transmission wave WIFI Wireless Network Architecture I. RADIO TRANSMISSION
A. GENERAL
1. Introduction:
Wireless technologies, as well as wired connections, are grouped into three categories depending on the scope of these links:
them WWAN (Wireless Wide Area Network): GSM, GPRS, UMTS-WLAN
( Wireless Local Area Network) WLAN (RadioLan): IEEE 802.11, HIPERLAN
-WPANs (Wireless Personal Area Network): Bluetooth, HomeRF
The WPAN covers few meters, the WLAN is measured in tens and hundreds of meters, the WWAN hundreds and thousands of meters.
2. Frequency bands:
Two groups are represented:
-technologies for mobile phones (from 824 to 2170 MHz)
-technologies used for computing, for WPAN and WLAN , operate on two bands: the ISM (Industrial Scientific Medical) (from 2400 to 2500 MHz) band and the U-NII (Unlicensed-National Information Infrastructure) (5150 MHz à5720).
ISM Band:
The ISM band is three sub-bands (902-928 MHz, 2.400-2.4835 GHz and 5725-5850 GHz) single band 2.400-2.4835 GHz with a bandwidth of 83.5 MHz is used by 802.11.
ISM bandwidth (the maximum is 83 .5 MHz) varies from country to country, as well as usable power. Also this band, specifically the sub-band 2.400-2.4835 GHz is heavily used by different standards and disrupted by devices (microwave ovens, wireless keyboard and mouse ...) operating in these frequencies.
band U-NII:
The band U-NII (5 .15-5.35 GHz 5725-5825 GHz) provides a bandwidth Total of 300 MHz, each using a different signal strength.
3. Organizations
a) Regulatory agencies:
These bands are recognized by international regulatory bodies for unlicensed use. These agencies are:
FCC: Federal Communications Commission for the United States, ETSI: European Telecommunications Standards Institute for Europe, MKK: for Japan, ART: Telecommunications Regulatory Authority for France.
b) French Regulations:
Since July 25, 2003, for France, the ART redefines the use ISM and U-NII, according to a release from the ISM band and part of the U-NII band by the Ministry of Defence and also for setting compliance on the new EU directives (known as "Telecoms Package").
It should be noted that the ISM band is said to be free is to say that it is free for now of any charge whether for private or public. Furthermore there is no authorization for the use of ISM band outside, only a declaration is required for public use.
These provisions are temporary and will serve as experimental setting pending the establishment of a law.
tables summarizing the provisions of ART
Tehnique Terms of Use frequency: Capacities are expressed in EIRP = equivalent isotropic radiated power, radiated power emission point average output of antenna.
* DFS: Dynamic Frequency Solution; TPC: Transmit Power Control.
c) Standards bodies:
Two organizations involved in standardization of wireless WLAN
ETSI
In Europe, the group Hiperlan (High Performance Radio LAN) from the ETSI (European Telecomunications Standards Institute) defines two standards, HIPERLAN 1 with a flow rate of 10 and 20 Mbps and HiperLAN 2 with a flow rate of 54 Mbit / s.
IEEE:
the United States is the committee 802 (so called by its creation date: February 1980) from the IEEE (Institute of Electrical and Electronics Engineers) standard that defines the IEEE 802.11 and its extensions (802.11b, 802.11a, ...).
These two standards are incompatible, more Hiperlan solely the band U-NII bands, while 802.11 uses the ISM and U-NII. Currently, only, products from the 802.11 standard are marketed.
WECA:
The term Wi-Fi (Wireless-Fidelity) is a standard issued by the WECA (Wireless Ethernet Compatibility Alliance) with 802.11b products. The latter, composed of 140 companies, tests and manages the interoperability between equipment to the standard 802.11.b. Lately the term
WiFi certified 802.11a 5.
B. THE IEEE 802.11 STANDARD
The 802.11 standard, like all the standards set by the Committee 802, covers the first two layers of the OSI model, that is to say, the physical layer (level 1) and datalink layer (level 2).
1. LAYER 1 (802.11 PHY)
General:
The physical layer defines the transmission technique (modulation of radio waves), encoding and signal transmission. The wireless signal will carry electrical information, it will be modified to carry information (binary data here). Sinusoidal electric signals can vary its amplitude (voltage in volts), frequency (in hertz) and phase (degrees). So on one of these three parameters that can change an electrical signal to encode. Usually associated frequency modulation and phase modulation to increase performance.
The physical layer is divided into two layers. PLCP (Physical Layer Convergence Protocol) is responsible for listening to the media and signaling by providing a CCA (Clear Channel Assessment) to the MAC layer and PMD (Physical Medium Dependent) processes the data encoding and modulation.
802.11 (LEV1) offers three layers using three different transmission techniques (FHSS, DSSS, IR). New techniques have since been added: 802.11b (DSSS / CCK), 802.11a (OFDM), 802.11g (OFDM) for key.
IMPORTANT NOTE: The rates shown do not represent useful rates but the actual flow rates needed for a reliable radio transmission.
802.11: The standard physical
802.11 (ratified in 1997) offers two types of transmission associated with frequency modulation and phase modulation technique used mainly infrared transmission in industrial environments and very much about computers. We saw only two types of transmission frequency modulation using more precise technique to "spread spectrum". This technique, developed by the military, has grown considerably because it has good performance against jamming and allows more easily coexist in the same frequency band several transmissions.
FHSS (Frequency Hopping Spread Spectrum) / GFSK (Gaussian Frequency Shift Keying) spread spectrum or frequency hopping. It changes the carrier frequency by a sequence of jumps. That is to say that the issuer changes its transmission frequency on a periodic basis and following a set sequence, it synchronizes with the receiver frames tags that contain the hop sequence and duration. In the 802.11 ISM frequency band defined from 2.400 to 2.4835 GHz is divided into 79 channels of 1 MHz and the jump occurs every 300 to 400 ms. The transmitter and receiver agree on a hopping sequence. The standard defines three sets of 26 possible sequences (78 sequences total). Signals (FHSS transformed data) are then modulated by a phase modulation type GFSK. The flow rates reach 1-2 Mbps Initially this technique was used for military purposes to encrypt the transmission, but the sequences of frequencies are now standardized, thus disclosed, the 802.11 standard is used to correct the interference phenomenon. Over the Bluetooth uses this technique but with different hopping sequences.
DSSS (Direct Sequence Spread Spectrum) or spread spectrum direct sequence. As for the MSDS, the DSSS is a technique called spread spectrum operating on the 2.4 GHz ISM band. This time the band is divided into 14 channels of 20 MHz, each 20 MHz channel consisting of four units of 5 MHz. Each channel is spaced 5 MHz, except channel 14, spaced 12 MHz channel 13.
bandwidth of 83.5 MHz is, can not be placed end to end 14-channel 20 MHz without having them overlap. When a channel is selected, the signal spectrum occupies a bandwidth of 10 MHz on either side of the peak frequency, that is why we can only use three separate channels (ie three networks) transmitting on the same cell without risk of interference.
Examples of combination of three channels:
It is essential to assign to each point of access channels that do not overlap. The major drawback is that the DSSS generates losses due to this overlap. A technique called "chipping" solves the information loss. This technique is to encode each bit in a sequence of 11 bits (called sequence Baker): 10110111000 when the bit is 1 and its complement where 01001000111 is 0. This allows for error control. This sequence or signal, called a "symbol", is transmitted at a rate of 1 MS / s (million symbols per second). The final rate in bit / s will be determined according to the phase modulation applied:
- BPSK (Binary Phase Shift Keying): This type of modulation will encode one bit at each phase change? rate of 1Mbit / s.
- QPSK (Quadrature Phase Shift Keying) will encode two bits per phase change? 2 Mbit / s. Using Channel
following countries:
802.11b (Wi-Fi)
The IEEE committee in 1999 identified a new physical layer, or 802.11HR 802.11b (High Rate), allowing for speeds of 5.5 to 11 Mbps s. This new physical layer, called Wi-Fi by WECA, is implemented on 802.11. This standard still uses the ISM band and DSSS modulation, making it fully compatible with 802.11 DSSS coding by cons is more sequence-based Baker, but encoding CCK (Complementary Code Keying). It uses a modulation scheme QPSK phase but at a rate of 1.375 MS / s, which allows it to reach speeds of 11 Mbps In addition, a mechanism for environmental adaptation can automatically adjust the speed (Variable Rate Shiting) according to reception conditions (interference, range of equipment ...).
Rate / Range:
type encoding and phase modulation:
802.11a:
In parallel to the previous standard in 1999, the IEEE has finalized a new layer Physical: 802.11a. Referred to as Wi-Fi by WECA 5, the physical layer uses the radio band U-NII 5GHz, which offers a higher bandwidth (300MHz) which is much less crowded than the ISM band. By cons, it is totally incompatible with other physical standards. In addition, the modulation frequency used, OFDM (Orthogonal Frequency Division Multiplexing) is different from other physical standards. It was found that over the long frames are more overlap, due to interference, inter frame is smaller. This shows that several low-flow channels are more effective than a single broadband.
OFDM: the first two subbands (Low and Middle) of the U-NII band is divided into 8 channels of 20 MHz. Each channel is then divided into 52 sub-channels of 300 MHz, 48 for data and 4 for error correction called FEC (Forward Error Correction).
8 CHANNEL
This parallel transmission of several sub-channels at low rate that will create, in fact, one broadband channel. In addition we can use eight channels disjoint, without interference, allowing eight Wi-Fi 5 to transmit simultaneously, while Wi-Fi only supports three. Cons by the disadvantage of OFDM is that it requires more power than the spread spectrum technology, which quickly empty the batteries of mobile devices. 802.11a provides data rates from 6 to 54 Mbps depending on the modulation phase used:
- BPSK achieves a throughput of 6Mbits / s
- 64QAM (64-level Quadrature Amplitude Modulation) allows a rate of 54 Mbit / s.
As for Wi-Fi, Wi-Fi 5 uses the "Variable Rate Shifting" when the environment deteriorates. Throughput from 54 Mbps to 48 then 36, 24, 12 and 6 Mbps to finish. It should be noted that the range is below the standards using the ISM band, because the higher the frequency, the greater the range decreases.
802.11g (validated in June 2003) Last
physical layer in 802.11 and made available since this year but not yet ratified by WECA. : 802.11g. This standard uses the ISM band as Wi-Fi as well as the CCK coding technique, which makes it compatible with Wi-Fi For cons as it uses OFDM transmission technique, which allows it to reach a max speed of 54Mbits / s but with a higher energy consumption. Products using the 802.11g sold on the market should offer full compatibility with 802.11b.
physical phenomena (altered signal / Solutions)
The phenomenon of propagation is a key element in the radio transmission. Depending on the type of environment, characteristics emission and reception will be profoundly changed. In direct propagation, the received signal will decrease linearly, while in a confined environment, the received signal undergoes changes during its route. It can be mitigated, diffracted, and also reflected. Generally we consider that there are four paths created when a signal undergoes an alteration: the direct path, the path with reflection on the ground, and two paths with reflection on the wall. Two additional elements modifying the signal during transmission is taken into consideration: the displacement of people, physical topology change (opening of doors ...) and use of equipment such as fans, microwave oven ...
Table of transmission over Wi-Fi:
It is these conditions propagation, which will determine all terms of use (transmission speed and flow). We will see now in terms of these weathering processes, techniques that allow a better radio electric signal carrying information.
Fading:
The radio signal as it travels, experiences, as we have seen previously, alterations and in particular impoverishment called "fading" in English.
Against this phenomenon using a technique called "diversity". This process involves collecting several transmissions of the same message. Several types of diversity exist:
-spatial diversity (or diversity antenna) is the most used:
The receiver has multiple antennas (minimum two). For information, the length between the antennas must be a multiple of the half-wavelength of the carrier frequency. Upon receipt of a frame can choose the best reception received by its antenna, it can also add or combine the signals, which improves very substantially the result. The
-frequency diversity, is to send a frame at different frequencies and choose the best, this requires having a broad frequency spectrum. The operation of
-time diversity imposes a waiting time between two frames (about 50 ms) which weakens the network performance.
Multiple Routes:
When sending a frame, the receiver receives this frame into several copies according to the different possible paths taken by the frame. The time of receipt is sent to his superior as the original frame and the echoes overlap products. It calculates a delta (SD propagation delay) if it exceeds 10% the length of the frame must be put in place techniques struggles against such interference. Different techniques:
-EQ: It makes a correction of digital transmission, this correction is calibrated by sending a frame of learning known to the receiver. The receiver samples the learning signal taken from a delay line which allows the equalizer to adjust its digital processing. This treatment is then applied to all frames.
-Spread Spectrum: This technique is very different from the previous is highly developed in the 802.11 standard since it is used in frequency bands where other transmissions coexist. The main families described above.
2. LAYER 2 (802.11 MAC)
· General:
At level 2, the data link layer is divided into two sublayers: LLC and MAC. The LLC sublayer, defined by the 802.11 standard is identical to the 802.2 layer allowing compatibility with any other network 802, while the MAC sublayer is reset by the standard 802.11 (LEV2). It characterizes the way of media access common to different physical standards 802.11, it is equivalent to the 802.3 Ethernet standard with features necessary for radio transmissions (the error rate is higher than the wire medium) that are normally assigned to higher protocols, such as fragmentation, control of error (CRC), packet retransmissions and acknowledgments. In addition, the MAC layer defines two different access methods, the Distributed Coordination Function (DCF) or CP (Contention Period), also called user access to competition, and the Point Coordination Function (PCF) or CFP (Contention Free Period ) called user access controlled. The DCF method is similar to Ethernet for transporting asynchronous data where stations have an equal opportunity to access support. The second method is the CPF, based on questioning in turn stations or polling controlled by the access point. A station can transmit only if it is allowed and it can only receive if selected. This method is designed for real-time applications (video, voice) requiring management time during data transmissions. The DCF method is used by the architectural forms Ad-Hoc and infrastructure, while the method PCF is used by the infrastructure mode.
History:
Historically there are two large families, controlled-access protocols and the protocols to competition: the first ran on time division multiplexing. Each host has a part of the communication available, there is reservation of bandwidth. Many disadvantages are inherent in this technology: a closed network, trouble management, inefficient, limited number of machines. The Aloha protocol, the first protocol to competition is based on a shared access with the support of collision risk between stations. Unable to prevent collisions, it uses a higher layer protocol in connected mode to allow the retransmission of the frame. More to prevent the collision from happening again, the stations have caused the collision, retransmit after a random time. This protocol enabled the rapid development of local networks. But as it does not achieve satisfactory performance on networks of larger size, it was improved by adding a carrier sense. A machine can listen on the bus to see if there is a current issue, the number of collisions is greatly reduced. This is the CSMA (Carrier Sense Multiple Access).
In CSMA / CD CSMA / CA:
machines using CSMA know if the line is busy, but if at the same time there are two machines emit collision. These collisions must be detected for the MAC layer can retransmit without passing through an upper layer. A system has been added to the protocol, allowing the machines to listen to the line while they emit is the collision detection (CD: Collision Detection). Protocol CSMA / CD the most used is called Ethernet. This system can be implanted in a radio for two reasons: the radio links used are not full-duplex (one can listen and transmit at the same time) and a machine that listens to the carrier is not sure to listen to all stations connected to the access point (If the station down). So we modified the CSMA / CD to arrive in CSMA / CA (Collision Avoidance) protocol called for collision avoidance.
DCF (CSMA / CA)
Protocol CSMA / CA uses several techniques to compensate for this inability to listen to broadcast. First a media access based on timers, a positive acknowledgment, management of recovery collision timers and a technical option for secure data transmission and avoid collisions with hidden nodes.
The Media Access:
Each frame is delimited by a space. This space allows management access to the support by delaying the sending of frames. By the type of timers used, it sets the priority of access. More IFS (Inter Frame Space) is short more access is a priority. There are three different types of inter frames:
-SIFS (Short IFS), the smallest of the IFS, so the highest priority. It is used for transmitting the same dialogue (data, ACK, ...) acknowledgment of the receiving station and data from the transmitting station remains a priority.
-PIFS (PCF IFS) inter frame space used for PCF (controlled access) by the access point. Allows priority access to the AP on the network stations. Its value corresponds to a plus one SIFS time (time slot).
-DIFS (DCF IFS) timer inter frame for distributed access used by stations to access the support (DCF mode).
Note: The fact that the PIFS interstitial frames are shorter than the DIFS interstitial frames show clearly that the data sent in the PCF mode has priority over the data sent in DCF mode.
VALUE SPACES (depending on the physical layer):
System acquittal positive
In an exchange between two stations, the station checks the destination CRC the frame and sends an ACK (acknowledgment) to the transmitter. If the transmitting station does not receive ACK, it assumes that a collision occurred, the frame is retransmitted after management using timers
Management pending transmission:
When a station wants
transmit data, it senses the medium. If the support is free for a DIFS, the station transmits, if it detects cons by a transmission, it uses a timer called NAV (Network Allocation Vector), allowing it to suspend its transmissions. NAV This applies to all stations and they have the ability to issue after the end of the NAV. The NAV is calculated on the TTL (Time To Live) of frames sent. This allows stations in the vicinity of the source and destination stations to know the full cycle of transmission to come. These different stations pending issue threatens to cause collisions if not using a management technique where the carrier will be free again. This process is called the restart backoff algorithm, each station calculates a random delay between 0 and 7 "time slot" (the time unit smaller, varying according to standard physical) and decrements this timer as soon as media is free. The station reached a value of 0 will transmit its first information, the other timer and block them again as soon as media is free again. If two stations have the same value of timer collision will occur. These stations will then regenerate a new meter, including this time between 0 and 15. This algorithm allows stations to access the media with the same probability, but not guaranteed period.
Technical secure transmission reservation (optional):
Listening to the media is at the physical layer with the PCS (Physical Carrier Sense) and at the MAC sublayer, with the VCS (Virtual Carrier Sense). The PCS detects the channel busy analyzing frames from the wireless carrier, the VCS is a reservation mechanism based on the use of frames RTS / CTS (Request To Send / Clear To Send) between source host and destination host . His Operation is simple: a station wishing to transmit sends an RTS, stations in the BSS read this RTS and initialize their NAV based on the parameters contained in the RTS. The destination station responds after a SIFS, with a CTS, again the other stations update their NAV based on the STC. The source station, having received the STC is satisfied that the support is reserved for its transmission. This method is optional and rather used to send large frames that would drop the bandwidth performance if there are collisions. We can precisely calculate a threshold for whether to use the RTS / CTS (RTS Threshold). A second application very useful for the universe of radio links, is the detection of hidden stations. Two stations in a. BSS may even be out of radio range from each other while being on the same PA. If they want to transmit at the same time there will be a collision. By cons if booking before transmission function RTS / CTS stations hidden from the source station will still detect the CTS from the access point. It should be noted that collisions can occur between RTS frames, but because of their small size, the bandwidth is not too affected
Protocol CSMA / CA is used to manage collisions while overcoming the constraints due to transmission radio. Cons by the mechanisms put in place weigh down trade (Specific WLAN frames) which makes the performance less than a wired network.
PCF (centralized access mode):
The 802.11 standard provides an opportunity for channel reservation to use real-time services. This system allows to implement controlled access type reservation. This control is made only by an access point (AP) which will, according to a TDM, organize a sequential polling stations (polling). In this mode they are more stations trying to access the media but the access point that controls the media. It selects the station may transmit. Mode PCF is optional in the 802.11 standard, it is still used in alternation with the DCF mode is always the primary mode of access. Moreover, a station can use both modes at once. It is in the process of association that the access point and station indicate if they implement this function. The activation is done for the access point (AP), through some fields beacon type frames, response association and the associated station, at the request frames of association and request verification. The AP compiled a list of interrogation (polling list) associated stations operating in PCF mode. The access point can manage stations operating in both modes. The PCF mode is organized around a "super frame" divided into two parts: one where the PCF mode is activated, it is the CFP (Contention Free Period), which corresponds to a period of time without restraints and another where it switches to conventional DCF (distributed access) is the CP (Contention Period), which corresponds to a period of time with restraint. The AP generates a tag, called Beacon Frame to indicate the change in PCF mode, after an inter frame PIFS. This tag type is DTIM (Delivery Traffic Information Map) then the station coordinator (PA) will examine each site by type frames CF.Poll (cotent Free Polling). The station interviewed may then transmit its data at the end of the PA takes his hand and asks the next station of the polling list. The PCF mode is a quality of service (QoS) for 802.11 but this function is not implemented in most facilities. A standard dealing with quality full service will appear with the 802.11e standard. Standards Sub
802.11 MAC
802.11e (QoS)
The PCF access method has never been used because no manufacturer has not implemented in products. IEEE 802.11e group aims to improve the QoS (Quality of Service) in both modes, DCF and PCF.en adding two new access methods, EDCF (Extended DCF) and HCF (Hybrid Coordination Function).
EDCF: Today Fi frames have the same priority regardless of the station. EDCF defines eight traffic categories (TC: Traffic Categories) are eight priority levels. Each station mode EDCF supports eight queues for each traffic type.
These eight traffic categories have their own parameters timers (IFS Backoff). Moreover the values of timers are no longer fixed. It keeps the ISF mode DCF plus a new timer, the AIFS (Arbritration IFS). This is the AIFS DIFS but its value is variable depending on the priority level of the transmitting station (AIFS = DIFS)
Another novelty of the ECDF: the TxOP (Transmission Opportunities). This mechanism of transmission management defines the right of access to a station and the time allocated according to its priority level. If several stations different classes of traffic accessing the medium at the same time, the TxOP, which is a time predetermined (category of the highest traffic? shortest time), give access to the highest priority category. This timer is added to the end of the backoff timer
HCF: This second method, as the PCF, the access point uses to manage traffic by defining periods with and without restraint (CP and CFP), d hence the term hybrid.
802.11f: see "Managing mobility (roaming)"
802.11h:
This standard aims to make it compatible with 802.11a equipment infrastructure Hiperlan2. Technology adoption DFS (Dynamic Frequency Solution) and TPC (Transmit Power Control) to comply with European standards. This allows the automatic assignment of frequencies and in control of the transmission power to avoid interference between access points.
802.11i: See "INTERNAL SOLUTIONS FUTURE"
802.11j:
Convergence of European and American Standard 802.11a Hiperlan.
