The Original Direct Sequence PHY

The physical layer itself consists of two components. The Physical Layer Convergence Procedure (PLCP) performs some additional PHY-dependent framing before transmission, while the Physical Medium Dependent (PMD) layer is responsible for the actual transmission of frames.

PLCP Framing and Processing

The PLCP for the DS PHY adds a six-field header to the frames it receives from the MAC. In keeping with ISO reference model terminology, frames passed from the MAC are PLCP service data units (PSDUs). The PLCP framing is shown in Figure 12-14.

Figure 12-14. DS PLCP framing

The FH PHY uses a data whitener to randomize the data before transmission, but the data whitener applies only to the MAC frame trailing the PLCP header. The DS PHY has a similar function called the scrambler, but the scrambler is applied to the entirety of the direct-sequence frame, including the PLCP header and preamble.

Preamble

The Preamble synchronizes the transmitter and receiver and allows them to derive common timing relationships. It is composed of the Sync field and the Start Frame Delimiter field. Before transmission, the preamble is scrambled using the direct-sequence scrambling function.

Sync

The Sync field is a 128-bit field composed entirely of 1s. Unlike the FH PHY, the Sync field is scrambled before transmission.

Start Frame Delimiter (SFD)

The SFD allows the receiver to find the start of the frame, even if some of the sync bits were lost in transit. This field is set to 0000 0101 1100 1111, which is different from the SFD used by the FH PHY.

Header

The PLCP header follows the preamble. The header has PHY-specific parameters used by the PLCP. Five fields comprise the header: a signaling field, a service identification field, a Length field, a Signal field used to encode the speed, and a frame-check sequence.

Signal

The Signal field is used by the receiver to identify the transmission rate of the encapsulated MAC frame. It is set to either 0000 1010 (0x0A) for 1-Mbps operation or 0001 0100 (0x14) for 2-Mbps operation.

Service

This field is reserved for future use and must be set to all 0s.

Length

This field is set to the number of microseconds required to transmit the frame as an unsigned 16-bit integer, transmitted least-significant bit to most-significant bit.

CRC

To protect the header against corruption on the radio link, the sender calculates a 16-bit CRC over the contents of the four header fields. Receivers verify the CRC before further frame processing.

No restrictions are placed on the content of the Data field. Arbitrary data may contain long strings of consecutive 0s or 1s, which makes the data much less random. To make the data more like random background noise, the DS PHY uses a polynomial scrambling mechanism to remove long strings of 1s or 0s from the transmitted data stream.

DS Physical Medium Dependent Sublayer

The PMD is a complex and lengthy specification that incorporates provisions for two data rates (1.0 and 2.0 Mbps). Figure 12-15 shows the general design of a transceiver for 802.11 direct-sequence networks.

Figure 12-15. Direct-sequence transceiver

 

Transmission at 1.0 Mbps

At the low data rate, the direct-sequence PMD enables data transmission at 1.0 Mbps. The PLCP header is appended to frames arriving from the MAC, and the entire unit is scrambled. The resulting sequence of bits is transmitted from the physical interface using DBPSK at a rate of 1 million symbols per second. The resulting throughput is 1.0 Mbps because one bit is encoded per symbol. Like the FH PMD, the DS PMD has a minimum power requirement and can cap the power at 100 mW if necessary to meet regulatory requirements.

Transmission at 2.0 Mbps

Like the FH PHY, transmission at 2.0 Mbps uses two encoding schemes. The PLCP preamble and header are transmitted at 1.0 Mbps using DBPSK. Although using a slower method for the header transmission reduces the effective throughput, DBPSK is far more tolerant of noise and multipath interference. After the preamble and header are finished, the PMD switches to DQPSK modulation to provide 2.0-Mbps service. As with the FH PHY, most products that implement the 2.0-Mbps rate can detect interference and fall back to lower-speed 1.0-Mbps service.

CS/CCA for the DS PHY

802.11 allows the carrier sense/clear channel assessment function to operate in one of three modes:

Mode 1

When the energy exceeds the energy detection (ED) threshold, it reports that the medium is busy. The ED threshold depends on the transmit power.

Mode 2

Implementations using Mode 2 must look for an actual DSSS signal and report the channel busy when one is detected, even if the signal is below the ED threshold.

Mode 3

Mode 3 combines Mode 1 and Mode 2. A signal must be detected with sufficient energy before the channel is reported busy to higher layers.

Once a channel is reported busy, it stays busy for the duration of the intended transmission, even if the signal is lost. The transmission's duration is taken from the time interval in the Length field. Busy medium reports must be very fast. When a signal is detected at the beginning of a contention window slot, the CCA mechanism must report a busy medium by the time the slot has ended. This relatively high performance requirement must be set because once a station has begun transmission at the end of its contention delay, it should seize the medium, and all other stations should defer access until its frame has concluded.

Characteristics of the DS PHY

Table 12-4 shows the values of a number of parameters in the DS PHY. In addition to the parameters in the table, which are standardized, the DS PHY has a number of parameters that can be adjusted to balance delays through various parts of an 802.11 direct-sequence system. It includes variables for the latency through the MAC, the PLCP, and the transceiver, as well as variables to account for variations in the transceiver electronics. One other item of note is that the total aggregate throughput of all direct-sequence networks in an area is much lower than the total aggregate throughput of all nonoverlapping frequency-hopping networks in an area. The total aggregate throughput is a function of the number of nonoverlapping channels. In North America and most of Europe, three direct-sequence networks can be deployed in an area at once. If each network is run at the optional 2 Mbps rate and the efficiency of the protocol allows 50% of the headline rate to become user data throughput, the total throughput is 3 Mbps, which is dramatically less than the frequency-hopping total aggregate throughput.

Table 12-4. DS PHY parameters

Parameter	Value	Notes
Slot time	20 ms
SIFS time	10 ms	The SIFS is used to derive the value of the other interframe spaces (DIFS, PIFS, and EIFS).
Contention window size	31 to 1,023 slots
Preamble duration	144 ms	Preamble symbols are transmitted at 1 MHz, so a symbol takes 1 ms to transmit; 144 bits require 144 symbol times.
PLCP header duration	48 ms	The PLCP header is 48 bits, so it requires 48 symbol times.
Maximum MAC frame	4-8,191 bytes
Minimum receiver sensitivity	-80 dBm
Adjacent channel rejection	35 dB	See text for measurement details.

Like the FH PHY, the DS PHY has a number of attributes that can be adjusted by a vendor to balance delays in various parts of the system. It includes variables for the latency through the MAC, the PLCP, and the transceiver, as well as variables to account for variations in the transceiver electronics.