Retransmission Policy | Wimax Radio

Retransmission policy (automatic repeat request – ARQ) is the set of rules or processes used by networks to define if, when, and how retransmissions of data or information will occur. Some types of services (such as real time digital audio) do not use retransmissions because the delay for retransmission would take too long to offer any benefit.

Retransmission uses error detection, feedback, retransmission processes and the retransmission of blocks of data in packets. The WiMAX system uses two forms of ARQ; hybrid automatic repeat request (HARQ) and selective repeat (SR).

Selective Repeat (SR)

Selective repeat automatic repeat request is a data transmission control process that allows the receiver to request the retransmission of selective blocks of data.

Data to be transmitted is grouped into blocks and given a block sequence number (BSN). The maximum size of a data block in the WiMAX system is 2040 bytes. When blocks are transmitted using ARQ, each block is given a sub-packet identifier (SPID). An SPID is an index value that can be used to identify specific packets that are awaiting conformation in an automatic repeat request (ARQ) process.

As blocks are transferred between the sending device and the receiving device, acknowledgement messages are sent. WiMAX acknowledgement message types include selective, cumulative and cumulative with selective. ARQ messages can be sent as separate messages or the may be combined with other messages.

Hybrid Automatic Repeat Request (HARQ)

Hybrid automatic repeat request is a data transmission flow control process that uses a combination of the physical layer (PHY) and medium access control (MAC) layer to allow the receiver to stop and restart the retransmission of data over a transmission channel.

HARQ is a variation of a stop and wait ARQ. Stop and wait ARQ is a flow control process that allows the data flow to stop when packets are not received and wait until a successful retransmission is received before the data flow is restarted.

Sub Channelization (Sub-carriers)

Sub channelization is the dividing of communication channels into smaller sub-parts. The air interface portions of the WiMAX system divide a wide radio channel into several sub-carriers. A sub-carrier is a modulation signal that is imposed on another carrier that can be used to independently transfer information from other sub-carriers located on the radio channel. The WiMAX system sub-carrier signal types include a pilot sub-carrier (reference signal), guard sub-carrier (interference protection) and data sub-carrier (user information).

A pilot sub-carrier is a reference signal that serves as a control signal for use in the reception of other sub-carrier signals. A guard sub-carrier is one or more sub-carriers on a communication channel that is not used (null channel). The guard sub-carrier is dedicated to the protection of a communication channel from interference due to radio signal energy or time overlap of signals. Data sub-carriers are transmission channels that carry user information or data. Sub-carrier signals are referenced from the center of the radio channel. The sub-carrier that is located at the center of the radio channel is called the DC sub-carrier.

The number of sub-carriers in the WirelessMAN-OFDM system is 256. Of these, 55 are reserved as guard bands and 8 sub-carriers are used to transfer reference pilot signals. This allows up to 192 sub-carriers to be used for data transfer.

For a 20 MHz WiMAX OFDMA radio channel, there may be 2048 subcarriers. Of these, approximately 70% can be used as data carriers, 25% are reserved to protect from interference (guard bands) and approximately 15% are used as reference signals (pilot channels).

Figure 1 depicts how the WirelessMAN-OFDM system divides a wide radio channel into several independent orthogonal channels with smaller bandwidth. The sub-channel in the center of the RF channel is called the DC subcarrier. Some of the sub-carriers are used as reference pilot channels and some are reserved for guard bands.

Figure 1: WiMax OFDMA Sub Channelization

Payload Header Suppression (PHS)

PHS is the process of removing or blocking the transfer of packet header information. Payload header suppression is usually performed to remove redundant or unnecessary information such as a source and destination address that does not change for packets that are sent on a fixed communication link.

The PHS process begins with requesting a PHS session and negotiating the parameters or rules on how PHS will operate during the session to determine which bits and how many bits of the header may be altered or removed. PHS suppression operation may include the removal of address bits and other control information (such as IP address port number) that may be part of the header.

The header information is stored at the sending end and receiving ends to allow for the detection and removal of the header (the sending end) and the reinsertion of the header at the receiving end. The header is removed by using a payload header suppression mask (PHSM). The PHSM is a code or binary sequence that is used to allow, block, or modify specific bits in a header to create the information that will be transmitted. To allow the receiver to recreate information that is located within the header that sequentially changed, a payload header suppression index (PHSI) is used. The PHSI is an incremental value that is used to identify the sequence of payload header suppression messages.

When the compressed packet arrives at the receiver, the packet header information is recreated and inserted into the packet so the original data packet (header and data) is completely recreated.

Figure 1 shows how PHS can be used to increase the data transmission rate through a communication channel. An IP communication session that occurs over an unchanging (circuit switched) wireless data link can use PHS to increase the efficiency (higher data throughput) by removing redundant packet header data. When an IP session is setup over a circuit switched connection, the system first identifies that PHS will be used. The system then negotiates for which parts of the header will be changed or removed during transmission (establishes PHS rules). The negotiation associates (maps) the IP communication system to the data link connection and stores the unchanging information at each end of the communication link (the header mask). For each IP packet that is received, the IP address information and some control information is removed prior to transmitting the packet on the data link (packet compression). When the compressed packet arrives at the receiver, the IP address and packet header information is re-inserted on the packet so the IP data packet is completely recreated.

Figure 1: Payload Header Suppression