CE Adaptivity tests support for 88W8887

[harshitshah]

Hi NXP team,

We are using the 88W8887 module in our product. We are using the "SD-UAPSTA-BT-8887-U16-MMC-W15.68.19.p38-15.26.19.p38-C4X15659_A2-MGPL" version and we have used the following scripts for AP and station for the CE Adaptivity tests.

We have got the following errors with the below comments from the test lab.

1) The payload duty cycle we are getting is less than 30% and we are not able to adjust the duty cycle. Actual results = 7.35%.

Expected result > 33% and it should be adjustable up to 100% to perform the adaptivity test. (See attached image-1)

2) Module is not stopping the signal transmission within maximum channel occupancy time during the interference signal insertion in the same channel.

Expected: The module has an adaptivity function and it should stop signal transmission when an interference signal is added to the channel. (See attached image-2.1, image-2.2)

3)  Short signaling transmission time is ~12 msec.

Expected: It should be less than 5mSec.

Setup

We have prepared the following command scripts for setting up the AP and station modes for the adaptive tests.

# sh wifi_scripts/2.412GHz_b_20MHZ/ap_2.412GHz_b_20MHZ.sh 1

# sh wifi_scripts/2.412GHz_b_20MHz/station_2.412GHz_b_20MHZ.sh 1

Queries:

1. Please suggest what can be the cause of the above issues and what we are missing here?

2. Please mention what should be the cause of the issue as mentioned in the image 2-1 and 2-2? 

3. Please share the scripts for adaptive and receiver blocking for the tests that is performed for this chipset.

Regards.

[riteshpatel]

1) Lab response "FAIL DATA is at 2412 b mode, the package can't make it 30%, so we stop here."

The lab has mentioned in the report that the duty cycle of the payload must be >30%. I believe, payload means the packet transmitted. 

3) FAIL DATA is at 2412 b mode.

- The Short control signalling time should be <5mSec. 

- Attached image showing in our case we are getting 11.75mSec

[nxf65420]

Hi @harshitshah,

For the 2nd test case failure, can you please provide us the EDMAC related configuration done and steps followed for the test case?

Regards,

Karan Gajjar

[riteshpatel]

@nxf65420 

The test procedure and test results are confidential and shared under NDA. So we are not able to share it here.

Thanks. 

[nxf65420]

Hi @riteshpatel,

Sure. However, you can provide us the details regarding : "the EDMAC related configuration and threshold value set"?

Regards,

Karan Gajjar

[harshitshah]

Hi @nxf65420 ,

We have used the same file for edmac configuration, please find this file for the same.

#       File : ed_mac_ctrl_V2_8887.conf
#
#   ed_mac_ctrl_v2 is used for 88W8897, 88W8887, 88W8797, 88W8782, 88W8787
#   ./mlanutl mlan0 hostcmd config/ed_mac_ctrl_V2_8887.conf ed_mac_ctrl_v2
#
## Set Energy Detect Threshold for EU Adaptivity test

ed_mac_ctrl_v2={
        CmdCode=0x0130                  #Command code, DO NOT change this line
        ed_ctrl_2g.enable:2=0x1	        # 0       - disable EU adaptivity for 2.4GHz band
                                        # 1       - enable EU adaptivity for 2.4GHz band

        ed_ctrl_2g.offset:2=0x0         # 0       - Default Energy Detect threshold
                                        #offset value range: 0x80 to 0x7F

        ed_ctrl_2g.on_period:2=0x12     #DO NOT Change this line
        ed_ctrl_2g.off_period:2=0x0     #DO NOT Change this line
        ed_ctrl_2g.bitmap:2=0x1         #DO NOT Change this line

        ed_ctrl_5g.enable:2=0x1         # 0       - disable EU adaptivity for 5GHz band
                                        # 1       - enable EU adaptivity for 5GHz band

        ed_ctrl_5g.offset:2=0x6         # 0       - Default Energy Detect threshold
                                        #offset value range: 0x80 to 0x7F

        ed_ctrl_5g.on_period:2=0x12     #DO NOT Change this line
        ed_ctrl_5g.off_period:2=0x0     #DO NOT Change this line
        ed_ctrl_5g.bitmap:2=0x1         #DO NOT Change this line
}

[nxf65420]

Hi @harshitshah, @riteshpatel,

We checked the EDMAC conf file provided. Please find below observations:

• The value of offset for 2g is 0x08, in default case, where this test is passed. Your offset value is 0x0, any specific reason for keeping this value to 0x0.
• The value of offset for 5g us 0x0c, in default case, where this test is passed. Your offset value is 0x06, any specific reason for keeping this value to 0x06.

Can you please check with above value? And please share the observation with the same.

• Can you please let us know if DUT is operating in 2.4GHz or 5 GHz?
• Also, please check with the latest release of 88W8887?

Can you please let us know the below details for the test setup?

• Can you please let us know the test scenario that you are using for this test? At which point are you seeing the Adaptivity test fail? Are you continuously transmitting the interfering signal?
• For Duty cycle failure for DUT, can you please confirm if the payload that you referring to is the payload of the data transmitted during iperf. If that is the case, can you check with iperf -c -i 1 -u -b 100M -t 1000? And share the result.

Regards,

Karan Gajjar

[riteshpatel]

@nxf65420 Thank you for your support.

We visited test lab today and for the 5GHz frequencies adaptivity is failing.
We checked with NXP documented ed_mac values 0×0C and 0×55, however with both these values adaptivity fails. (With 0×55 all the results are failing.) So, the 7F value for the ed_mac that will fail as per the observation.

For 2.4GHz the adaptivity test is passing (Threshold 0×08 - NXP default).

From the observations, It looks like ed_mac might not fix the issue for the 5G adaptivity failure. Please suggest the further plan for the same.

[harshitshah]

Hi @nxf65420 ,

Please find the complete pre-compliance report for the CE adaptivity failure. Please look into the logs from page 15 for the 5G CE adaptivity tests.

[nxf65420]

Hi @harshitshah,

Can you please help us with the dmesg logs with drvdbg=0x20037 and sniffer capture for the 5G failed test case?

Also please share the setup details, steps used and the configuration file for the failed case.

Regards,

Karan Gajjar

[harshitshah]

Hi @nxf65420 ,

We might not add more debug prints currently, as we do not have that setup available to test in-house. We were using the external lab for the same.

Have you got the chance to perform the setup at your end with the same version that we are using?

Regards.

[nxf65420]

Hi @harshitshah,

From the test report shared we can see that MAX COT and Idle period probability cases are failing. And MAX COT and idle period probability are in Option-A which is optional in V2.1.1 of the standard.

Regards,

Karan Gajjar

[riteshpatel]

@nxf65420 

We are not able to find those requirements as an optional in the standard.

Can you please let us know where it is mentioned in the standard?

We are referring to EN 301 893 v2.1.1.

[nxf65420]

Hello @harshitshah,

We are analyzing your query and would require your inputs to better understand your query.

Can you please help us explain your below test case failure?

1) The payload duty cycle we are getting is less than 30% and we are not able to adjust the duty cycle. Actual results = 7.35%.

Expected result > 33% and it should be adjustable up to 100% to perform the adaptivity test. (See attached image-1)

[Karan]Which payload duty cycle are you referring to? Can you let us know the test case here?

3) Short signaling transmission time is ~12 msec.

[Karan]Can you please explain the test case here and short signaling transmission time?

3. Please share the scripts for adaptive and receiver blocking for the tests that is performed for this chipset.

[Karan]Can you please let us know which scripts are you referring to? And can you please explain the adaptive and receiver blocking? Is it referring to any test case?

Regards,

Karan Gajjar

[riteshpatel]

Hi Karan,

Can you please let us know how can we pass the maxCOT and idle period probability Test cases with option B?

Appreciate your earliest response on this.

Regards,

Ritesh Patel.