ERP Physical Medium Dependent (PMD) Layer

Once the PLCP frame is ready for transmission, it is dispatched to the Physical Medium Dependent (PMD) layer. The PMD is responsible for taking the data and sending it out the antenna. Due to the wide variety of modulation schemes that might be used, an 802.11g transceiver must implement several different transmission modes, either wholly or partially, and switch between them as needed. Some functions, however, are shared by all transceivers regardless of operating mode.

Sidebar 3 1 Atheros Super G Extensions

Although the speed has increased dramatically in the past several years, wireless networks are still not fast. Under optimum conditions, a standards-based wireless network can achieve about 30 Mbps. In the race to get the highest number possible on the side of the box, nearly all chipset manufacturers have implemented extended functions to increase speed. By far, the most notable (or notorious) was Atheros' Super G enhancements, with three good features, and one that has drawn a great deal of fire.

       

Block acknowledgment (also called frame bursting)

In the standard 802.11 radio link management scheme, every data frame must be followed by an acknowledgment, after which the station must re-contend for the medium. Block acknowledgments allow a station to transmit several frames, separated only by the short interframe space so there is no contention for the medium. The sequence is then followed by a single block acknowledgment. Block acknowledgment developed according to the current 802.11e drafts is generally interoperable between vendors.

Packet clustering

The 802.11 frame can hold much more data than a standard 1,500 byte Ethernet-size frame. Using the full payload size improves the payload-to-overhead ratio, and increases speed.

Hardware data compression

By compressing data prior to transmission, it requires less time to transmit and gives an effective throughput increase to the link. Hardware compression is most effective when there is redundant data, such as uncompressed text. It is not as effective when transmitting compressed images.

Channel bonding (also called "turbo mode")

Instead of using a single 22 MHz channel, the bonding feature doubles the bandwidth. With more spectrum to spread the signal over, a higher number of bits can be pushed through the now-wider channel. To ensure that the wider bandwidth stays within regulatory limits, channel bonding is restricted in some products to operate in the middle of the band on channel 6.

By far the most controversial feature of the Super G feature set is the channel bonding. In late 2003, Broadcom accused Super G of interfering with regular 802.11g transmissions, and had a private demonstration at Comdex. (I was barred from observing.) Later tests revealed some interference, but found that Broadcom's chipsets were more susceptible to degredation than those of other vendors. Atheros has implemented a mode which monitors for standard 802.11g transmissions, and refrains from using channel bonding if other networks are present.

The main lesson is that any channel bonding feature is likely to increase interference by sucking up more of the available scarce spectrum. Spectrum-hungry transmissions modes are unsuitable for use in a large area requiring multiple access points because they cut the number of available channels from an already-too-low three to one. Deploy a channel bonding device in your home, but keep it off your company's network.

Figure 14-7. DSSS-OFDM frame format

 

Clear Channel Assessment (CCA)

Only one CCA mode is defined for 802.11g, which combines a minimum energy threshold with the ability to decode a signal. Energy detection is based on receiving a valid signal at the start of a transmission slot with a signal power of -76 dBm or greater. As a performance requirement, within a specified window, the PHY should have a high probability of correctly reporting the medium busy. Both the time window and the probability are shown in Table 14-2.

Table 14-2. 802.11g CCA performance requirements

	Long slot (20 ms)	Short slot (9 ms)
CCA time	15 ms	4 ms
Detection probability	>99%	>90%

CCA is integrated with a PLCP-level virtual carrier sense. When a PLCP header is received, it will include a length field that indicates the amount of time the medium will be busy. The physical layer will continue to report the medium busy for that time period, even if the physical signal is lost. (Note that this is similar in concept and operation to the Network Allocation Vector at the MAC layer.) Part of the reason for doing this is that not all implementations will support all the transmission modes, so it is important that the physical layer correctly avoids interfering with transmissions it cannot demodulate.

Reception Procedure

802.11g stations have a more complicated procedure for receiving frames than chips that implement other standards because of the choice and backwards compatibility. When an incoming frame is detected, an 802.11g station will need to detect it and then demodulate it with the correct physical layer.

1. Is the preamble an OFDM preamble (like 802.11a), or the traditional single-carrier preamble used by 802.11b? Frames modulated with OFDM will be processed exactly the same as 802.11a frames, just received on a different frequency.
2. If the frame is not an OFDM frame, it must decode the preamble, and find the end of the preamble to see the PCLP SIGNAL and SERVICE fields.
3. By decoding the PLCP header, the appropriate modulation can be employed to demodulate the frame body.
   1. If the SERVICE field indicates that the frame is modulated with PBCC, the PBCC reception procedures will be triggered.
   2. If the SERVICE field doesn't indicate the presence of PBCC, the data rate is checked. Data rates of 1 and 2 Mbps are processed with the DSSS reception algorithm, and data rates of 5.5 Mbps and 11 Mbps are processed according to the CCK reception algorithm; both are described in Chapter 12.
   3. If the SIGNAL field indicates a speed of 3 Mbps, DSSS-OFDM reception is used. It will switch the demodulator to receiving OFDM at the end of the PCLP header.

Characteristics of the ERP PHY

802.11g has very similar characteristics to 802.11a, with one notable exception. Although each channel has similar performance to an 802.11a channel, there are only three channels. If each channel is run at the highest data rate and 50% efficiency, the total aggregate throughput is only 81 Mbps. It is a high number, but does not begin to approach 802.11a.

Table 14-3. ERP PHY parameters

Parameter	Value	Notes
Maximum MAC frame length	4,095 bytes
Slot time	20 ms 9 ms	If the network consists only of 802.11g stations, the slot time may be shortened from the 802.11b-compatible value to the shorter value used in 802.11a.
SIFS time	10 ms	The SIFS is used to derive the value of the other interframe spaces (DIFS, PIFS, and EIFS).
Signal extension time	6 ms	Every 802.11g packet is followed by the signal extension time.
Contention window size	15 or 31 to 1,023 slots	If the station supports only 802.11b rates, it will be 31 slots for compatibility. Otherwise, the contention window may be shorter.
Preamble duration	20 ms