The Wireless LAN is a relatively new technology that uses radio frequencies as a medium instead of physical wires. WLANs use the same principles as an Ethernet network, with a couple of minor changes. Ethernet uses Carrier Sense Multiple Access with Collision Detection (CSMA/CD). This is not a possibility, as there is no way to accurately detect a collision on a Radio Frequency. Instead, WLAN technology uses Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) to ensure that packets are not dropped due to a collision. A WLAN device, or Mobile Unit (MU), uses two methods of collision avoidance: Physical Carrier Sense and Virtual Carrier Sense.
Physical Carrier Sense is quite simple. The MU checks to see if another station is transmitting at all times. This serves two purposes. The first is to determine if another station is trying to communicate with it. If the listening device senses that the medium is busy, it will attempt to synchronize with that transmission, and receive the transmission for processing by the network adapter if it is the intended recipient. The station also listens to the physical medium to determine if the radio frequency is free prior to transmitting data. This is referred to as Clear Channel Assessment (CCA). The station has to assess the channel to determine whether the ambient radio frequency energy is high enough to indicate that another station is transmitting. The 802.11 standard for this threshold is -82dBm. The physical medium must be clear before a station will begin to transmit.
Virtual Carrier Sense is a bit more complex. This 802.11 Collision Avoidance mechanism is based on a Virtual Carrier Detect (VCD) mechanism that uses a Network Allocation Vector (NAV) countdown timer to prevent a station from transmitting during a period for which it has accumulated advance knowledge that the medium will be busy. The network device may not be able to directly sense that the medium is busy, but it will not transmit until the NAV countdown reaches zero. Since the station is listening to the medium constantly, it hears every transmission that is within range. This process allows the station to predict when the medium might be free. When a station hears a frame it will look at the header of the frame and extract the Duration ID field. This field contains a value between 0 and 32,767. the Duration ID value is above zero the station will set its NAV timer to that value. It will then count down from that value and will not attempt to transmit anything until the NAV reaches zero. Once the NAV is at zero the station will not attempt to transmit right away – instead it must first contend for the medium. When the NAV is at zero the station will use CCA to ensure that the radio waves are clear, and then will wait for a pre-defined Interframe Spacing Gap (IFS Gap). There are four different Interframe Spacing Gaps in the 802.11 standard. These are Short Interframe Space (SIFS), which is 16 μs, Point Coordination Function Interframe Space (PIFS) which is not widely used, Distributed Coordination Function Interframe Space (DIFS) having a duration of 34 μs, and Extended Interframe Space which has a variable length. As we can see, these are all very short delays, but they must be taken into account when calculating delay on our WLAN. The IFS Gap can not be used to define QoS; instead there is a standard 802.11e that includes eight different Interframe spaces that are used to define QoS. After waiting for the specified IFS Gap, the station will begin a process called Request To Send (RTS)/Clear To Send (CTS). The station sends a short RTS frame to the access point (AP) which allows all access points within range to set a NAT timer. The receiving station transmits a CTS frame, letting all network stations hear the CTS frame. The transmitting station then is able to send its data frame, knowing that it has the all-clear signal from the AP to send data. Once the data is sent the receiving station acknowledges it with an ACK frame. In an environment that supports multiple 802.11 specifications (i.e. 802.11b/g) the stations will transmit the RTS and CTS signals at a lower speed, and will then transmit data at the higher speed. This allows all stations to set NAV timers appropriately.
In addition to the original 802.11 standard there are four additional standards that define operating parameters for a WLAN. These standards are 802.11a, 802.11b, 802.11g, and 802.11N. I would like to briefly discuss these four standards before I go any further.
The 802.11a Standard defines the operating range of the WAN as residing in the 5 GHz Frequency Range. 802.11a offers up to 54Mbps of raw throughput, and is able to realize an actual throughput of approximately 20 Mbps (after overhead). There are twelve channels that can be used in 802.11a, each operating independently of others. 802.11a offers an advantage over other standards that operate in the 2.4GHz range, as there are an incredible number of devices that operate in the 2.4 GHz range (microwave ovens, cordless telephones, etc.). By operating in the 5 GHz range 802.11a reduces the amount of interference that can be expected. There is a downside, however – 802.11a signals are more readily absorbed by things that stand in their way, such as walls.
802.11b is a standard hat operates on the 2.4 GHz range at speeds of up to 11 Mbps. 802.11b has a range of up to 38m and uses a technology called Complementary Code Keying (CC) to chip information bits. The process of chipping involves sending a continuous stream of pseudo-noise symbols which is a seemingly random sequence of 1 and -1 values. This process begins when the carrier signal is modulated with a code sequence. The code sequence contains the chipping code which determines how much spreading (how much of the frequency range) is going to be used, as well as information regarding the rate of transmission. There are 14 802.11b signals, and each one operates in a 22 MHz-wide range. All channels are spaced 5 MHz apart. This means that if two channels are in use in the same location they must be at least five channels apart (i.e. 25 MHz apart) to avoid interference from other channels. In North America the channels that are used are 1, 6, and 11. Europe licenses channels 1-13, however only 1, 5, 9, and 13 are commonly used. Japan licenses all 14 channels in the 802.11 spectrum.
802.11g is the most common standard currently in use. It is backwards compatible with 802.11b and also operates in the 2.4 GHz range. 802.11g operates at speeds of up to 54 Mbps. This provides a normal throughput of up to 19 Mbps, very similar to the speeds of 802.11a. The reason for the slight decrease is the additional overhead required for backwards compatibility with 802.11b. Again, there are 14 channels available, each operating in a 22 MHz range and spaced 5 MHz apart. This results in a logical limit of three channels within range of each other, like 802.11b. It uses the same modulation scheme as 802.11a, but includes the functionality of the 802.11b chipping scheme.
Finally, 802.11n is the most recent standard. It is not yet finalized by the IEEE, however there are many devices that already support the draft of the standard. The 802.n standard is expected to be finalized in late 2009, possibly as early as September, but a more realistic expectation would be that it is finalized in November. 802.11n is able to achieve much more acceptable levels of speed through Multiple-In Multiple-Out (MIMO) technology and operating in a 40 MHz range, called Channel Bonding. Using MIMO and Channel Bonding, 802.11n is able to achieve maximum speeds of up to 600 MHz (Current devices have been operating at up to 450 MHz). This yields an actual throughput of somewhere between 110 Mbps and 130 Mbps. Once it is finalized, the 802.11n standard, or Wireless-N as it is often referred to, will allow for an optimal configuration of wireless devices using VoIP. While I won’t get into the technical details of how 802.11n works, I will say that it uses multiple antennas to process multiple data streams simultaneously, increasing the raw throughput of devices on a wireless-N network. Further information on 802.11n can be found in a number of places, including Wikipedia.
Physical Carrier Sense is quite simple. The MU checks to see if another station is transmitting at all times. This serves two purposes. The first is to determine if another station is trying to communicate with it. If the listening device senses that the medium is busy, it will attempt to synchronize with that transmission, and receive the transmission for processing by the network adapter if it is the intended recipient. The station also listens to the physical medium to determine if the radio frequency is free prior to transmitting data. This is referred to as Clear Channel Assessment (CCA). The station has to assess the channel to determine whether the ambient radio frequency energy is high enough to indicate that another station is transmitting. The 802.11 standard for this threshold is -82dBm. The physical medium must be clear before a station will begin to transmit.
Virtual Carrier Sense is a bit more complex. This 802.11 Collision Avoidance mechanism is based on a Virtual Carrier Detect (VCD) mechanism that uses a Network Allocation Vector (NAV) countdown timer to prevent a station from transmitting during a period for which it has accumulated advance knowledge that the medium will be busy. The network device may not be able to directly sense that the medium is busy, but it will not transmit until the NAV countdown reaches zero. Since the station is listening to the medium constantly, it hears every transmission that is within range. This process allows the station to predict when the medium might be free. When a station hears a frame it will look at the header of the frame and extract the Duration ID field. This field contains a value between 0 and 32,767. the Duration ID value is above zero the station will set its NAV timer to that value. It will then count down from that value and will not attempt to transmit anything until the NAV reaches zero. Once the NAV is at zero the station will not attempt to transmit right away – instead it must first contend for the medium. When the NAV is at zero the station will use CCA to ensure that the radio waves are clear, and then will wait for a pre-defined Interframe Spacing Gap (IFS Gap). There are four different Interframe Spacing Gaps in the 802.11 standard. These are Short Interframe Space (SIFS), which is 16 μs, Point Coordination Function Interframe Space (PIFS) which is not widely used, Distributed Coordination Function Interframe Space (DIFS) having a duration of 34 μs, and Extended Interframe Space which has a variable length. As we can see, these are all very short delays, but they must be taken into account when calculating delay on our WLAN. The IFS Gap can not be used to define QoS; instead there is a standard 802.11e that includes eight different Interframe spaces that are used to define QoS. After waiting for the specified IFS Gap, the station will begin a process called Request To Send (RTS)/Clear To Send (CTS). The station sends a short RTS frame to the access point (AP) which allows all access points within range to set a NAT timer. The receiving station transmits a CTS frame, letting all network stations hear the CTS frame. The transmitting station then is able to send its data frame, knowing that it has the all-clear signal from the AP to send data. Once the data is sent the receiving station acknowledges it with an ACK frame. In an environment that supports multiple 802.11 specifications (i.e. 802.11b/g) the stations will transmit the RTS and CTS signals at a lower speed, and will then transmit data at the higher speed. This allows all stations to set NAV timers appropriately.
In addition to the original 802.11 standard there are four additional standards that define operating parameters for a WLAN. These standards are 802.11a, 802.11b, 802.11g, and 802.11N. I would like to briefly discuss these four standards before I go any further.
The 802.11a Standard defines the operating range of the WAN as residing in the 5 GHz Frequency Range. 802.11a offers up to 54Mbps of raw throughput, and is able to realize an actual throughput of approximately 20 Mbps (after overhead). There are twelve channels that can be used in 802.11a, each operating independently of others. 802.11a offers an advantage over other standards that operate in the 2.4GHz range, as there are an incredible number of devices that operate in the 2.4 GHz range (microwave ovens, cordless telephones, etc.). By operating in the 5 GHz range 802.11a reduces the amount of interference that can be expected. There is a downside, however – 802.11a signals are more readily absorbed by things that stand in their way, such as walls.
802.11b is a standard hat operates on the 2.4 GHz range at speeds of up to 11 Mbps. 802.11b has a range of up to 38m and uses a technology called Complementary Code Keying (CC) to chip information bits. The process of chipping involves sending a continuous stream of pseudo-noise symbols which is a seemingly random sequence of 1 and -1 values. This process begins when the carrier signal is modulated with a code sequence. The code sequence contains the chipping code which determines how much spreading (how much of the frequency range) is going to be used, as well as information regarding the rate of transmission. There are 14 802.11b signals, and each one operates in a 22 MHz-wide range. All channels are spaced 5 MHz apart. This means that if two channels are in use in the same location they must be at least five channels apart (i.e. 25 MHz apart) to avoid interference from other channels. In North America the channels that are used are 1, 6, and 11. Europe licenses channels 1-13, however only 1, 5, 9, and 13 are commonly used. Japan licenses all 14 channels in the 802.11 spectrum.
802.11g is the most common standard currently in use. It is backwards compatible with 802.11b and also operates in the 2.4 GHz range. 802.11g operates at speeds of up to 54 Mbps. This provides a normal throughput of up to 19 Mbps, very similar to the speeds of 802.11a. The reason for the slight decrease is the additional overhead required for backwards compatibility with 802.11b. Again, there are 14 channels available, each operating in a 22 MHz range and spaced 5 MHz apart. This results in a logical limit of three channels within range of each other, like 802.11b. It uses the same modulation scheme as 802.11a, but includes the functionality of the 802.11b chipping scheme.
Finally, 802.11n is the most recent standard. It is not yet finalized by the IEEE, however there are many devices that already support the draft of the standard. The 802.n standard is expected to be finalized in late 2009, possibly as early as September, but a more realistic expectation would be that it is finalized in November. 802.11n is able to achieve much more acceptable levels of speed through Multiple-In Multiple-Out (MIMO) technology and operating in a 40 MHz range, called Channel Bonding. Using MIMO and Channel Bonding, 802.11n is able to achieve maximum speeds of up to 600 MHz (Current devices have been operating at up to 450 MHz). This yields an actual throughput of somewhere between 110 Mbps and 130 Mbps. Once it is finalized, the 802.11n standard, or Wireless-N as it is often referred to, will allow for an optimal configuration of wireless devices using VoIP. While I won’t get into the technical details of how 802.11n works, I will say that it uses multiple antennas to process multiple data streams simultaneously, increasing the raw throughput of devices on a wireless-N network. Further information on 802.11n can be found in a number of places, including Wikipedia.
No comments:
Post a Comment