LoRaWAN Communication Debug

1

**~ Table of Contents:**

2

= 1. Join process page check =

8

9

10

These pages are useful to check what is wrong on the Join process. Below shows the four steps that we can check the Join Process.

11

\\**If user has checked below steps and still can't solve the problem, please send us (support @ dragino.com) the sceenshots for each step to check. They include:**

12

13

* End node console to show the Join freuqency and DR. (If possible)

14

15

* Gateway (from gateway UI) traffic to show the packet got from end node and receive from Server. (If possible)

16

17

* Gateway traffic (from server UI) to shows the data exchange between gateway and server. (Normaly possible)

18

19

* End Node traffic (from server UI) to shows end node activity in server. (Normaly possible)

20

21

* End Node Keys screen shot shows in end node and server. so we can check if the keys are correct. (In most case, we found keys doesn't match, especially APP EUI)

22

23

(% style="color:blue" %)**1. End Device Join Screen shot, we can check:**

24

25

* If the device is sending join request to server?

26

27

* What frequency the device is sending?

28

29

[[image:image-20240129142147-2.png||height="736" width="964"]]

30

31

Console Output from End device to see the transmit frequency.

32

33

34

(% style="color:blue" %)**2. Gateway packet traffic in gateway web or ssh. we can check:**

35

36

* If the gateway receive the Join request packet from sensor? (If this fail, check if the gateway and sensor works on the match frequency)

37

38

* If the gateway gets the Join Accept message from server and transmit it via LoRa?

39

40

[[image:image-20240129151608-6.jpeg||height="725" width="1256"]]

41

42

Console Output from Gateway to see packets between end node and server.

43

44

45

(% style="color:blue" %)**3. Gateway Live data in LoRaWAN Server**

46

47

* Does the gateway real-time data contain information about Join Request? If not, check the internet connection and gateway LoRaWAN server Settings.

48

49

* Does the server send back a Join Accept for the Join Request? If not, check that the key from the device matches the key you put into the server, or try to choose a different server route for that end device.

50

51

* If the Join Accept message are in correct frequency? If you set the server to use US915 band, and your end node and gateway is EU868, you will see the Join Accept message are in US915 band so no possible to Join success.

52

53

[[image:image-20240129150821-5.jpeg||height="522" width="1264"]]

54

55

The Traffic for the End node in the server, use TTN as example.

56

57

58

(% style="color:blue" %)**4. Data Page in LoRaWAN server**

59

60

* If this data page shows the Join Request message from the end node? If not, most properly you have wrong settings in the keys. Keys in the server doesn't match the keys in End Node.

61

62

[[image:image-20240129142557-3.png||height="488" width="1267"]]

63

64

The data for the end device set in server

65

66

67

[[image:image-20240129142631-4.png||height="637" width="1256"]]

68

69

Check if OTAA Keys match the keys in device.

70

71

72

= 2. Notice of US915/CN470/AU915 Frequency band =

(((

If user has problem to work with LoRaWAN server in band US915/AU915/CN470, he can check:

)))

* (((

What **sub-band** the server support?

81

)))

82

* (((

83

What is the **sub-band** the gateway support?

84

)))

85

* (((

86

What is the **sub-band** the end node is using?

)))

(((

All of above should match so End Node can properly Join the server and don't have packet lost.

)))

(((

)))

(((

In LoRaWAN protocol, the frequency bands US915, AU915, CN470 each includes at least 72 frequencies. Many gateways support only 8 or 16 frequencies, and server might support 8 frequency only. In this case, the OTAA join time and uplink schedule is long and unpredictable while the end node is hopping in 72 frequencies, because the end node will send data in many frequency that the gateway or server doesn't support.

)))

(((

)))

(((

Here are the frequency tables for these bands as reference:

107

)))

108

109

[[image:image-20220526163801-6.png]]

US915 Channels

[[image:image-20220526163926-7.png]]

AU915 Channels

[[image:image-20220526163941-8.png]]

(((

CN470 Channels

)))

(((

If we look at the [[TTN network server frequency plan>>url:https://www.thethingsnetwork.org/docs/lorawan/frequency-plans.html]], we can see the US915 frequency band use the channel 8~~15.So the End Node must work at the same frequency in US915 8~~15 channels for TTN server.

129

)))

130

131

[[image:image-20240123151225-3.png||height="434" width="902"]]

(((

TTN FREQUENCY PLAN

(% style="display:none" %) (%%)

)))

(((

In dragino end node, user can use AT+CHE command to set what frequencies set the end node will use. The default settings for Dragino end node are preconfigure for TTN server, so use 8~~15 channels, which is **AT+CHE=2**. (AT+CHE=1 for first 8 channels, AT+CHE=2 for second 8 channels.. etc, and AT+CHE=0 for all 72 channels. )

141

)))

142

143

(% style="display:none" %) (%%)

144

145

= 3. Why I see data lost/ is not periodically uplink？ Even the signal strength is good =

146

147

148

In this case, we can check if the frequency band matches in End Node, Gateway and LoRaWAN server. A typical case is using US915 in ChirpStack server as below:

149

150

* (% style="color:blue" %)**End node** (%%) ~-~-> Use Sub-band2 (Channel 8,9,10,11,12,13,14,15) for Dragino Sensor. ADR is also enable, this is the default settings for dragino sensors.

151

152

* (% style="color:blue" %)**Gateway** (%%) ~-~-> Use Sub-band2 (Channel 8,9,10,11,12,13,14,15) for Dragino Gateway. this is the default settings for dragino sensors.

153

154

* (% style="color:blue" %)**LoRaWAN server** (%%) ~-~-> ChirpStack default installation and use Sub-band1, **enabled_uplink_channels=[0, 1, 2, 3, 4, 5, 6, 7]** in the file chirpstack-network-server.toml.

155

156

(((

157

When Sensor power on, it will use sub-band2 to join the network, the frequency matches the settings in gateway so all Join Request will be passed to the server for Join. Server will ask the sensor to change to Sub-band1 in the Join Accept downlink message. Sensor will change to sub-band1 for data upload. This cause the sensor and gateway have different frequencies so user see lost of most data or even no data.

)))

(((

Use Subband2 as a default subband cause the sensor to have problem to work with the LoRaWAN server which use other subband, and use need to access to the end node to change the subband by console. that is not user frendily,. So since Dragino LoRaWAN Stack version DLS-005(release on end of 2020), we have changed the device to use All Subbands for OTAA join, for example, device will use the first frequency in Sub-Band1 as firt OTAA join packet, then use the first frequency in Sub-Band 2 , then first frequency in sub-band 3, and so on. LoRaWAN server will normally provide the required subband in the OTAA accept process, so end node will know what subband it use after join. If LoRaWAN server doesn't provide subband info in OTAA join, end node will use the subband which join success as the working subband. So the new method cause a longer OTAA Join time but will be compatible with all LoRaWAN server. And new method won't affect the normal uplink after Join Success.

)))

= 4. Transmision on ABP Mode =

(((

In ABP mode, there is a Frame Counter Checks. With this check enabled, the server will only accept the frame with a higher counter. If you reboot the device in ABP mode, the device will start from count 0, so you won't be able to see the frame update in server.

)))

(((

So in ABP mode, first check if the packet already arrive your gateway, if the packet arrive gatewat but didn't arrive server. Please check if this is the issue.

)))

(((

To solve this, disable the Frame Counter Check will solve this issue , or reset the frame counter in the device page.

179

180

[[image:image-20240123161737-4.png||height="395" width="763"]]

181

)))

182

183

[[image:image-20240123161853-6.png||height="599" width="771"]]

184

185

Disable Frame Counter Check in ABP Mode

186

187

188

= 5. Downstream Debug =

189

190

== 5.1 How it work ==

191

192

193

LoRaWAN End node will open two receive windows to receive the downstream data. If the downstream packets arrive the end node at these receive windows, the end node will be able to get this packet and process it.

194

195

(((

196

Depends on Class A or Class C, the receive windows will be a little difference.

197

)))

198

199

[[image:image-20220531161828-1.png]]

200

201

receive windows for Class A and Class C

202

203

204

Below are the requirement for the End Device to receive the packets.

205

206

* The End Device must open the receive windows: RX1 or RX2

207

208

* The LoRaWAN server must send a downstream packet, and the gateway forward this downstream packet for this end node.

209

210

* This downstream packet must arrive to the end node while RX1 or RX2 is open.

211

212

* This packet must match the frequency of the RX1 or RX2 window.

213

214

* This packet must match the DataRate of RX1(RX1DR) or RX2 (RX2DR). (% style="color:red" %)**This is the common fail point, because different lorawan server might use different RX2DR and they don't info End Node via ADR message so cause the mismatch. If this happen, user need to change the RX2DR to the right value in end node. In OTAA, LoRaWAN Server will send the RX2DR setting in Join Accept message so the end node will auto adjust. but ABP uplink doesn't support this auto change.**

215

216

== 5.2 See Debug Info ==

(((

(% style="color:blue" %)**For LoRaWAN Server**

)))

(((

We can check if there is downlink message for this end node, use TTN for example:

)))

(((

Configure a downlink to the end device

229

230

[[image:image-20240129152412-8.png||height="486" width="1206"]]

)))

(((

Set a downstream in TTN and see it is sent

236

)))

237

238

(% style="color:red" %)**Note: After the downlink command is successfully sent from the platform to the node, the downlink command is executed only after the platform receives the next uplink package from the node.**

(((

This downlink info will then pass to the gateway downlink list. and the DR which is used (SF7BW500) in US915 is DR5.

243

)))

244

245

[[image:image-20240129152049-7.png||height="463" width="1166"]]

246

247

(((

248

Gateway Traffic can see this downlink info

)))

(((

(% style="color:blue" %)**For LoRaWAN Gateway**

)))

(((

When the downlink packet appear on the traffic of Gateway page. The LoRaWAN gateway can get it from LoRaWAN server and transmit it. In Dragino Gateway, this can be checked by running "logread -f" in the SSH console. and see below:

259

)))

260

261

[[image:image-20240129154321-9.png]]

262

263

(((

264

Gateway Sent out this packet

)))

(((

(% style="color:blue" %)**For End Node**

)))

(((

we can use AT Command (AT+CFG) to check the RX1 configure and RX2 configure. as below:

)))

(((

* (% style="color:#037691" %)**AT+RX2FQ=869525000** (%%) **~-~-->** The RX2 Window frequency

279

280

* (% style="color:#037691" %)**AT+RX2DR=3** (%%) **~-~-->** The RX2 DataRate

281

282

* (% style="color:#037691" %)**AT+RX1DL=1000** (%%) ** ~-~-->** Receive Delay 1

283

284

* (% style="color:#037691" %)**AT+RX2DL=2000** (%%) **~-~--> ** Receive Delay 2

)))

(((

(% style="color:blue" %)**when the device running, we can see below info:**

289

)))

290

291

{{{ [12502]***** UpLinkCounter= 0 *****

292

[12503]TX on freq 868500000 Hz at DR 0

293

[13992]txDone

294

[15022]RX on freq 868500000 Hz at DR 0 --> RX1 window open at frequency: 868500000, DR0, after 15022-13992= 1030ms of txdone

295

[15222]rxTimeOut --> no packet arrive in RX1 window. (duration: 200ms)

296

[15987]RX on freq 869525000 Hz at DR 3 --> RX2 window open at frequency: 869525000, DR3, after 15987-13992= 1995ms of txdone

297

[16027]rxTimeOut --> no packet arrive in RX2 window. (duration: 40 ms)}}}

(((

)))

(((

(% style="color:blue" %)**Another message:**

307

)))

308

309

{{{ [12502]***** UpLinkCounter= 0 *****

310

[12503]TX on freq 868100000 Hz at DR 0

311

[13992]txDone

312

[15022]RX on freq 868100000 Hz at DR 0

313

[15222]rxTimeOut

314

[15987]RX on freq 869525000 Hz at DR 3

315

[16185]rxDone --> We have got the downstream packet.

Rssi= -64

Receive data

1:0012345678}}}

== 5.3 If problem doesn't solve ==

322

323

324

(% style="color:red" %)**If user has checked below steps and still can't solve the problem, please send us (support @ dragino.com) the sceenshots for each step to check. They include:**

325

326

* End node console to show the transmit freuqency and DR.

327

328

* Gateway (from gateway UI) traffic to show the packet got from end node and receive from Server.

329

330

* Gateway traffic (from server UI) to shows the data exchange between gateway and server.

331

332

* End Node traffic (from server UI) to shows end node activity in server.

333

334

= 6. Downlink Issue ~-~- Packet REJECTED, unsupported frequency =

(((

In LoRaWAN, the gateway will use the frequency specify by the server to transmit a packet as downlink purpose. Each Frequency band has different downlink frequency. and the gateway has a frequency range limited to transmit downlink.

)))

(((

)))

(((

So if the LoRaWAN server is an AS923 server which ask the gateway to transmit at 923.2Mhz frequency, but the gateway is IN868 frequency band (support 865~~867Mhz to transmit). In the gateway log it will show something like below:

347

)))

348

349

{{{Sat Nov 21 08:04:17 2020 daemon.info lora_pkt_fwd[1680]: ERROR~ Packet REJECTED, unsupported frequency - 923200000 (min:865000000,max:867000000)}}}

(((

)))

(((

In this case, please double check the gateway frequency and the server frequency band.

)))

= 7. Decrypt a LoRaWAN Packet =

361

362

363

(% style="color:blue" %)**1. LHT65N End device configure:**

364

365

**Change to ABP Mode: AT+NJM=0**

366

367

**Change to fix frequency: AT+CHE=1**

368

369

370

**AT+CFG(Print configuration):**

371

372

[[image:image-20240129170603-7.png||height="697" width="545"]][[image:image-20240129163741-3.png||height="694" width="565"]]

**Configuration: **

[[image:image-20240129164219-4.png||height="612" width="440"]]

(% style="color:blue" %)**2. In LPS8-v2, configure to receive above message**

383

384

[[image:image-20240129164326-5.png||height="506" width="1114"]]

385

386

387

In LPS8-v2 console, we can see the Base64 receive are:

388

389

[[image:image-20240129170137-6.png||height="459" width="1116"]]

(% style="color:blue" %)**3. Decode the info in CMD(Command prompt window)**

394

395

LoRa packet Base64 format: QP~/~/~/~/+AFQACZv8Hjmc8gFTAkhMzU+75 **(from LPS8-v2)**

396

397

Then the instructions and format parsed in SecureCRT are: ./node_modules/.bin/lora-packet-decode ~-~-base64 QP~/~/~/~/+AFQACZv8Hjmc8gFTAkhMzU+75

398

399

400

**Step1: Open CMD, Enter the gateway IP and port.(ssh root@gateway IP -p 22)**

401

402

[[image:image-20240129190752-17.png||height="338" width="901"]]

403

404

[[image:image-20240129191937-21.png||height="450" width="901"]]

405

406

407

**Step2: Enter the command to download the LoRa parsing package.(npm install lora-packet)**

408

409

[[image:image-20240129192239-22.png||height="416" width="902"]]

410

411

[[image:image-20240129192549-23.png||height="459" width="898"]]

412

413

414

**Step3: Parse the gateway raw payload.(./node_modules/.bin/lora-packet-decode ~-~-base64 QP~/~/~/~/+AFQACZv8Hjmc8gFTAkhMzU+75)**

[[image:image-20240129192908-24.png||height="477" width="907"]]

420

421

422

[[image:image-20240129192954-25.png||height="485" width="916"]]

= 8. Why I see uplink 0x00 periodically on the LHT65 v1.8 firmware =

431

432

433

Since firmware v1.8, LHT65 will send MAC command to request time, in the case if DR only support max 11 bytes, this MAC command will be bundled to a separate uplink payload with 0x00.

434

435

436

= 9. Why do I see a "MIC Mismatch" error message from the server? =

(((

1) If the user receives a "MIC Mismatch" message after registering the node on the server.

)))

(((

It is likely that the user filled in the wrong APPKEY when registering the node. Many users fill in "APPSKEY".

)))

* (((

Please note the distinction between "APPKEY" and "APPSKEY".

)))

(((

2）If the node works on the server for a period of time, the device stops working and receives a "MIC Mismatch" message.

)))

(((

The user needs a USB-TTL adapter to connect the serial port to modify the node APPKEY.

)))

* (((

If a node is registered with multiple servers, it may also cause the "mic mismatch" error.

461

)))

462

463

(% class="wikigeneratedid" %)

464

3）Wrong Regional Parameters version selected

465

We generally use versions above 1.0.2

466

467

(% class="wikigeneratedid" %)

468

[[image:image-20230322163227-1.png]]

469

470

(% class="wikigeneratedid" %)

471

4）We have had cases where it was automatically fixed the next day despite no manual changes, probably a server side issue

472

473

474

= 10. Why I got the payload only with "0x00" or "AA~=~="? =

475

476

477

(% style="color:blue" %)**Why sensor sends 0x00?**

478

479

For US915, AU915 or AS923 frequencies, the max payload lenght is 11 bytes for DR0. Some times sensor needs to send MAC command to server, because the payload is 11 bytes, The MAC command + Payload will exceed 11 bytes and LoRaWAN server will ignore the uplink. In this case, Sensor will send two uplinks together: one uplink is the payload without MAC command, another uplink is **0x00 payload + MAC Command.** For the second uplink, in the server side, it will shows the payload is 0x00. Normally, there are several case this will happen.

**Possible Case 1**:

Sensor has ADR=1 enable and sensor need to reply server MAC command (ADR request) while sensor has DR=0.

**Possible Case 2:**

For the sensor which has Datalog Feature enable, the sensor will send TimeRequest MAC Command to sync the time. This Time Request will be sent once Sensor Join Network and Every 10 days. While they send such command with DR=0, sensor will send this command with 0x00 payload.

489

490

491

(% style="color:blue" %)**How to solve:**

Solution:

~1. Use the decoder to filter out this 0x00 packet. (**Recommand**)

496

497

2. Data rate changed from DR3 to DR5, increasing upload byte length

AT+ADR=0

AT+DR=3

Downlink：

[[http:~~/~~/wiki.dragino.com/xwiki/bin/view/Main/End%20Device%20AT%20Commands%20and%20Downlink%20Command/#H7.4DataRate>>http://wiki.dragino.com/xwiki/bin/view/Main/End%20Device%20AT%20Commands%20and%20Downlink%20Command/#H7.4DataRate]]

504

505

Some node decoders may not have the filter function, or you need decoders of other servers and formats. Please send an email to [[support@dragino.com>>mailto:support@dragino.com]]

506

507

508

= 11. Why my Dev EUI and APP EUI is 0x000000000000, how to solve? =

(((

It is possible the keys is erased during upgrading of firmware. and the console output shows below after AT+CFG

)))

(((

AT+APPKEY=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

)))

(((

AT+NWKSKEY=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

)))

(((

AT+APPSKEY=00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

)))

(((

AT+APPEUI=00 00 00 00 00 00 00 00

)))

(((

)))

(((

You can get the keys from the box sticker or send mail to Dragino Support to check keys with the provided SN number.

)))

(((

You can rewrites the keys by running commands in AT Console

)))

(((

**For example：**

)))

(((

AT+APPKEY=85 41 47 20 45 58 28 14 16 82 A0 F0 80 0D DD EE

)))

(((

AT+NWKSKEY=AA CC B0 20 30 45 37 32 14 1E 14 93 E2 3B 20 11

)))

(((

AT+APPSKEY=11 23 02 20 30 20 30 60 80 20 20 30 30 20 10 10

)))

(((

AT+APPEUI=2C 45 47 E3 24 12 23 24

)))

(((

(Any combination of 16 bit codes can be used）

567

568

569

= 12. I set my device is LoRaWAN Class C mode, why I still see Class A after boot? =

)))

Class C only refers to status after OTAA Join successfully. The OTAA Join Process will use Class A mode.

574

575

576

= 13. Why it takes longer time for OTAA joined in US915/CN470/AU915 band? =

577

578

579

In US915, AU915 or CN470 frequency band, there are 8 sub-bands, totally 72 channels. and LoRaWAN server normally use only one sub-band, for example Sub-band 2 in TTN. The gateway also configured to Sub-band 2 and cover eight channels in this sub-band. If the end node transfer data in Sub-band 2, it will reach to gateway and to the LoRaWAN server. If the end node transfer packets in other sub-bands, for example sub-band 1, the packet won't arrive both gateway or LoRaWAN server.

580

581

582

In Dragino Sensors old version firmware (before early 2022), the sub-band is fixed the sub-band to 2 , but this cause a problem, the end node is hard to use in other subband and need program. So the new logic is as below:

583

584

We have improved this, the end node will use frequency 1 from sub-band1, then frequency 1 from sub-band2, then frequency 1 from sub-band3, etc to process the OTAA join, in this case, in this case, the end node can support LoRaWAN servers with different sub-bands. To make sure the end node will only transmit the proper sub-band after OTAA Joined successfully, the end node will:

585

586

* (((

587

Check what sub-band the LoRaWAN server ask from the OTAA Join Accept message and switch to that sub-band.

588

)))

589

* (((

590

Use the Join successful sub-band if the server doesn't include sub-band info in the OTAA Join Accept message (TTN v2 doesn't include).

591

)))

592

593

This change will make the activation time a little longer but make sure the device can be used in any sub-band.

594

595

596

Below is a photo to show why it takes longer time for OTAA Join. We can see in 72 channels mode, why it takes more time to join success. If users want to have faster OTAA Join success, he can change default CHE to the sub-band he uses.

597

598

599

[[image:image-20221215223215-1.png||height="584" width="1280"]]

Wiki source code of LoRaWAN Communication Debug

Applications

Navigation