List of Tables
2.9 Resource Mapping of DL Control Channels [10]
Again this section is mainly taken from [10].
Figure 2.35: Physical processing block diagram for the S-SFH (Figure 519 in [10]).
2.9.1 Superframe Header
Primary Superframe Header
The P-SFH IE shall be appended with 5 bits CRC. The generating polynomial is G(x) = x5 + x4 + x2 + 1. The resulting sequence of bits shall be encoded by the TBCC with parameter M = NRep,P −SF H and Kbuf size = 4L, where L is the number of information bits and NRep,P −SF H is the number of repetition for effective code rate of 1/16 or 1/24.
The modulated symbols shall be mapped to two transmission streams using SFBC. The two streams using SFBC shall be precoded and mapped to the transmit antennas.
Antenna specific symbols at the output of the MIMO encoder/precoder shall be mapped to the resource elements.
Secondary Superframe Header
Figure 519 shows the physical processing block diagram for the S-SFH.
The S-SFH IE shall be appended with a 16-bit CRC.
The resulting sequence of bits shall be encoded by the TBCC with parameter M = NRep,S−SF HKbuf size and Kbuf size = 4L, where L is the number of information bits.
The value of NRep,S−SF H is indicated in P-SFH.
The encoded bit sequences shall be modulated using QPSK.
The modulated symbols shall be mapped to two transmission streams using SFBC. The
Figure 2.36: Chain of non-user specific A-MAP IE to non-user specific A-MAP symbols (Figure 520 in [10]).
two streams using SFBC shall be precoded and mapped to the transmit antennas.
Antenna specific symbols at the output of the MIMO encoder/precoder shall be mapped to the resource elements.
Advanced MAP (A-MAP)
SFBC with precoding shall be used for the A-MAP region.
Non-user Specific A-MAP
The coding chain for non-user-specific A-MAP IE to non-user-specific A-MAP symbols is shown in Figure 2.36. The non-user specific A-MAP IE is encoded by TBCC with parameter M = 3Kbuf size and Kbuf size = 4L for 1/12 code, where L is the number of information bits.
In FFR configurations, the non-user specific A-MAP is also encoded with a code rate of 1/12 when it is in the frequency reuse 1 partition. When the non-user specific A-MAP is in the power-boosted frequency reuse 3 partition, it should be encoded with parameter M = Kbuf size = 4L for 1/4 code rate. The encoded sequence is modulated using QPSK.
For each Tx antenna, symbols at the output of MIMO encoder, denoted by SN U S[0] to SN U S[LN U S − 1], are mapped to tone-pairs from RMP [(LHF + LP C)/2] to RMP [(LHF + LP C + LN U S)/2 − 1], where RMP refers to the renumbered A-MAP tone-pairs and LHF is the number of tones required to transmit the entire HARQ feedback A-MAP; LP C is the number of tones required to transmit the entire power control A-MAP; LN U S is the number
Figure 2.37: Chain of HF-A-MAP IE to HF-A-MAP symbols (Figure 521 in [10]).
of tones required to transmit the non-user specific-A-MAP.
HARQ Feedback A-MAP
HARQ feedback A-MAP (HF-A-MAP) contains HARQ-feedback-IEs for ACK/NACK feed-back information to uplink data transmission.
Figure 2.37 shows the construction procedure of HF-A-MAP symbols from HF-A-MAP-IE. Each HF-A-MAP IE carries 1 bit information. Firstly, it is repeated NRep,HF −A−M AP
times, where NRep,HF −A−M AP is 8. Then, Repeated HF-A-MAP IE bits are scrambled by the NRep,HF −A−M APLSBs of the STID of the associated AMS. Depending on the channel conditions, power scaling can be applied to each scrambled sequence. Before MIMO encod-ing, each scrambled sequence is mapped to either real part or imaginary part in the signal constellation and multiplexed with other scrambled sequence, if exist.
Figure 2.37 shows a cluster of HF-A-MAP channels, which consists of 4 HF-A-MAP channels numbered as 4c, 4c + 1, 4c + 2, 4c + 3 where c is the HF-A-MAP cluster index in a HF-A-MAP region. Channel 4c in the cluster occupies the real part of the first symbol in each tone pair before the SFBC encoder. Channel 4c + 1 in the cluster occupies the imaginary part of the first symbol in each tone pair before the SFBC encoder. Channel
4c + 2 in the cluster occupies the real part of the second symbol in each tone pair before the SFBC encoder. Channel 4c + 3 in the cluster occupies the imaginary part of the second symbol in each tone pair before the SFBC encoder.
For each Tx antenna, symbols at the output of MIMO encoder, denoted by SHF[0] to SHF[LHF−1], are mapped to tone-pairs from RMP [0] to RMP [LHF/2−1], where RMP refers to the renumbered A-MAP tone-pairs and LHF is the number of tones required to transmit the entire HARQ feedback A-MAP. Clusters of the HF-A-MAP are indexed sequentially from index 0 within an HF-A-MAP region in the mapping process.
There is one HF-A-MAP region in each DL AAI subframe. Within each HF-A-MAP region, the index for HF-A-MAP channel is calculated as follows.
For the deallocation of a persistent allocation, the HF-A-MAP resource index is specified in the HFA the of UL Persistent Allocation A-MAP IE.
For group resource allocation, the HF-A-MAP resource index for the lth AMS in the GRA allocation is (istart+ ⌊l · NHF −A−M AP/NGRA⌋) mod NHF −A−M AP, where istart is the ACK Channel Offset in the UL group resource allocation A-MAP IE, NHF −A−M AP is the total number of HF-A-MAP configured per HF-A-MAP region, and NGRA is the User Bit Map Size in the UL group resource allocation A-MAP IE.
For the BR-ACK A-MAP IE, the HF-A-MAP resource index for the lth AMS grant in the BR-ACK bitmap is (istart+ ⌊l ·NHF −A−M AP/NBR−ACK⌋) mod NHF −A−M AP, where istart
is the HFA start offset in the BR-ACK A-MAP IE, NHF −A−M AP is the total number of HF-A-MAP channels configured per HF-HF-A-MAP region, and NBR−ACK is the number of AMSs with grants in the BR-ACK A-MAP IE.
For resource allocation using the UL Basic Assignment A-MAP IE, UL Subband Assign-ment A-MAP IE, CDMA Allocation A-MAP IE and the UL Persistent Allocation A-MAP
Figure 2.38: Chain of PC-A-MAP IE to PC-A-MAP symbols (Figure 522 in [10]).
IE, the HF-A-MAP resource index is (M(j) + n) mod NHF −A−M AP, where j is HF-A-MAP Index Parameter in the Non-user specific A-MAP IE, n is a 3 bit HFA value in each assign-ment A-MAP IE, NHF −A−M AP is the total number of HF-A-MAP channels configured per HF-A-MAP region. M(j) is STID for the UL Basic Assignment A-MAP IE and UL Subband Assignment A-MAP IE and RA-ID for CDMA Allocation A-MAP IE when j = 0 and M(j) is the lowest LRU index of the corresponding UL transmission when j = 1. For the UL persistent allocation A-MAP IE, M(j) is always STID regardless of value j.
Power Control A-MAP
Power Control A-MAP (PC-A-MAP) contains PC-A-MAP-IEs for closed-loop power control of the uplink transmission. The ABS shall transmit PC-A-MAP-IE to every AMS which operates in closed-loop power control mode.
Figure 2.38 shows the construction procedure of PC-A-MAP symbols from PC-A-MAP-IE. The ith PC-A-MAP-IE shall have the size of 2 bits according to power correction value.
The ith and (i + 1)th PC-A-MAP IEs shall be mapped to two QPSK symbols as depicted in Figure 2.38. Only the ith PC-A-MAP may also be mapped to two QPSK symbols for transmitting to the corresponding MS with poor channel quality.
Power scaling by sqrt(Pi)(0 ≤ i ≤ NP C−A−M AP −IE) shall be applied to the ith
PC-A-Figure 2.39: Chain of A-A-MAP IE to A-A-MAP symbols (PC-A-Figure 523 in [10]).
MAP-IE where NP C−A−M AP −IE is the number of PC-A-MAP-IEs and sqrt(Pi) is the value determined by the management entity to satisfy the link performance.
The QPSK symbols are repeated Nrep,P C−A−M AP −IE times, where Nrep,P C−A−M AP −IE
equals two.
Figure 2.38 shows a cluster of PC-A-MAP channels, which consists of 2 PC-A-MAP channels numbered as 2c and 2c + 1 where c is the PC-A-MAP cluster index in the A-MAP region. Channel 2c in the cluster occupies the real part of both symbols in each tone pair before the SFBC encoder. Channel 2c + 1 occupies the imaginary part of both symbols in each tone pair before the SFBC encoder.
For each Tx antenna, symbols at the output of MIMO encoder, denoted by SP C[0] to SP C[LP C − 1], are mapped to tone-pairs from RMP [(LHF/2] to RMP [(LHF + LP C)/2 − 1], where RMP refers to the renumbered A-MAP tone-pairs and LHF is the number of tones required to transmit the entire HARQ feedback A-MAP; LP C is the number of tones required to transmit the entire power control A-MAP. Clusters of the PC-A-MAP are indexed sequentially in the mapping process.
Assignment A-MAP
The Assignment A-MAP (A-A-MAP) shall include one or multiple A-A-MAP IEs and each A-A-MAP IE is encoded separately. Figure 2.39 describes the procedure for constructing A-A-MAP symbols.
A 16-bit CRC is generated based on the contents of the assignment A-MAP IE. Denote the
assignment A-MAP IE by m(x) = bN −1xN −1+bN −2xN −2+...+b1x+b0, where bN −1is the MSB of the assignment A-MAP IE and b0 is the LSB of the assignment A-MAP IE. The 16-bit CRC is calculated as the remainder of dividing m(x)x16 by the 16-bit CRC generator polynomial g(x) = x16+x12+x5+1. The resulted CRC is denoted by p(x) = p15x15+p14x14+...+p1x+p0
where p15 is the MSB of the CRC and p0 is the LSB of the CRC.
The 16-bit CRC mask is then applied to the 16-bit CRC by bitwise XOR operation.
The masked CRC is then appended to the assignment A-MAP IE, resulting in a bit sequence of m(x) ∗ x16+ p(x) ⊕ q(x). The resulting sequence of bits shall be encoded by the TBCC.
Coded bits can be repeated to improve the robustness of an A-A-MAP channel based on the link condition of a particular AMS.
For a given system configuration, assignment A-MAP IEs can be encoded with two different effective code rates. The set of code rates is (1/2, 1/4) or (1/2, 1/8) and is explicitly signaled in the S-SFH.
In case of FFR, two code rates, either (1/2, 1/4) or (1/2, 1/8), can be used in the reuse 1 partition. 1/2 or 1/4 code rate is used in the power-boosted reuse 3 partition and signaled in the S-SFH.
The parameters for TBCC are M = Kbuf size= 2L for 1/2 code rate, M = Kbuf size= 4L for 1/4 code rate, and M = 2Kbuf size and Kbuf size = 4L for 1/8 code rate where L is the number of information bits. The encoded bit sequences shall be modulated using QPSK.
2.9.2 Downlink Power Control
The ABS should be capable of controlling the transmit power per AAI subframe and per user. An ABS can exchange necessary information with neighbor ABS through backbone
network to support downlink power control.
Power Control for A-MAP
Downlink transmit power density of A-MAP transmission for an AMS may be set in order to satisfy target error rate for the given MCS level which is used for the A-MAP transmission.
Detail algorithm is left to vendor-specific implementations.
Chapter 3
Introduction to IEEE 802.16m MIMO
We introduce the MIMO transmission mechanism in IEEE 802.16m. Much of the material is taken from [10].