Now let's take a look the 802.11 PHY protocols. Of the three available 802.11 PHY protocols, 802.11a, b and g, 802.11b is the most widely deployed. This is largely because it is the least expensive. The 802.11a standard was defined first, but was more costly to implement, and the operational RF spectrum it utilized varied across the globe. Expense, limited distance and lack of backward compatibility and global compatibility greatly limited 802.11a's deployment to only specific environment with operational requirements for high-speed wireless. A good rule of thumb when comparing 802.11a to 802.11b is one-third the distance and twice the AP density. 802.11g achieves the compromise between 802.11a and 802.11b, offering both speed and distance. In terms of cost, 802.11g is slightly more expensive then 802.11b, but is getting less expensive as wireless becomes a standard feature in laptops.
Due to its wide deployment and maturity, it seems proper to start with 802.11b. 802.11b uses direct-sequence spread spectrum (DSSS) radio transmission for data delivery. DSSS works by transmitting the signal across several frequencies simultaneously, with the idea that one of the transmissions will make it to the receiver. The 802.11b DSSS model uses fourteen carrier signal channels. These carrier channels are the starting point for the transmission, which spreads into the frequency ranges above and below carrier frequency. Four data rates are supported: 1 Mbps, 2 Mbps, 5.5 Mbps and 11 Mbps. To ensure data integrity, 802.11b uses chipping schemes to encapsulate the actual data. The use of the chipping code scheme adds to the size of the data message (utilizing bandwidth for delivering the data, instead of moving actual data). This is because sending the data as chips increases the resilience of the data transmission, making it possible to reconstruct the data in the event of transmission interference. Once the data has been encoded, the chip is modulated and transmitted over the carrier signal channel.
The chipping and modulation schemes used with 802.11b differ as the transmission rate increases. Mbps 802.11b uses the Barker code chipping sequence (10110111000 using 11 bits to encode 1 bit, coupled with Binary Phase Shift Keying (BPSK) modulation. BPSK works by shifting the phase of the carrier signal 180 degrees in accordance with the digital data stream. Differences in the signal are detected by comparing the phase of each incoming bit to the phase of the preceding bit. When the bit takes a value "1"or "0," the carrier phase changes between 180 degrees and 0 degrees. The Barker code and Quaternary Phase Shift Keying (QPSK) modulation is used for Mbps. QPSK sends 2-bit symbols using four carrier phase shifts. 802.11b transmission rates of 5.5 and 11 Mbps use a combination of QPSK modulation and Complementary Code Keying (CCK) or the chipping sequencing. Actually CCK is used for 5.5 Mbps and CCK2 is used for 11 Mbps transmissions. CCK uses 64 unique code words and works by using complementary code sequences in combination with turning bits to transmit 4 and 8-bit data chips. So, at the operational rate of 5.5 Mbps, QPSK/CCK moves 4 data bits and at 11 Mbps, 8 data bits are moved. At each operation speed, a consistent chip transmission rate of 11 Mchip/s is maintained, with changes in encoding and modulation increasing the throughput over the same amount of bandwidth.
802.11b utilizes the 83 MHz Industrial, Scientific & Medical (ISM) band, which is crowded and prone to attenuation. This band is utilized by everything from wireless phones and microwaves to garage door openers. And just about everything (walls, posts, power lines, windows, people, etc.) can absorb, reflect and scatter the signal. When deploying an 802.11b wireless network, one of the most common mistakes made is working from the assumption that all of the channels are usable. This is not the case. The carrier frequency channels, by design, bleed into one another. Each 802.11b transmission occupies between 22 MHz of bandwidth, with only 5 MHz of passband bandwidth separating each of the 802.11b carrier channels. The DSSS transmission results in a 10Mhz bleed-through on either side of the CF. That limits 802.11B to three non-overlapping channels (1,6, and 11) spaced 25 MHz apart, limiting infrastructure AP deployments to three discrete access points within range to one another.
802.11a uses Orthogonal Frequency Division Multiplexing (OFDM). OFDM works by dividing the data transmission into multiple bit streams. The bit streams are then transmitted over parallel narrowband carriers or sub-carriers, carved out of the available channel bandwidth. The receiver reconstructs the sub-carrier into the original transmission signal. The 802.11a IEEE specification for OFDM defines 52 sub-carriers, 48 of which carry data, the remaining four carriers carry pilot data.
The IEEE protocol defines eight data transmission rates for 802.11a. Transmission rates of 6, 12 and 20 Mbps are mandatory. Support for 9, 18, 36, 48 and 54 Mbps transmission rates are optional, but are supported on most vendor's products. To accommodate the different transmission rates, 802.11a utilizes different modulation schemes. 802.11a does not utilize a chipping code, like 802.11b. OFDM is far more resistant to interference then DSSS. The lower 802.11a transmission rates use modulation schemes we covered in the 802.11b overview. BPSK modulation is used to transmit data at 6 and 9 Mbps transmission rates. QPSK modulation is used for transmission rates running at 12 and 18 Mbps. For the higher speed transmission rates, 24 thru 54 Mbps Quadrature Amplitude Modulation (QAM) is used. QAM is a digital frequency modulation technique that represents data as phase and amplitude symbols, each representing 4 data bits. 16-QAM, which supports 16 symbols is used for the 24 through 48 Mbps transmission rates. 64-QAM is used for 54 Mbps (and on some vender implementations 48 Mbps) transmissions.
802.11a operates using 300MHz of bandwidth in the 5Ghz Unlicensed National Information Infrastructure (U-NII) RF spectrum. The 300 Mhz of bandwidth is sub-divided in three 100 Mhz domains, each with different maximum operating power. The first 200 MHz is contiguous, operating between 5.200 GHz to 5.320. The 5.200 to 5.240 supports 50 mW max output, and 5.260 to 5.320 runs up to 250 mW. The last 100 MHz operates between 5.745 and 5.805 GHz, with a maximum output power of 1W. Each domain has four non-overlapping 20 MHz bandwidth channels, each of which can be utilized for transmission (unlike 802.11b, where the available channels bleed over one another).