Pusch upstream hopping (2)-type2 frequency hopping

"Pusch (1)-type1 frequency hopping" mentioned why to use the Pusch hopping frequency, as well as detailed introduction of the TYPE1 mode of frequency hopping, this article continues this topic, introduces the TYPE2 mode of frequency hopping.

1. Problems needing attention when using Pusch frequency hopping

In the upstream sub-frame, the PUCCH channel is at the high and low sides of the bandwidth, or at the edge of the band, and the Pusch channel is in the middle of the bandwidth. The PUCCH channel is also a RB pair as the basic unit, each RB in the frequency domain is 12 sub-carrier, the time domain is 1 time slots. It is necessary to pay attention to the two RB positions of each RB pair of the PUCCH channel: The PUCCH channel of the first time slot is located in the low frequency position of the bandwidth and the PUCCH channel of the second timeslot is at the high frequency position of the bandwidth. As shown in.

(Fig. 1)

The reason for such a set of PUCCH and Pusch position, there are two aspects of consideration: first, the pucch on the edge of the spectrum, you can make the control signaling frequency diversity maximization (the PUCCH channel is the bearer of control signaling); second, Pusch must use continuous RB, which restricts the allocation of Pusch resources if the PUCCH is placed in the middle of the spectrum. For example, a certain time, the system only 1 UE in the upstream big data volume transmission, if the pucch in the middle of the band, will cause ENB can only give the UE to allocate a limited RB resources, resulting in waste of resources at the same time, can not meet the UE traffic demand. As shown in.

(Fig. 2)

If there is overlap between the Pusch and the PUCCH channel in the same sub-frame, the decoding of the data in the Pusch and PUCCH channels is affected. ENB When considering the Pusch frequency hopping, the mapped prb position cannot overlap with the pucch position .

2.PUSCH frequency Hopping Mode 2 (Type2 PUSCH hopping)

In this paper, we first introduce the frequency hopping method of Type2, and then combine the mathematical formula to make further explanation.

Type2 is a sub-band frequency hopping method based on the cell-specific frequency hopping pattern (hopping pattern) and the image pattern (mirroring pattern) . There are three things to be aware of in this sentence:

First, Type2 is based on the frequency hopping of the sub-band, we can divide the entire bandwidth 1~4 sub-band, which is different from Type1, Type1 and no sub-band concept.

Second, thefrequency hopping pattern of the Type2 is specific to the cell, which means that the frequency hopping pattern is not the same for different communities, and the frequency hopping pattern used by different UE is the same in the same cell . However, there is an exception, that is, if the entire bandwidth is divided into 1 sub-bands, then the frequency hopping pattern of different cells may be the same, this can be confirmed by the mathematical formula in the later.

Thirdly, the Type2 has a mirroring mode, which can further increase the complexity of the frequency hopping, and can further avoid the same frequency hopping results as the adjacent area.

is a Pusch sub-band, in this figure, the entire upstream bandwidth corresponds to 50 RB, a total of 4 sub-bands, each of which includes 11 RB. It is important to note that the sub-band does not cover the entire upstream band , and the PUCCH channel on both sides of the band is not in the sub-band range.

(Fig. 3)

Here in a word to summarize the steps of Type2 frequency hopping: based on the frequency hopping of the sub-band, is based on the ul_grant dispatching authorization given VRB location, using a cell-specific frequency hopping pattern, through the shift mapping to the corresponding PRB location . This frequency hopping pattern provides different distances for different time slots to be shifted . As shown, ENB assigns a continuous set of VRB to a UE, RB27, RB28, and RB29, respectively. In the first time slot, the predefined frequency hopping pattern value is 1 (that is, hopping pattern= 1, corresponding to the calculation results of the frequency hopping function f_hop (i) in the following formula, which is first understood here), meaning that the VRB pattern needs to be moved to the right 1 sub-bands , so that the location of PRB is RB38, RB39, RB40, respectively. When calculating the second timeslot, the HP value is 3, which means that the VRB pattern (i.e. RB27, RB28, RB29) is moved to the right by 3 sub-bands , so the location of PRB is RB16, RB17, RB18, respectively.

(Fig. 4)

Say the frequency hopping pattern (hopping pattern), say the mirror pattern (mirroring pattern). A mirrored pattern is the use of mirroring within a sub-band of a timeslot for the allocated resource. In other words, mirroring mode does not shift the position of PRB from one sub-band to another, but the mirrored symmetry offset within the sub-band, which understands the mirroring pattern. For example, the second time slot in Figure 4 above is mirrored offset, then the structure will be shown in Figure 5 below (this figure of the dotted line labeled VRB to PRB mapping relationship and not, should be the corresponding location of the red Line callout, refer to Figure 11 below).

(Fig. 5)

In Figure 5, the second time slot uses the mirror pattern (that is, the mirroring pattern=yes, corresponding to the following formula in the image function F_m (i) of the calculation results, here first understand), so the original book with 1 within the location of PRB is RB16, RB17, RB18 , after the deflection of the mirror, was shifted to the RB20, RB21, RB22 of the genus 1.

The above summarizes the two methods of Type2 frequency hopping (frequency hopping pattern and image pattern), and then uses the mathematical formula to illustrate the results of the two methods. If you just want to know about the Type2 frequency hopping, then you can see here.

In the previous tutorial on Type1 's frequency hopping, it was mentioned that in the first timeslot, the position of PRB is actually the same as the position of VRB, and in the second time slot, the final prb has a certain offset from the VRB. When explaining the Type1 calculation of the PRB location formula, it can be noted that the formula for the first and second time slots is different, and the second time slot has a much more complex formula. for the Type2 type of frequency hopping, the PRB position of the two time slots is calculated by the same formula , and the time slot NS is used as a parameter to calculate the location of PRB. The specific formula is as follows, where the PRB position of the NS timeslot is indicated by the parameter N_PRB (NS) .

(Fig. 6)

Here are some explanations for this formula:

(1) Our ultimate goal is to calculate the location of PRB in two slots, which is the PRB position N_PRB (NS)of the time slot NS, which is related to parameters such as NSB, N_HO_RB, Hoppingmode, and so on. The NSB parameter indicates that the current Pusch area is divided into several sub-bands, the range is 1~4, that is, a maximum of 4 sub-bands, the value of the RRC configuration parameters N-SB obtained. The N_HO_RB parameter is pusch-hoppingoffset by the parameters of the RRC configuration, and thehoppingmode parameter determines whether the "sub-frame hopping mode" or "sub-frame inter-frame hopping mode" is used. The parameters N-SB, Pusch-hoppingoffset, and Hoppingmode parameters of the RRC configuration are derived from the SIB2 cells in Pusch-config as shown in.

(Fig. 7)

(2) in the formula, use N_SB_RB to represent the number of RB currently occupied by each sub-band. If nsb=1, then N_SB_RB equals the entire bandwidth of the RB number N_UL_RB. If the current 10M bandwidth, RRC configuration N-SB = Pusch-hoppingoffset = 4, then according to the formula above, each sub-band occupies the number of RB N_sb_rb=floor ((50-4-4 mod 2)/4) = 11, that is, each sub-band occupies 11 RB, At this time, the Pusch of the sub-band RB distribution as shown, the bandwidth on both sides of the 6 RB for the use of the PUCCH channel.

(Fig. 8)

(3) in the formula, the meaning ofcurrent_tx_nb in the Type1 type of frequency hopping is also useful to indicate the current TB block HARQ transmission times. This parameter is only valid in "sub-frame frequency hopping" mode, because in the sub-frame hopping mode, the location of PRB with the same sub-frame, non-simultaneous gap needs to be consistent, refer to the blog "Pusch Uplink frequency hopping (1)-type1 frequency FH" in the figure (3). While the frequency hopping does not guarantee that the two time slot prb location is consistent, so it is necessary to make appropriate adjustments according to the CURRENT_TX_NB value, to ensure that the frequency hopping between sub-frames, the two time slots of the PRB position consistent.

(4) The image function in the formula f_m (i) evaluates to only two values: 0 (FALSE) or 1 (TRUE), indicating whether a time slot introduces a "mirror pattern" (mirroring) frequency hopping, If the calculation of the function f_m (i) is equal to 1 in a timeslot NS, then the location of PRB for that time slot requires a "mirror" operation. The previous Figure 5 depicts an example of a mirroring pattern, where a formula is used to calculate the PRB position of the second timeslot after mirroring mode 15:

Depending on the sub-band configuration in Figure 5, the conditions that can be determined are: (a) Upstream 10M bandwidth, nsb=4, n_sb_rb=11 (b) n_vrb=27, 28, 293 RB (c) The first time slot is not mirrored, f_m (i) = 0, the second slot is mirrored, F_m (j) =1 (d) When the first timeslot is not mirrored, the final position of PRB is n_prb=38, 39, 40, and the final position of PRB is n_prb=16, 17, 18, respectively, when the second timeslot is not mirrored. To know the final position of PRB after the second time slot is mirrored, you also need to know the value of the n_ho_rb and function f_hop , depending on the formula. However, since the second time slot does not take the image of the PRB location has been determined, we can use this to reverse the introduction of N_HO_RB and f_hop values, and then you can calculate the second time slot in the mirror mode after the PRB location. Here is the calculation process: (Figure 9) from the above results, you can see that after the predefined frequency hopping pattern and image pattern are executed, the number 27th VRB will be mapped to the number 22nd PRB. Similarly, it can be inferred that number 28th VRB will be mapped to number 21st prb,29 VRB will be mapped to PRB 20th. If you do not do the mirroring mode of PRB location 10, then the PRB location after mirroring will be shown in 11. (Fig. 10) (Fig. 11)

(5) The C sequence used in the formula, please refer to the post "LTE downlink Physical layer transport mechanism (1)-Antenna port antenna port and cell-specific reference signal CRS" in the relevant description.

(6) The frequency hopping function in the formula F_hop (i) is a recursive function, F_hop (-1) = 0.

3. Summary

Pusch frequency hopping related content summary as shown in 12, through this graph we can deepen the understanding of Pusch frequency hopping.

Reference documents:

(1) 3GPP TS 36.212 V9.4.0 (2011-09) multiplexing and channel coding

(2) 3GPP TS 36.213 V9.3.0 (2010-09) Physical layer procedures

(3) 3GPP TS 36.211 V9.1.0 (2010-03) physical Channels and modulation

(4) http://www.sharetechnote.com

(5) << 4G lte/lte-advanced for Mobile Broadband >>

Pusch upstream hopping (2)-type2 frequency hopping