Before you start
Objectives: Learn the basics about frequencies, architecture, and media access method used in wireless networks.
Prerequisites: you should know the types of network topologies used.
Key terms: wireless, network, device, BSS, data, frequency, AP, topology, method, access
Radio Frequency
Radio Frequency waves replace physical wires as the medium which we use to transfer data on our network. Radio waves use specific frequency ranges that are specified by the Federal Communications Commission (FCC) agency in the United States of America. Those ranges are classified as “Industrial, Scientific and Medical (ISM) Radio Frequency Bands”. The reserved frequency ranges that can be used for wireless networks are:
- 2.4 GHz
- 5.8 GHz
- 5.15 GHz
- 5.25 GHz
- 5.725 GHz
The popular range is the 2.4 GHz. In addition to wireless networks, there are some other devices that also use that range. For example, cordless phones and microwave ovens. This fact can cause problems in a form of interference.
Dealing With Interference With Spread Spectrum
Wireless standards are defined in the IEEE 802.11 standards. These standards specify ways to increase bandwidth and to reduce interference when transferring data using radio waves. One way to do all that is to use Spread Spectrum (SS) signal transmission technique. In SS the radio transmitter varies the frequency used during the transmission. In other words, data is spread across a wide band of frequencies. So, the data is sent by using more than one frequency. In order for this to work, the transmitter and the receiver have to be in tune. The Spread Spectrum technology provides better bandwidth in contrast to using a single frequency to send data. It also deals with errors caused by interference from other devices. If we have interference on one frequency, with Spread Spectrum we can compensate that by sending data on another frequency.
Spread Spectrum can be implemented in two different ways. The first method is the Frequency Hopping Spread Spectrum (FHSS). In FHSS the transmitter only uses one frequency at the time, but it changes that frequency several times each second, in a predictable sequence. The transmitter and the receiver have to be in sync in order for this to work correctly. FHSS method also increases security since it makes data capture more difficult. The second method is called the Direct Sequence Spread Spectrum (DSSS). With DSSS the transmitter divides the information to multiple pieces, and then sends those pieces on different frequencies, all at once. This really increases the bandwidth, but this method is more susceptible to interference and less secure then FHSS.
Architecture and Topology
Every wireless network has some typical components. The device with the wireless network card (interface) installed is often referred to as a station. Also, we could also refer to that device as a host, but if it has for example, two wireless interfaces, then we would have two stations. So, what makes a station is the wireless interface, not the device itself.
In general, wireless network can operate in two different modes. In first mode we have two wireless devices that communicate directly to each other. This is called an Ad hoc wireless network. Ad hoc network works in peer-to-peer mode and it uses physical mesh topology with logical bus topology. The mesh topology means that all devices have to communicate directly to each other. As we add devices to the network, each device has to keep track of every other device on the network in order to communicate. The ad hoc wireless networks are typically used to create temporary connections between two hosts.
Mesh Topology
The second mode in which a wireless network can operate is an Infrastructure network. This type of wireless network uses a wireless Access Point (AP). The AP in wireless networks acts as a hub in Ethernet networks. Infrastructure network uses a physical star topology and a logical bus topology. The AP is in the middle and communicates with all stations on the network. Stations that need to send data to other stations first send data to the AP, and the AP then sends that message to the end device. With the infrastructure network we can easily add new hosts and we can easily connect the wireless network with the wired network.
Star Topology
The smallest networking unit in a wireless network is called a Basic Service Set (BSS). BSS is a collection of all devices that can communicate together using the same channel. Channel is a portion of a wireless frequency that all devices use. If we have an Ad hoc network, the BSS is the collection of devices that communicate together. If we have an Infrastructure network, the BSS consists of all devices and the AP.
In larger environments we can have multiple AP on the same network. On each AP there are different clients connected to that portion of the wireless network. Note that all clients and an AP form a separate BSS. Multiple BSSs will typically use different channels. Wireless APs can be connected together, and be connected to a wired network. If we connect them to a wired network, APs are acting as bridges connecting the wireless clients to the wired network. The form in which we have multiple BSS units connected together is called the Extended Service Set (ESS).
BSS and ESS
The ESS is identified by a name, which is often referred to as Service Set Identifier (SSID). Wireless clients that want to connect to the ESS are configured with the SSID, as are all the APs in the ESS. This configuration allows a wireless host in one BSS to move to another BSS. The wireless client uses the appropriate channel to connect to the BSS. The client identifies the BSS by using the APs MAC address. This MAC address is called the BSSID (Basic Service Set Identifier). This is typically configured automatically on the Access Point. When the wireless client tries to connect to the specific SSID, it finds the broadcasting AP within its location and automatically chooses the appropriate channel to use in order to join the BSS. When the wireless client moves outside of the range of the current BSS, it still uses the same SSID to locate additional APs that are the part of the same ESS. When it comes in the range of the next BSS, it will typically use the different channel to join the BSS. So, the client will change channel as it moves from one BSS to another. Again, it will use the BSSID or the MAC address of the AP to keep track of which BSS it is currently in.
Many wireless APs include bridging features, and include switching and routing features which enable us to easily connect the wireless network to the wired portion of our network.
Media Access Control Method
Wireless networks use the Carrier Sense Multiple Access/Collision Avoidance (CSMA/CA) media access method. This is similar to media access method used by Ethernet, where only the last part is different (Collision Detection instead of Avoidance). The whole process starts the same, but ends differently. The first that happens when a device needs to send data on the network is the Carrier Sense. This means that the device listens if anybody else is currently transmitting. If somebody is using the media, the device will wait a random period of time (a backoff period) and then try again. This part of the access method is the same in Ethernet and in wireless. In wireless networks, if the media appears to be free, the transmitting device sends the Request-to-Send (RTS) message to the receiving device or AP. The RTS includes the information about the source, destination and the duration of the requested transmission. The RTS message will also be received by any other device in the area. This way all devices will know that there is a device on the network which will be transmitting the next period of time, so they will delay attempting to send. When the receiving device receives the RTS, it transmits back a Clear-to-Send (CTS) message, which also includes information about the duration of the transmission. The CTS message means that the device is ready to receive data. After the RTS and CTS messages, the transmitting device sends actual data to the receiving device, and then waits for the acknowledgement of data receival. If the acknowledgement isn’t received, then the sending device assumes that there was a collision on the network and it retransmits the original data. After the time in the RTS and CTS has expired, the other devices can go trough the same process to see if they can transmit data. So, wireless devices have to take turns when transmitting since they all share the same media. This means that the wireless communication operates in half-duplex mode. This means that devices can both send and receive, but not at the same time. Note that with typical wireless infrastructure implementation we use the physical star topology, but we use the logical bus topology. Messages in a logical bus topology are received by all devices on the network.
Considerations
There are some common factors that will more or less affect our wireless network performance. The transfer speed on our Wi-Fi network depends on the type of equipment that we are using. There are several Wi-Fi specifications, like 802.11g, 802.11n, 802.11ac, etc. For example, with 802.11g the theoretical maximum is 54 Mbps, but the actual speed will probably be a lot less. The reason for that is that our Wi-Fi NIC and WAP will negotiate the optimal speed depending on environmental factors like distance and obstacles between them. If the signal is bad, the speed can even drop to as low as 2 Mbps. If the signal drops even more, the connection will usually be dropped. Rarely, the speed will go down to 1 Mbps. So, the first thing to keep in mind is that wireless networks have limited range.
One other factor that will affect our Wi-Fi performance is Electro Magnetic Interference (EMI). EMI is actually any radio wave coming from the environment. If EMI is strong enough, it can degrade the performance of our Wi-Fi. 802.11b,g,n networks all operate in 2.4 GHz range. If we have other devices that also work in that range (microwave, cordless phone, other APs on the same or similar channel), it can really impact our Wi-Fi. It probably won’t drop our connection, but the data rate will be lower. If there are other APs in the area, we can try to avoid interference by changing the channel used on our access point.
One other factor that can impact Wi-Fi performance is the antenna. The better (quality) the antenna, the better the range. Also, the antenna orientation might have a small effect on signal strength. There are two types of antennas: Directional antenna and omni-directional antenna. Directional antenna creates a narrow, focused signal in a particular direction. Focused signal provides greater signal strength increasing the transmission distance. It can also provide a stronger point-to-point connection, better equipping them to handle obstacles. Omni-directional antenna disperses the RF wave in an equal 360-degree pattern. It is used to provide access to many clients in a radius.
The placement of the AP and hosts can also impact performance. We should place APs in central locations since radio waves are broadcast in each direction. Also, devices often get better reception from APs that are above or below. For security reasons we should not place APs near outside walls since the signal will extend outside the walls. Tick concrete and metal can have a great impact on Wi-Fi signal. In that case we will probably have to install multiple APs which are bridged together to accommodate signal loss. In this scenario we will have overlapping wireless networks, so we should use different channels to ensure that they don’t conflict with each other.