, C estimates the duration of the transmission (including ACK), so C asserts itself a “virtual channel busy” and this period is called “NAV (Network Allocation Vector)”. D cannot hear RTS but can hear CTS, so it also asserts the NAV.

A collision can occur when more than two stations send RTS simultaneously. Then they backoff and retry.

For the above two methods, there is no central control. All station contends each other as in Ethernets.

3) PCF (optional)—Point Coordination Function

The access point (base station) polls stations according to a list asking them if they have any frame to send. There will be no collision since the list is controlled by the base station.

DCF and PCF can coexist. It can be done by carefully controlling the waiting times such as DIFS, SIFS, etc. For more details, read the IEEE 802.11 standard. IEEE standards can be obtained from http://standards.ieee.org/getieee802/portfolio.html

8.2 Wireless LAN applications

· LAN Extension: Wireless LAN can save the cost of wiring in an existing building. Wireless LAN can easily interconnect existing LANs together, of course without wires.

· Cross-building interconnect: LANs between buildings can easily be connected without cross-street wiring.

· Normadic access: A laptop with wireless network card can connect to an existing wired LAN if an access point is attached to the wired LAN

· Ad Hoc networking: Computers equipped with wireless network cards can communicate each other if they are in the range of each others’ range of radio signals. Therefore, a LAN can be set up instantaneously.

There are some other aspects to consider about wireless LANs:

· An access point normally supports up to 10 stations even though there are access points that can support up to 100. Obviously, using more computers than recommended will cause performance and reliability problems.

· Multiple access points to a wired LAN: Multiple access points can be connected to a wired LAN if an area is too large.

· Extension point: This is a wireless relay (repeater) which extends the range of a single access point. Multiple extension points can be strung together to cover a wider area.

· Roaming.
A user can move from Area 1 to Area 2 transparently. The Wireless networking hardware automatically swaps to the Access Point with the best signal.