CRC Failures

[JieyuZhao-nmg]

Dear Developers,

Thank you very much for open-sourcing this excellent application. I am currently trying to use NR-Scope to monitor a real 5G base station signal, but I have encountered some issues.From the logs, I can see a large number of CRC check failures. Occasionally, there is a successful DCI detection (crc=OK and Found DCI), but after that, I don't see any SIB 1 Decoded log entries.

Here's my target base station information

Operator: Sunrise (5G)
Station: VD110-1 CZ1
Type: Outdoor Macro Cell (Aussen)
Frequency: 3750.0 MHz (n77 band)
PCI: 632
Transmission Power: Up to 5000W
Antenna Height: 30 meters
Azimuth: 265°
Adaptive Mode: Yes
Activation Date: February 4, 2025

And this is my environment configuration

Branch: main
Commit: 158ef1419d8d638101aac7e13688086985573486
Hardware: USRP B210 (type=b200)
Band: n77
Subcarrier Spacing: 30kHz (scs_index: 1)

Here is my configuration file:

nof_usrp_dev: 1
usrp_setting_0:
  ssb_freq: 3750240000  # Set to the ssb frequency of the cell, could be obtained with the cell scan function
  rf_args: "type=b200" # For B210
  rx_gain: 80 # for x310 CBX, max rx gain is 31.5, for b210, it's around 80, for x310 TwinRX, max rx gain is 90 
  band_list: [77]
  min_rx_gain: 10
  max_rx_gain: 80
  srate_hz: 23040000 # the sampling rate of USRP, integer division of 200MHz for TwinRX, and 5G sampling rate for CBX: 11520000 or 23040000
  srsran_srate_hz: 23040000 # the sampling for real signal processing, should be in 5G sampling rate, multiple of 1.92 MHz. 
  nof_carriers: 1 # srsRAN rf setting, always set to be 1 for now. 
  nof_antennas: 1 # srsRAN rf setting, always set to be 1 for now.
  scs_index: 1 #(0: 15kHz, 1: 30kHz, ..., the u in standard)
  rf_log_level: "debug"
  nof_rnti_worker_groups: 16 # number of of threads for DCI decoding, each will divide the UEs into small groups
  nof_bwps: 1 # number of BWP of the cell
  nof_workers: 16 # using worker pool to asynchronously process the slot data
  cpu_affinity: false # pin the worker's thread into CPU or not.

### nrscope commit: 158ef1419d8d638101aac7e13688086985573486

And this is my observed behavior

Consistent CRC Failures

The vast majority of PDCCH decode attempts result in crc=KO:

PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=4, al=0, EPRE=+11.83, RSRP=+3.16, corr=0.136; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=5, al=0, EPRE=+13.65, RSRP=+11.85, corr=0.660; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=6, al=0, EPRE=+12.47, RSRP=+9.86, corr=0.549; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=7, al=0, EPRE=+12.51, RSRP=+11.32, corr=0.761; nof_bits=39; crc=KO;

Occasional Successful DCI Detection

Occasionally, a valid DCI is found:

PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=3, EPRE=+12.75, RSRP=+7.15, corr=0.276; nof_bits=39; crc=OK;
SIBDecoder -- Found DCI: si-rnti=0xffff dci=1_0 ss=common0 L=3 cce=0 f_alloc=0x5f t_alloc=0x1 vrb_to_prb_map=0 mcs=1 rv=0 sii=0 coreset0_bw=48 reserved=0x0

PDSCH configuration shows:

PDSCH_cfg:
  DMRS: type=1, add_pos=2, len=single, typeA_pos=2
  Grant: rnti=0xffff, k=0, mapping=A, t_alloc=2:10, f_alloc=0:48
  CW0: mod=QPSK, mcs=1, tbs=1256, R=0.153, rv=0

Despite finding DCI, no successful SIB1 decoding is logged

Note: We have successfully tested this setup with srsRAN and Open5GS using 20 MHz and 40 MHz bandwidth configurations, but not with this real gNB.

[WanHaoRan]

Hello, thanks for testing our tool! In the Consistent CRC Failure, the high EPRE and RSRP could be caused by the DCI for normal communication, instead of the SIB DCI. Is this cell serving many users already? If not, these could be the DCI for the SIBs.

For the Occasional Successful DCI Detection, it seems the DCI is found but the code fails to decode the PDSCH for the SIB1's content. May I ask if you are using the GPSDO on both sides (NR-Scope and the base station)? In our codebase, we rely on the GPSDO for synchronization, because as the bandwidth grows, syncing with the software processing leads to blocking and incontinuous data collection. So using GPSDO is required for the real base station.

[h57jafari]

Hi JieyuZhao-nmg, Could you solve this. I have a same setup and having same problem. While you could decode come DCI, I can not even decode a single PDCCH/DCI. (all of the PDCCH crc=KO, SIBDecoder -- No DCI found). I am using USRP B210 without any GPSDO.

[WanHaoRan]

Hello, seems you have successfully decoded the cell's MIB message. Decoding SIB message has a higher demand on the signal quality. Could you share the signal strength and perhaps EPRE and RSRP values to confirm that the signal strength is acceptable?

Also, the code heavily depends on the GPSDO to keep synchronization, please consider using one. B210 is good at decoding low center frequency cells (600MHz center frequency for example), and X310 generally has better signal quality and will make the decoding easier.

[h57jafari]

Hi, yes I am looking for ~600MHz, n71 band and could decode MIB. I am using only USRP B210 without GPSDO. Please check the following config file and screenshot from sample nrscope result, thank you:

[h57jafari]

and here is the sample nrscope result for 1935.85MHz, n25 band:

[h57jafari]

Hi, I tested nrscope close to multiple gNBs (@b100m-200m), but most of the time the result is as following:

########## it can find SSB and MIB, here is example: ##########
Cell Found!
N_id: 430
Decoding MIB...
MIB: sfn=840 ssb_idx=0 hrf=n scs=15 ssb_offset=4 dmrs_typeA_pos=pos2 coreset0=7 ss0=2 barred=n intra_freq_reselection=y spare=0
Coreset parameter:

• coreset_id=0
• mapping_type=interleaved
• duration=1
• freq_res=11111111000000000000000000000000000000000...
• dmrs_scrambling_present=false (id=0)
• precoder_granularity=reg_bundle
• interleaver_size=2
• reg_bundle_size=6
• shift_index=430
• offset_rb=0
  After calling coreset_zero_t_f_nrscope
  [AGC gain adj] new rx gain: 0.000000
  current value: 9999
  [AGC test] trigger process
  producer time_spend: 743(us)
  current value: 9998

########## but most of the time no valid values for EPRE, RSRP, and correlation is zero: ##########
SIBDecoder -- No DCI found :'(
[AGC test] trigger process
producer time_spend: 1087(us)
current value: 1
consumer time_spend: 53(us)
Processing sf_round: 2, sfn: 899, slot.idx: 5
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=1, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=2, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=3, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=4, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=5, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=6, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=7, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=1, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=2, al=1, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=4, al=1, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=6, al=1, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=2, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=4, al=2, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=3, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;

########## in some case even they have value, correlation is too low as following: ##########
SIBDecoder -- No DCI found :'(
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=0, EPRE=+25.12, RSRP=+16.46, corr=0.136; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=1, al=0, EPRE=+33.10, RSRP=+25.09, corr=0.158; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=2, al=0, EPRE=+31.39, RSRP=+23.53, corr=0.164; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=3, al=0, EPRE=+33.32, RSRP=+18.53, corr=0.033; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=4, al=0, EPRE=+30.79, RSRP=+19.69, corr=0.078; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=5, al=0, EPRE=+31.95, RSRP=+17.58, corr=0.037; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=6, al=0, EPRE=+31.62, RSRP=+21.28, corr=0.093; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=7, al=0, EPRE=+26.27, RSRP=+11.77, corr=0.036; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=1, EPRE=+30.73, RSRP=+19.51, corr=0.075; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=2, al=1, EPRE=+32.46, RSRP=+20.38, corr=0.062; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=4, al=1, EPRE=+31.41, RSRP=+22.14, corr=0.118; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=6, al=1, EPRE=+29.72, RSRP=+17.06, corr=0.054; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=2, EPRE=+31.68, RSRP=+0.03, corr=0.001; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=4, al=2, EPRE=+30.64, RSRP=+4.85, corr=0.003; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=3, EPRE=+31.19, RSRP=+16.94, corr=0.038; nof_bits=39; crc=KO;

Any suggestion? Thank you.

[WanHaoRan]

Hey @h57jafari , are you using a GPSDO for your measurement? That's quite curial. Also, could you share your config file with us so that we can check?

[h57jafari]

I am not using GPSDO, only USRP B210.

Here is the config file:

nof_usrp_dev: 1
usrp_setting_0:
ssb_freq: 627650000 # Set to the ssb frequency of t>
#band_list: [71] #[1,2,3,78] # Scan specific bands with the cell scan function. Will do full scan if left commented.
pci: 549 # Specify the desired Physical cell id. If specified, NR>
rf_args: "type=b200" # "clock=external,type=b200" For B210
rx_gain: 70 # 75 # for x310 CBX, max rx gain is 31.5, for b210, it's around 80, for x310 TwinRX, max rx gain is 90
min_rx_gain: 0 # 0
max_rx_gain: 75 #80
srate_hz: 15360000 #23040000 # the sampling rate of USRP, integer division of 200MHz for TwinRX, and 5G sampli>
srsran_srate_hz: 15360000 #23040000 # the sampling for real signal processing, should be in 5G sampling rate, >
nof_carriers: 1 # srsRAN rf setting, always set to be 1 for now.
nof_antennas: 1 # srsRAN rf setting, always set to be 1 for now.
scs_index: 0 #(0: 15kHz, 1: 30kHz, ..., the u in standard)
rf_log_level: "debug" #"info" #"debug"
nof_rnti_worker_groups: 4 #4 # number of of threads for DCI decoding, each will divide the UEs into small groups
nof_bwps: 1 # number of BWP of the cell
nof_workers: 4 # 4 # using worker pool to asynchronously process the slot data
cpu_affinity: false # pin the worker's thread into CPU or not.

[WanHaoRan]

Hello, GPSDO is required to make the code work with the live cells. If you don't have one, you can try the sc2430 branch, which will try decode SSB periodically and print out the frequency offset information, like:
SSBTracker Found SSB: sf_idx=0, sfn=776, ssb_idx=1, cfo=103.890 Hz, delay_us=423.521, ssb_sfn_parity=1
But this can't replace a GPSDO, especially when the cell's bandwidth increases. Please use a GPSDO for reliable DCI decoding.

[WanHaoRan]

Without the GPSDO, the device's sampling frequency will drift and synchronization can't be maintained. Ideally, the code itself should pick up the SSBs and stays synchronized with the cell, but there are two problems:

1. the SSB is not always properly decoded due to the signal quality, thus the code will have to try many times to recover synchronization, causing data lost during the retries, as no DCI can be decoded without synchronization.
2. syncing with SSB requires the global buffer sample shifts, which is a sequential process before the multi-thread consumers. Doing so blocks the multi-thread consumers and causes overflow in the RF frontend, specially when the bandwidth is high as 100MHz.

With the GPSDO, the code only needs to correctly decode the SSB once and use the GPSDO reference signal remain synchronization with the cell, no data will be lost.

[h57jafari]

Hi,

I connected GPSDO to my USRP B210 (10MHz to ref in), also I went closer to gNBs site (100m to 200m). But still I can not decode any DCI or SIB.
I tried two bands of 627.65MHz and 1935.85MHz (at 23.04Msps). Following show result for a gNB on two different bands (where both captured in the same location, distance from gNB)
any suggestion?

##########################################################################################################################
#here is some output for 627.65MHz:
##########################################################################################################################

SIBDecoder -- No DCI found :'(
consumer time_spend: 280(us)
Processing sf_round: 0, sfn: 869, slot.idx: 1
producer time_spend: 639(us)
current value: 2
consumer time_spend: 237(us)
Processing sf_round: 0, sfn: 869, slot.idx: 2
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=0, EPRE=+12.20, RSRP=+5.63, corr=0.220; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=1, al=0, EPRE=+32.61, RSRP=+25.34, corr=0.188; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=2, al=0, EPRE=+16.14, RSRP=+4.79, corr=0.073; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=3, al=0, EPRE=+13.45, RSRP=-0.95, corr=0.036; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=4, al=0, EPRE=+31.43, RSRP=+21.19, corr=0.095; nof_bits=39; crc=KO;
SIBDecoder -- No DCI found :'(
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=1, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=2, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=3, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=4, al=0, EPRE=-inf, RSRP=-inf, corr=0.000; nof_bits=39; crc=KO;

##########################################################################################################################
#here is my config file:
##########################################################################################################################

nof_usrp_dev: 1
usrp_setting_0:
ssb_freq: 627650000 #1935850000 #627650000 #622850000 # Set to the ssb frequency of the cel>
pci: 877 #494 for 1.93585GHz #877 for 627.65MHz # pci: 339 # Specify the desired Physical c>
rf_args: "clock=external,type=b200,num_recv_frames=1024" # "clock=external,type=b200" For B>
rx_gain: 70 # 75 # for x310 CBX, max rx gain is 31.5, for b210, it's around 80, for x310 Tw>
min_rx_gain: 0
max_rx_gain: 75 #80
srate_hz: 23040000 #15360000 # the sampling rate of USRP, integer division of 200MHz for Tw>
srsran_srate_hz: 23040000 #15360000 # the sampling for real signal processing, should be in>
nof_carriers: 1 # srsRAN rf setting, always set to be 1 for now.
nof_antennas: 1 # srsRAN rf setting, always set to be 1 for now.
scs_index: 0 #(0: 15kHz, 1: 30kHz, ..., the u in standard)
rf_log_level: "debug"
nof_rnti_worker_groups: 16 # number of of threads for DCI decoding, each will divide the UE>
nof_bwps: 1 # number of BWP of the cell
nof_workers: 16 # 4 # using worker pool to asynchronously process the slot data
cpu_affinity: false # pin the worker's thread into CPU or not.

log_name: "PCI_549_1536.csv"

##########################################################################################################################
#here is the sample output for 1935.85MHz:
##########################################################################################################################
SIBDecoder -- No DCI found :'(
consumer time_spend: 226(us)
Processing sf_round: 1, sfn: 210, slot.idx: 3
producer time_spend: 1052(us)
current value: 1
consumer time_spend: 199(us)
Processing sf_round: 1, sfn: 210, slot.idx: 4
[AGC test] trigger process
producer time_spend: 1011(us)
current value: 1
consumer time_spend: 199(us)
Processing sf_round: 1, sfn: 210, slot.idx: 5
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=0, EPRE=+nan, RSRP=+nan, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=1, al=0, EPRE=+nan, RSRP=+nan, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=2, al=0, EPRE=+nan, RSRP=+nan, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=3, al=0, EPRE=+nan, RSRP=+nan, corr=0.000; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=4, al=0, EPRE=+nan, RSRP=+nan, corr=0.000; nof_bits=39; crc=KO;
SIBDecoder -- No DCI found :'(
producer time_spend: 929(us)
current value: 1
consumer time_spend: 206(us)
Processing sf_round: 1, sfn: 210, slot.idx: 6
producer time_spend: 971(us)
current value: 1
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=0, al=0, EPRE=-4.23, RSRP=-24.45, corr=0.009; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=1, al=0, EPRE=-1.47, RSRP=-16.83, corr=0.029; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=2, al=0, EPRE=-1.51, RSRP=-9.04, corr=0.177; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=3, al=0, EPRE=-3.09, RSRP=-11.59, corr=0.141; nof_bits=39; crc=KO;
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=4, al=0, EPRE=-7.47, RSRP=-18.54, corr=0.078; nof_bits=39; crc=KO;

##########################################################################################################################
#also following are the sample spectrogram of those two bands. the top one is for 1935.85MHz and bottom one for 627.65MHz (both with 23.04Msps).
##########################################################################################################################

[WanHaoRan]

Hello @h57jafari , thanks for your information and the updated setup. I can clearly see the MIB/SSB block from your spectrograms at the center frequency, and your log indicates that you have decoded them successfully.

Your log snapshots show high signal strength, but low PDCCH DMRS correlation (corr=0.177). Could you search through your log file to find out what's the highest correlation number? The code is configured to process with a DMRS correlation result 0.05, but generally higher correlation yields better decoding chances. Can you see a decoded DCI for SIB 1 (with RNTI 0xffff) but the code fails to decode its content (PDSCH)? Or the DCI for SIB 1 itself is not decoded?

[h57jafari]

Hi, the max corr in my logs are some times around 0.3 (but not go higher like 0.4 or 0.5):
PDCCH: si-rnti=0xffff, crst_id=0, ss_type=common0, ncce=3, al=0, EPRE=-15.65, RSRP=-20.76, corr=0.308; nof_bits=39; crc=KO;

[WanHaoRan]

I see. Then original correlation threshold in srsRAN UE code is 0.5, we lowered it but 0.3 still seems to be too low.
Could you fix the gain to the maximum by setting both the min and max rx gain to 75? Also, no need to use a high number for nof_rnti_worker_groups: 16, you can use nof_rnti_worker_groups: 1. Is the antenna the same one used for generating the spectrum graph?

[h57jafari]

Thank you Wan, I set the rx_gain, min_rx_gain, and max_rx_gain to the same values like 75, 70, 65, 60, 55, ... but it did not help ( I also set the nof_rnti_worker_groups: 1). Yes, antenna is the same for running nrscope and capturing RF for spectrum graph.