sys/net: add some definitions for loramac by aabadie · Pull Request #7592 · RIOT-OS/RIOT

aabadie · 2017-09-11T12:52:35Z

This PR adds required definitions and constants used for using LoRaMAC. It's mainly based on what is required for #7505 but I tried to keep them as generic as possible by checking the specs of the MAC layer of LoRaWAN.

There are probably missing definitions but those can be added later if needed.

miri64

Minor thing. Will give it a proper review tomorrow.

miri64 · 2017-09-11T15:25:38Z

sys/include/net/loramac.h

+#endif
+
+#endif /* NET_LORAMAC_H */
+/** @} */


Missing newline or something.

fixed (missing newline)

jia200x · 2017-09-13T14:46:43Z

Looks good to me (at least for class A devices ;)

aabadie · 2017-09-19T13:45:02Z

@miri64, any chance for you to look into this one ?

miri64 · 2017-10-10T17:19:31Z

With gnrc_netdev close to being deprecated, maybe this can be ported to gnrc_netif2 directly?

miri64 · 2017-10-10T17:21:05Z

Sorry, I thought this PR introduces gnrc_netdev_loramac_init(). Where is that?

aabadie · 2017-10-11T15:49:20Z

I thought this PR introduces gnrc_netdev_loramac_init(). Where is that?

Well, they were supposed to be added in #7505, but I just realized that the function definitions are missing or use a wrong name (gnrc_loramac_recv, gnrc_loramac_send, gnrc_netdev_loramac_init).
Do you have any advice ?

aabadie · 2017-10-11T16:15:18Z

@miri64, I moved the gnrc_netdev_loramac_init function from #7505 to here. The implementation is here as well. It's totally not tested.

dylad · 2017-10-13T19:41:27Z

Just a question here completely unrelated :
Why is there an implementation of loramac within GNRC and a LMIC package ? What's the point ?

aabadie · 2017-10-16T11:58:41Z

Why is there an implementation of loramac within GNRC and a LMIC package ? What's the point ?

The idea will be at some point to call LMIC API (or another one) via gnrc_netdev_loramac_init. A single API for working with different underlying implementations. But I have not yet a clear vision right now myself on what it should look like in the end.

miri64 · 2017-11-07T20:33:18Z

Removing Hack'n'ACK label since this contains gnrc(_netdev) related stuff.

aabadie · 2017-12-14T09:00:13Z

@miri64, I removed the netdev related commit from this one. So now this PR only contains LoRaMAC useful definitions.
Used in #7505 (needs adaption though) and the coming Semtech LoRaMAC PR.

miri64

Apart from this (purely opinion-based) value and naming comments, I'm fine with this PR. After a short discussion, I think we can merged.

miri64 · 2017-12-18T16:23:51Z

sys/include/net/loramac.h

+ */
+#ifndef LORAMAC_DEV_EUI_DEFAULT
+#define LORAMAC_DEV_EUI_DEFAULT        { 0x00, 0x00, 0x00, 0x00, \
+                                         0x00, 0x00, 0x00, 0x00 }


Not sure if this is also true for LoRA, but from experience with IEEE 802.15.4 a random EUI might be better than an all 0 one. Of course, this should be in accordance with the standard, so if certain bits have semantics, they should be set accordingly.

Leaving 0 values will prevent the node from the joining a LoRaWAN network in any case. That's why by default they use those arrays. For example, the RN2483 module (with the stack already in it) has those values by default.

The arrays are provided by the LoRaWAN provider when one register a new node and they depend on the type of join procedure (either ABP or OTAA).

I can verify what the LoRaWAN specs mention about that to be sure.

Sounds convincing enough to me :-)

miri64 · 2017-12-18T16:24:12Z

sys/include/net/loramac.h

+#define LORAMAC_APP_KEY_DEFAULT        { 0x00, 0x00, 0x00, 0x00, \
+                                         0x00, 0x00, 0x00, 0x00, \
+                                         0x00, 0x00, 0x00, 0x00, \
+                                         0x00, 0x00, 0x00, 0x00 }


Those keys should definitely not be 0....

Since this is something that is set by the user retrieved from the LoRaWAN backend, I think these list can be kept like this by default.
The other problem is that some keys are not needed depending on the activation mode (ABP: requires all keys, OTAA only requires dev eui, app eui and app key).

Are you convinced ?

miri64 · 2017-12-18T16:24:35Z

sys/include/net/loramac.h

+ * @brief   Default device address
+ */
+#ifndef LORAMAC_DEV_ADDR_DEFAULT
+#define LORAMAC_DEV_ADDR_DEFAULT       { 0x00, 0x00, 0x00, 0x00 }


Same applies here as for the EUI-64.

miri64 · 2017-12-18T16:28:53Z

sys/include/net/loramac.h

+ * @brief   Default NetID (only valid with ABP join procedure)
+ */
+#ifndef LORAMAC_DEFAULT_NETWORK_ID
+#define LORAMAC_DEFAULT_NETWORK_ID      (1U)


In the doc you name it NetID. Why not in the name (would have the benefit of being much shorter)

I'll change that

miri64

ACK

aabadie added the Area: network Area: Networking label Sep 11, 2017

aabadie added this to the Release 2017.10 milestone Sep 11, 2017

aabadie assigned miri64 and jia200x Sep 11, 2017

miri64 requested changes Sep 11, 2017

View reviewed changes

aabadie mentioned this pull request Sep 11, 2017

drivers/rn2xx3: basic initial implementation (without netdev) #7505

Merged

aabadie force-pushed the loramac_definitions branch 2 times, most recently from 6fa3716 to a397db3 Compare September 12, 2017 10:26

aabadie force-pushed the loramac_definitions branch 2 times, most recently from 423b0c2 to 08cc456 Compare September 19, 2017 10:43

aabadie force-pushed the loramac_definitions branch 2 times, most recently from 41eb188 to ebdba7c Compare October 8, 2017 10:50

aabadie force-pushed the loramac_definitions branch from ebdba7c to f833e29 Compare October 11, 2017 05:38

aabadie force-pushed the loramac_definitions branch 2 times, most recently from 8b6072e to de19479 Compare October 11, 2017 16:10

aabadie added the Area: LoRa Area: LoRa radio support label Oct 13, 2017

aabadie force-pushed the loramac_definitions branch 2 times, most recently from 390c31c to 1dc9bf3 Compare October 13, 2017 09:18

kYc0o mentioned this pull request Nov 6, 2017

Low-level Hardware Support #7888

Closed

23 tasks

aabadie added Community: Hack'n'ACK candidate This PR is a candidate for review and discussion during one of RIOT's monthly Hack'n'ACK parties CI: ready for build If set, CI server will compile all applications for all available boards for the labeled PR labels Nov 7, 2017

miri64 removed the Community: Hack'n'ACK candidate This PR is a candidate for review and discussion during one of RIOT's monthly Hack'n'ACK parties label Nov 7, 2017

smlng modified the milestones: Release 2017.10, Release 2018.01 Nov 16, 2017

aabadie force-pushed the loramac_definitions branch from 1dc9bf3 to 3617499 Compare December 14, 2017 08:56

aabadie mentioned this pull request Dec 14, 2017

pkg: Semtech LoRaMAC #8264

Merged

aabadie force-pushed the loramac_definitions branch from 3617499 to bdc25f4 Compare December 14, 2017 16:18

miri64 reviewed Dec 18, 2017

View reviewed changes

sys/net: add some definitions for loramac

0512ac5

aabadie force-pushed the loramac_definitions branch from bdc25f4 to 0512ac5 Compare December 18, 2017 17:26

miri64 approved these changes Dec 18, 2017

View reviewed changes

miri64 merged commit 1601341 into RIOT-OS:master Dec 18, 2017

aabadie deleted the loramac_definitions branch June 8, 2018 20:49

Conversation

aabadie commented Sep 11, 2017

Uh oh!

miri64 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jia200x commented Sep 13, 2017

Uh oh!

aabadie commented Sep 19, 2017

Uh oh!

miri64 commented Oct 10, 2017

Uh oh!

miri64 commented Oct 10, 2017

Uh oh!

aabadie commented Oct 11, 2017

Uh oh!

aabadie commented Oct 11, 2017

Uh oh!

dylad commented Oct 13, 2017

Uh oh!

aabadie commented Oct 16, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

miri64 commented Nov 7, 2017

Uh oh!

aabadie commented Dec 14, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

miri64 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

miri64 left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

aabadie commented Oct 16, 2017 •

edited

Loading

aabadie commented Dec 14, 2017 •

edited

Loading