Tech-invite3GPPspaceIETFspace
21222324252627282931323334353637384‑5x

Content for  TR 38.844  Word version:  18.0.0

Top   Top   Up   Prev   Next
0…   6…   6.1.2…   6.1.2.2   6.1.2.3   6.1.2.4   6.1.3   6.2…   6.2.2…   6.2.3   6.3…   6.4…   6.5…   6.6…   6.7…   7…   A…

 

6.7  RAN4 standard impact identificationp. 31

6.7.1  Larger Channel BW than licensed BWp. 31

UE asymmetric bandwidth combinations for the corresponding spectrum scenarios are needed, handled on band by band basis or by generic requirement.

6.7.2  Combined UE CBW (one cell)p. 31

Overall impact limited since both the "main RF carrier" and the "additional RF carrier" would conform to existing 3GPP requirements, to guarantee co-existence

6.7.3  Overlapping CA (two cells)p. 31

The CA framework defines the transmitter emission and receiver blocking at the edge of carriers. Limited impacted is expected to enable coexistence with neighbouring channels for this scenario. Overlapping CA channel spacing need to be updated to consider channel raster, minimum guard band and RB alignment.
For 15 KHz SCS and 100 KHz channel raster,
formula#2 in TR6#32;38.444
For NR operating bands with 15 kHz channel raster,
formula#3 in TR6#32;38.444
with
n=μ0
Where BW is the bandwidth not aligned with the existing NR channel bandwidths, BWchannel(1) and BWchannel(2) are channel bandwidths of the overlapping carriers respectively.
For overlapping CA approach, the only difference compared to normal CA is that overlapping CA can further adjust the channel spacing to fit the irregular spectrum. Hence gNB supporting the corresponding normal CA could also meet the existing unwanted emission and ACS requirement when operating in overlapping CA case and no need to redo Base Station conformance testing.
Up

6.7.4  Overlapping UE CBWs from Network Perspectivep. 32

The required guard band is based on the next smaller channel bandwidth for the legacy UE. Hence BS can use the same implementation as overlapping CA, i.e. to use two existing BS filters (other implementations are not excluded). If manufacturer declares to use two existing filter implementation, and gNB supports the corresponding normal CA, the gNB could also meet the existing unwanted emission and ACS requirement and hence no need to redo conformance testing.

6.7.5  Dedicated BS channel bandwidthsp. 32

If the irregular channel bandwidths are explicitly defined in the specification, it will create huge technical specification work. For each channel bandwidth, the guard band size and the transmission bandwidth configuration need to be specified and used as a basis for defining transmitter and receiver requirements. The identified impact to BS core specification for a new channel BW is shown in following Table 6.7.5-1.
Subject Clause in 38.104 Requirement Assessment for new channel BW
General 5.3.2Transmission bandwidth configurationthe Transmission bandwidth configuration NRB for the new CBW need to be defined
5.3.3Minimum guardband and transmission bandwidth configurationThe minimum guardband for the new CBW need to be defined
5.3.5BS channel bandwidth per operating bandnew CBW need defined per band
Transmitter 6.3.3Total power dynamic rangeit is a NRB related requirement
6.6.2Occupied bandwidthBS channel bandwidth should be defined
6.6.3ACLRThe filter are set using transmission bandwidth configuration (BWConfig). It need to be defined for testing
6.7Transmitter intermodulationthe interfering is defined according to BS channel bandwidth
Receiver 7.2Reference sensitivity levelfor 15 KHz SCS, 25 RB and 106 RB FRC which can be reused
7.3Dynamic rangeinterfering signal level is according to BS channel bandwidth
7.4In-band selectivity and blockingthe position of interfering signal is defined according to BS channel bandwidth/transmission bandwidth configuration
7.7Receiver intermodulationthe position of interfering signal is defined according to BS channel bandwidth/transmission bandwidth configuration
7.8In-channel selectivityit is defined according to BS channel bandwidth
It is concluded that new dedicated BS channel bandwidths for irregular channel bandwidths are not defined explicitly in the specification.
Up

6.7.6  Additional enhancementsp. 33

6.7.6.1  Channel raster enhancementsp. 33

All FR1 low-frequency bands have channel raster in steps of 100kHz defined by RAN WG4 as the global raster (5kHz) x 20. RAN WG2 signalling also uses the global raster of 5kHz in principle not preventing a channel of being on any 5kHz channel raster point. While configuring the NR channel on the 100kHz raster is not a big limitation for standard spectrum blocks, such as 5 or 10MHz, it may result in less efficient spectrum utilisation in certain irregular channels. Furthermore, it might be difficult (and sometimes even impossible) to configure the dedicated UE channel bandwidth so that it is aligned on the 100kHz raster and at the same time is the RB aligned with the SIB1 channel bandwidth if the latter is also on the 100kHz raster.
One exemplary benefit of allowing carriers on the non-100kHz raster is presented in Figure 6.7.6.1-1 below for the 7MHz irregular channel. If, for the sake of example, we assume overlapping carriers from the network perspective as potential solution, then the first and the second carrier must have offsets in multiple of 900kHz, whereupon 900kHz is the common multiplier between the 100kHz raster and 180kHz RB size. So, for the 7MHz channel we can shift the second carrier by 1800kHz, i.e., 10RBs; but the resulting guard bands will be noticeably larger than the minimum requirements for the 5MHz carrier. If there is a way to put the carrier on the non-100kHz raster, then the overall system capacity can be increased to 36 RBs still ensuring the minimum guard-band requirements of the 5MHz carrier.
Copy of original 3GPP image for 3GPP TS 38.844, Fig. 6.7.6.1-1: Exemplary usage of overlapping carriers with 100kHz and non-100kHz alignment (7MHz irregular channel)
Up
Thus, a potential enhancement is to extend available channel raster points so that the 100kHz raster alignment is not mandated at least for the dedicated UE channel bandwidth. It can be implemented as a common feature of the whole release or as a UE capability. Nevertheless, irrespective how it is introduced, once the network knows the UE capabilities, it can decide whether the dedicated UE channel bandwidth should stay on the 100kHz raster (for legacy devices) or can be re-configured to a non-100kHz raster channel.
In the example of Figure 6.7.6.1-1, the network can configure two carriers (each with its own SSB, shown at the top and in the middle of Figure 6.7.6.1-1) in accordance with the 100kHz requirement and these carriers will serve legacy devices. And since the network is aware of the UE capabilities, it can always consider configuring another overlapping carrier on the non-100kHz raster after the UE completed the attachment procedure through the legacy carrier shown in the middle row of Figure 6.7.6.1-1. This way the network can handle both legacy and new devices and it will be the network scheduler responsibility to decide how to schedule resources over three overlapping carriers.
Up

6.7.6.2  UE channel larger than SIB1 channel enhancementp. 33

When a UE camps on the cell, it reads the corresponding system information to understand the position of the channel, its size, the initial bandwidth part configuration, etc. Once a UE enters the cell, the network has an ability to re-configure the UE whereupon the network can also provide a new channel configuration that will be specific to that UE. While the most common re-configuration case is to configure a UE with the dedicated channel and/or bandwidth part that is within the SIB1 channel bandwidth, there are also use cases when a UE might be configured with the channel size or position of which is outside the SIB1 configuration. Thus, to enable better utilisation of irregular channels it would be preferrable that UEs can support a re-configuration scenario when the dedicated UE channel is outside the SIB1 boundaries.
Up

Up   Top   ToC