- OpenBSC - lists.osmocom.org

[PATCH] use osmocom auth API instead of direct calls
by Max Suraev 22 Jul '14

22 Jul '14

--- openbsc/src/libmsc/auth.c | 68 +++++++++++++++++++---------------------------- 1 file changed, 28 insertions(+), 40 deletions(-) diff --git a/openbsc/src/libmsc/auth.c b/openbsc/src/libmsc/auth.c index 10d8edf..ab84975 100644 --- a/openbsc/src/libmsc/auth.c +++ b/openbsc/src/libmsc/auth.c @@ -25,46 +25,11 @@ #include <openbsc/auth.h> #include <openbsc/gsm_data.h> -#include <osmocom/gsm/comp128.h> +#include <osmocom/crypt/auth.h> #include <stdlib.h> -static int -_use_xor(struct gsm_auth_info *ainfo, struct gsm_auth_tuple *atuple) -{ - int i, l = ainfo->a3a8_ki_len; - - if ((l > A38_XOR_MAX_KEY_LEN) || (l < A38_XOR_MIN_KEY_LEN)) { - LOGP(DMM, LOGL_ERROR, "Invalid XOR key (len=%d) %s\n", - ainfo->a3a8_ki_len, - osmo_hexdump(ainfo->a3a8_ki, ainfo->a3a8_ki_len)); - return -1; - } - - for (i=0; i<4; i++) - atuple->sres[i] = atuple->rand[i] ^ ainfo->a3a8_ki[i]; - for (i=4; i<12; i++) - atuple->kc[i-4] = atuple->rand[i] ^ ainfo->a3a8_ki[i]; - - return 0; -} - -static int -_use_comp128_v1(struct gsm_auth_info *ainfo, struct gsm_auth_tuple *atuple) -{ - if (ainfo->a3a8_ki_len != A38_COMP128_KEY_LEN) { - LOGP(DMM, LOGL_ERROR, "Invalid COMP128v1 key (len=%d) %s\n", - ainfo->a3a8_ki_len, - osmo_hexdump(ainfo->a3a8_ki, ainfo->a3a8_ki_len)); - return -1; - } - - comp128(ainfo->a3a8_ki, atuple->rand, atuple->sres, atuple->kc); - - return 0; -} - /* Return values * -1 -> Internal error * 0 -> Not available @@ -76,6 +41,11 @@ int auth_get_tuple_for_subscr(struct gsm_auth_tuple *atuple, { struct gsm_auth_info ainfo; int i, rc; + static struct osmo_sub_auth_data auth = { + .type = OSMO_AUTH_TYPE_GSM + }; + struct osmo_auth_vector _vec; + struct osmo_auth_vector *vec = &_vec; /* Get subscriber info (if any) */ rc = db_get_authinfo_for_subscr(&ainfo, subscr); @@ -109,13 +79,23 @@ int auth_get_tuple_for_subscr(struct gsm_auth_tuple *atuple, return 0; case AUTH_ALGO_XOR: - if (_use_xor(&ainfo, atuple)) - return 0; + auth.algo = OSMO_AUTH_ALG_XOR; + if ((ainfo.a3a8_ki_len > A38_XOR_MAX_KEY_LEN) || (ainfo.a3a8_ki_len < A38_XOR_MIN_KEY_LEN)) { + LOGP(DMM, LOGL_ERROR, "Invalid XOR key (len=%d) %s\n", + ainfo.a3a8_ki_len, + osmo_hexdump(ainfo.a3a8_ki, ainfo.a3a8_ki_len)); + return -1; + } break; case AUTH_ALGO_COMP128v1: - if (_use_comp128_v1(&ainfo, atuple)) - return 0; + auth.algo = OSMO_AUTH_ALG_COMP128v1; + if (ainfo.a3a8_ki_len != A38_COMP128_KEY_LEN) { + LOGP(DMM, LOGL_ERROR, "Invalid COMP128v1 key (len=%d) %s\n", + ainfo.a3a8_ki_len, + osmo_hexdump(ainfo.a3a8_ki, ainfo.a3a8_ki_len)); + return -1; + } break; default: @@ -124,6 +104,14 @@ int auth_get_tuple_for_subscr(struct gsm_auth_tuple *atuple, return 0; } + memcpy(auth.u.gsm.ki, ainfo.a3a8_ki, sizeof(auth.u.gsm.ki)); + + if (osmo_auth_gen_vec(vec, &auth, atuple->rand) < 0) + return -1; + + memcpy(atuple->sres, vec->sres, 4); + memcpy(atuple->kc, vec->kc, 8); + db_sync_lastauthtuple_for_subscr(atuple, subscr); DEBUGP(DMM, "Need to do authentication and ciphering\n"); -- 1.9.1

3 3

[osmo-pcu v2 1/1] bts: Return the special type for {ul, dl}_tbf_by_* functions
by Daniel Willmann 16 Jul '14

16 Jul '14

Start returning the special type instead of the base gprs_rlcmac_tbf when retrieving a TBF. Ticket: SYS#389 Sponsored-by: On-Waves ehf --- src/bts.cpp | 36 ++++++++++++++++++------------------ src/bts.h | 16 ++++++++-------- 2 files changed, 26 insertions(+), 26 deletions(-) diff --git a/src/bts.cpp b/src/bts.cpp index f614dab..79cae4c 100644 --- a/src/bts.cpp +++ b/src/bts.cpp @@ -217,15 +217,15 @@ int BTS::add_paging(uint8_t chan_needed, uint8_t *identity_lv) } /* search for active downlink tbf */ -gprs_rlcmac_tbf *BTS::dl_tbf_by_tlli(uint32_t tlli) +gprs_rlcmac_dl_tbf *BTS::dl_tbf_by_tlli(uint32_t tlli) { - return tbf_by_tlli(tlli, GPRS_RLCMAC_DL_TBF); + return static_cast<gprs_rlcmac_dl_tbf *>(tbf_by_tlli(tlli, GPRS_RLCMAC_DL_TBF)); } /* search for active uplink tbf */ -gprs_rlcmac_tbf *BTS::ul_tbf_by_tlli(uint32_t tlli) +gprs_rlcmac_ul_tbf *BTS::ul_tbf_by_tlli(uint32_t tlli) { - return tbf_by_tlli(tlli, GPRS_RLCMAC_UL_TBF); + return static_cast<gprs_rlcmac_ul_tbf *>(tbf_by_tlli(tlli, GPRS_RLCMAC_UL_TBF)); } /* search for active downlink or uplink tbf */ @@ -250,9 +250,9 @@ gprs_rlcmac_tbf *BTS::tbf_by_tlli(uint32_t tlli, enum gprs_rlcmac_tbf_direction return NULL; } -gprs_rlcmac_tbf *BTS::dl_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts) +gprs_rlcmac_dl_tbf *BTS::dl_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts) { - struct gprs_rlcmac_tbf *tbf; + struct gprs_rlcmac_dl_tbf *tbf; /* only one TBF can poll on specific TS/FN, because scheduler can only * schedule one downlink control block (with polling) at a FN per TS */ @@ -262,14 +262,14 @@ gprs_rlcmac_tbf *BTS::dl_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts) && tbf->poll_fn == fn && tbf->trx->trx_no == trx && tbf->control_ts == ts) { OSMO_ASSERT(tbf->direction == GPRS_RLCMAC_DL_TBF); - return tbf; + return static_cast<gprs_rlcmac_dl_tbf *>(tbf); } } return NULL; } -gprs_rlcmac_tbf *BTS::ul_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts) +gprs_rlcmac_ul_tbf *BTS::ul_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts) { - struct gprs_rlcmac_tbf *tbf; + struct gprs_rlcmac_ul_tbf *tbf; /* only one TBF can poll on specific TS/FN, because scheduler can only * schedule one downlink control block (with polling) at a FN per TS */ @@ -279,22 +279,22 @@ gprs_rlcmac_tbf *BTS::ul_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts) && tbf->poll_fn == fn && tbf->trx->trx_no == trx && tbf->control_ts == ts) { OSMO_ASSERT(tbf->direction == GPRS_RLCMAC_UL_TBF); - return tbf; + return static_cast<gprs_rlcmac_ul_tbf *>(tbf); } } return NULL; } /* lookup downlink TBF Entity (by TFI) */ -gprs_rlcmac_tbf *BTS::dl_tbf_by_tfi(uint8_t tfi, uint8_t trx) +gprs_rlcmac_dl_tbf *BTS::dl_tbf_by_tfi(uint8_t tfi, uint8_t trx) { - return tbf_by_tfi(tfi, trx, GPRS_RLCMAC_DL_TBF); + return static_cast<gprs_rlcmac_dl_tbf *>(tbf_by_tfi(tfi, trx, GPRS_RLCMAC_DL_TBF)); } /* lookup uplink TBF Entity (by TFI) */ -gprs_rlcmac_tbf *BTS::ul_tbf_by_tfi(uint8_t tfi, uint8_t trx) +gprs_rlcmac_ul_tbf *BTS::ul_tbf_by_tfi(uint8_t tfi, uint8_t trx) { - return tbf_by_tfi(tfi, trx, GPRS_RLCMAC_UL_TBF); + return static_cast<gprs_rlcmac_ul_tbf *>(tbf_by_tfi(tfi, trx, GPRS_RLCMAC_UL_TBF)); } /* lookup TBF Entity (by TFI) */ @@ -1047,12 +1047,12 @@ gprs_rlcmac_tbf *gprs_rlcmac_pdch::tbf_from_list_by_tfi(struct llist_head *tbf_l return NULL; } -gprs_rlcmac_tbf *gprs_rlcmac_pdch::ul_tbf_by_tfi(uint8_t tfi) +gprs_rlcmac_ul_tbf *gprs_rlcmac_pdch::ul_tbf_by_tfi(uint8_t tfi) { - return tbf_from_list_by_tfi(&bts_data()->ul_tbfs, tfi, GPRS_RLCMAC_UL_TBF); + return static_cast<gprs_rlcmac_ul_tbf *>(tbf_from_list_by_tfi(&bts_data()->ul_tbfs, tfi, GPRS_RLCMAC_UL_TBF)); } -gprs_rlcmac_tbf *gprs_rlcmac_pdch::dl_tbf_by_tfi(uint8_t tfi) +gprs_rlcmac_dl_tbf *gprs_rlcmac_pdch::dl_tbf_by_tfi(uint8_t tfi) { - return tbf_from_list_by_tfi(&bts_data()->dl_tbfs, tfi, GPRS_RLCMAC_DL_TBF); + return static_cast<gprs_rlcmac_dl_tbf *>(tbf_from_list_by_tfi(&bts_data()->dl_tbfs, tfi, GPRS_RLCMAC_DL_TBF)); } diff --git a/src/bts.h b/src/bts.h index a0b808e..8f99942 100644 --- a/src/bts.h +++ b/src/bts.h @@ -62,8 +62,8 @@ struct gprs_rlcmac_pdch { BTS *bts() const; uint8_t trx_no() const; - struct gprs_rlcmac_tbf *ul_tbf_by_tfi(uint8_t tfi); - struct gprs_rlcmac_tbf *dl_tbf_by_tfi(uint8_t tfi); + struct gprs_rlcmac_ul_tbf *ul_tbf_by_tfi(uint8_t tfi); + struct gprs_rlcmac_dl_tbf *dl_tbf_by_tfi(uint8_t tfi); #endif uint8_t m_is_enabled; /* TS is enabled */ @@ -193,12 +193,12 @@ public: /** add paging to paging queue(s) */ int add_paging(uint8_t chan_needed, uint8_t *identity_lv); - gprs_rlcmac_tbf *dl_tbf_by_tlli(uint32_t tlli); - gprs_rlcmac_tbf *ul_tbf_by_tlli(uint32_t tlli); - gprs_rlcmac_tbf *dl_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts); - gprs_rlcmac_tbf *ul_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts); - gprs_rlcmac_tbf *dl_tbf_by_tfi(uint8_t tfi, uint8_t trx); - gprs_rlcmac_tbf *ul_tbf_by_tfi(uint8_t tfi, uint8_t trx); + gprs_rlcmac_dl_tbf *dl_tbf_by_tlli(uint32_t tlli); + gprs_rlcmac_ul_tbf *ul_tbf_by_tlli(uint32_t tlli); + gprs_rlcmac_dl_tbf *dl_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts); + gprs_rlcmac_ul_tbf *ul_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts); + gprs_rlcmac_dl_tbf *dl_tbf_by_tfi(uint8_t tfi, uint8_t trx); + gprs_rlcmac_ul_tbf *ul_tbf_by_tfi(uint8_t tfi, uint8_t trx); int tfi_find_free(enum gprs_rlcmac_tbf_direction dir, uint8_t *_trx, int8_t use_trx); -- 1.8.4.2

1 0

osmo-trx with RAD1?
by Ralph A. Schmid, dk5ras 16 Jul '14

16 Jul '14

Hi, When having a look at http://discourse.criticalengineering.org/t/howto-gsm-base-station-with-the-b eaglebone-black-debian-gnu-linux-and-a-usrp/56 I wonder if there is a way to use the osmo-trx with rangenetworks RAD1/SDR1 radio. I do not yet own an Ettus B210, so it would be fun using the BeagleBone with the RAD1/SDR1, what is some shrinked and at the same time extended USRP1 radio. Ralph. -- Ralph A. Schmid Mondstr. 10 90762 Fürth +49-171-3631223 ralph(a)schmid.xxx http://www.bclog.de/

5 9

Re: TRX Clock calibration
by Alexander Chemeris 16 Jul '14

16 Jul '14

Choukoumoun, The question is related to the osmo-trx more than to UmTRX, so I'm moving it to the Osmocom mailing list: On Tue, Jul 15, 2014 at 8:17 PM, choukoumoun <choukoumoun(a)gmail.com> wrote: > Hello guys. > > I have a UMTRX my UHD/OPENBTS or UHD/OSMO-TRX same working fine but wen i > start osmo-trx and openbts i have this error Could you specify where did you get sources and which exact versions and commit numbers do you use? Did you try to use osmo-trx with Osmocom software rather than with OpenBTS? > Osmo-trx : > > linux; GNU C++ version 4.8.1; Boost_105300; UHD_003.004.000-4fa97a1 > > Config Settings > Log Level............... NOTICE > Device args............. addr=192.168.10.2 > TRX Base Port........... 5700 > TRX Address............. 127.0.0.1 > Channels................ 1 > Samples-per-Symbol...... 4 > External Reference...... Disabled > C0 Filler Table......... Disabled > Diversity............... Disabled > Tuning offset........... 0 > > -- Opening a UmTRX device... > -- Current recv frame size: 1472 bytes > -- Current send frame size: 1472 bytes > -- Setting UmTRX 4 SPS > -- Transceiver active with 1 channel(s) > > ALERT 139940970051328 02:13:33.7 UHDDevice.cpp:993:writeSamples: UHD: Device > send timed out > "Constantely" ..... > > > anybody have a idea ? > > > my usrp probe out : > > linux; GNU C++ version 4.8.1; Boost_105300; UHD_003.004.000-4fa97a1 > > -- Opening a UmTRX device... > -- Current recv frame size: 1472 bytes > -- Current send frame size: 1472 bytes > _____________________________________________________ > / > | Device: UmTRX Device > | _____________________________________________________ > | / > | | Mboard: UMTRX-REV0 > | | hardware: 64000 > | | mac-addr: 00:50:c2:85:3f:ff > | | ip-addr: 192.168.10.2 > | | gpsdo: none > | | serial: 7 > | | name: UmTRX > | | tcxo-dac: 2048 > | | > | | Time sources: none, external, _external_, mimo > | | Clock sources: internal, external, mimo > | | Sensors: ref_locked > | | _____________________________________________________ > | | / > | | | RX DSP: 0 > | | | Freq range: -6.500 to 6.500 Mhz > | | _____________________________________________________ > | | / > | | | RX DSP: 1 > | | | Freq range: -6.500 to 6.500 Mhz > | | _____________________________________________________ > | | / > | | | RX Dboard: A > | | | ID: LMS6002D (0xfa07) > | | | Serial: 7.A > | | | _____________________________________________________ > | | | / > | | | | RX Subdev: 0 > | | | | Name: LMS6002D (0xfa07) - 0 > | | | | Antennas: RX0, RX1, RX2, RX3, CAL > | | | | Sensors: lo_locked > | | | | Freq range: 232.500 to 3720.000 Mhz > | | | | Gain range VGA1: 0.0 to 126.0 step 1.0 dB > | | | | Gain range VGA2: 0.0 to 30.0 step 3.0 dB > | | | | Connection Type: IQ > | | | | Uses LO offset: No > | | | _____________________________________________________ > | | | / > | | | | RX Codec: A > | | | | Name: LMS_RX > | | | | Gain Elements: None > | | _____________________________________________________ > | | / > | | | RX Dboard: B > | | | ID: LMS6002D (0xfa07) > | | | Serial: 7.B > | | | _____________________________________________________ > | | | / > | | | | RX Subdev: 0 > | | | | Name: LMS6002D (0xfa07) - 0 > | | | | Antennas: RX0, RX1, RX2, RX3, CAL > | | | | Sensors: lo_locked > | | | | Freq range: 232.500 to 3720.000 Mhz > | | | | Gain range VGA1: 0.0 to 126.0 step 1.0 dB > | | | | Gain range VGA2: 0.0 to 30.0 step 3.0 dB > | | | | Connection Type: IQ > | | | | Uses LO offset: No > | | | _____________________________________________________ > | | | / > | | | | RX Codec: B > | | | | Name: LMS_RX > | | | | Gain Elements: None > | | _____________________________________________________ > | | / > | | | TX DSP: 0 > | | | Freq range: -6.500 to 6.500 Mhz > | | _____________________________________________________ > | | / > | | | TX DSP: 1 > | | | Freq range: -6.500 to 6.500 Mhz > | | _____________________________________________________ > | | / > | | | TX Dboard: A > | | | ID: LMS6002D (0xfa09) > | | | Serial: 7.A > | | | _____________________________________________________ > | | | / > | | | | TX Subdev: 0 > | | | | Name: LMS6002D (0xfa09) - 0 > | | | | Antennas: TX0, TX1, TX2, CAL > | | | | Sensors: lo_locked > | | | | Freq range: 232.500 to 3720.000 Mhz > | | | | Gain range VGA1: -35.0 to -4.0 step 1.0 dB > | | | | Gain range VGA2: 0.0 to 25.0 step 1.0 dB > | | | | Connection Type: IQ > | | | | Uses LO offset: No > | | | _____________________________________________________ > | | | / > | | | | TX Codec: A > | | | | Name: LMS_TX > | | | | Gain Elements: None > | | _____________________________________________________ > | | / > | | | TX Dboard: B > | | | ID: LMS6002D (0xfa09) > | | | Serial: 7.B > | | | _____________________________________________________ > | | | / > | | | | TX Subdev: 0 > | | | | Name: LMS6002D (0xfa09) - 0 > | | | | Antennas: TX0, TX1, TX2, CAL > | | | | Sensors: lo_locked > | | | | Freq range: 232.500 to 3720.000 Mhz > | | | | Gain range VGA1: -35.0 to -4.0 step 1.0 dB > | | | | Gain range VGA2: 0.0 to 25.0 step 1.0 dB > | | | | Connection Type: IQ > | | | | Uses LO offset: No > | | | _____________________________________________________ > | | | / > | | | | TX Codec: B > | | | | Name: LMS_TX > | | | | Gain Elements: None > > > I make the VCXO configuration procedure with the GPS but no success... The issue with "UHD: Device send timed out" have nothing to do with the VCXO calibration. The issue is because of a issue in communication between the host system and the UmTRX. What is your hardware setup? Particularly: - Do you use a switch between the UmTRX and the host system? - Do you run it on a real hardware or in a VM? - Is your host system powerful enough? -- Regards, Alexander Chemeris. CEO, Fairwaves, Inc. https://fairwaves.co

1 0

OpenBSC unsubscribe notification
by mailman-bounces＠lists.osmocom.org 13 Jul '14

13 Jul '14

mki(a)mki-consult.de has been removed from OpenBSC.

1 0

[RFC/PATCH] tlv/msgb: Be able to generate and parse TLV as used by LLDP
by Holger Hans Peter Freyther 10 Jul '14

10 Jul '14

Hi! we have the need to generate and parse LLDP frames and would like to use the TLV infrastructure of libosmocore for it. In LLDP the tag is only 7bit and the length is 9bit long. To make this work with the tlv parsing routines we need to shift the tag. But this means that when somebody is a tlv_definition with the T7L9 type and others we might match the wrong thing. I would like to include the feature as it is right now. The alternative would be to create a separate tlv_parse and tlv_parse_one routine for the other type. comments? holger

3 4

Re: Re: Need help with configuration of Timeslots and of multible USRPs
by Alexander Chemeris 10 Jul '14

10 Jul '14

Norbert, On Thu, Jul 10, 2014 at 11:01 AM, Norbert Englberger <Norbert.Englberger(a)unibw.de> wrote: > Alexander Chemeris wrote: >>Just for the sake of full understanding, what you're asking for is called "Late Assignment". Check the specs for the exact details or Wikipedia for a brief explanation: >>http://en.m.wikipedia.org/wiki/Um_interface#Mobile-Originating_Call_.28MOC.… >>I don't have the code near me at the moment, but this might be implemented in the "jolly" branch. > > i was already using the jolly/testing branch. > > Now that you hinted me to the "Late Assignment" I googled that and found that there is a "jolly/new_handover" branch using it. > Is this still the branch I should use, because when I tried to compile it, make failed. (I am right now trying to google if there are > other dependencies I have to change the branch and to recompile) I have never worked with this branch, so can't comment unfortunately. AFAIK, this branch is not being used in real environments, so it's most likely bit-rotten. My suggestion is to take the implementation there as an inspiration and re-implement it for master. That would make a good work for one of your students probably. Please don't move the conversation into private and keep the mailing list CC'ed. -- Regards, Alexander Chemeris. CEO, Fairwaves, Inc. https://fairwaves.co

1 0

Need help with configuration of Timeslots and of multible USRPs
by Norbert Englberger 09 Jul '14

09 Jul '14

Hello everybody, after starting on my new Job at an University I have been tasked to build up an BTS for our students. Being a complete noob to both Linux and GSM it took me some time, but I finally got OpenBSC and Osmo-BTS up and running on Ubuntu Linux 12.04 with our ETTUS N210. After doing some experiments with this setup, our students now ask me to do some changes which I couldn't figure out yet, how to do. 1. The students told me, that OpenBSC doesn't behave like all the BTS from commercial Providers they analysed: The say that on commercial nets at the beginning of a phone call there will be send an "Immediate assignment" sending the MS always to a SDCCH on TS 1 or TS2 where the encryption will be startet. After establishing the encryption another "Immediate assignment" will occour and send the MS to a Traffic Channel (which also contains the FACCH / SACCH) for the actual Communication. OpenBSC also sends the "Immediate assignment" on TS0 but sends the MS directly to Timeslot with a Traffic Channel where all the signalling, like encryption, will be done in the SACCH. Is there any way to configure OsmoBSC to behave like the commercial BTS the students described? 2. I have been asked if I could enable "Frequency Hopping" Right now I already fail to bring up a second TRX with OpenBSC. I start 2 instances of "osmo-trx" with 2 differents IP-Addresses but the same port, after that I start "osmobts-trx" with the "-t 2" parameter. OpenBSC seems to see both USRPs but the LED'S an the USRPs indicate that only one is completely powered up and transmitting (a quick check with an Motorola with RSSI also shows only one ARFC). So how do I have to start "osmo-trx" to enable transmitting on both URSPs as TRX0 and TRX1 in one BTS? After that, is it possible to configure "Frequency hopping" with this setup? (Because when I changed the value "hopping enabled" of the timeslots in open-bts.cfg to "1" I get a message that this type of BTS does not support hopping) I hope that anyone can clarify those two points to me. Thanks in advance Norbie

3 2

Re:Re: Need help with configuration of Timeslots and of multible USRPs
by Alexander Chemeris 09 Jul '14

09 Jul '14

Hi Norbert, 08 июля 2014 г. 16:34 пользователь "Norbert Englberger" < Norbert.Englberger(a)unibw.de> написал: > > > Hello again, > > Holger Hans Peter Freyther wrote: > > > it is not a question of BTS but the BSC. The role of the BSC is to manage > the > > channels. Inside src/libbsc/gsm_04_08_utils.c you will see a table that > > determines how the BSC/NITB will respond to "channel request" inside the > > RACH request by the phone. > > > > E.g. if you change get_ctype_by_chreq to always return LCHAN_SDCCH you > > should be close to the behavior you have described. > > > > First off all: Thank you for your fast response. > > I was hoping that it would be enough to change the settings in the > open-bsc.cfg and I wouldn't have to temper the source code, as I said, I'm > new to all the GSM stuff and Linux. Just for the sake of full understanding, what you're asking for is called "Late Assignment". Check the specs for the exact details or Wikipedia for a brief explanation: http://en.m.wikipedia.org/wiki/Um_interface#Mobile-Originating_Call_.28MOC.… I don't have the code near me at the moment, but this might be implemented in the "jolly" branch. Please excuse typos. Written with a touchscreen keyboard. -- Regards, Alexander Chemeris CEO Fairwaves, Inc. https://fairwaves.co

1 0

Re:Re: Need help with configuration of Timeslots and of multible USRPs
by Norbert Englberger 08 Jul '14

08 Jul '14

Hello again, Holger Hans Peter Freyther wrote: > it is not a question of BTS but the BSC. The role of the BSC is to manage the > channels. Inside src/libbsc/gsm_04_08_utils.c you will see a table that > determines how the BSC/NITB will respond to "channel request" inside the > RACH request by the phone. > > E.g. if you change get_ctype_by_chreq to always return LCHAN_SDCCH you > should be close to the behavior you have described. > First off all: Thank you for your fast response. I was hoping that it would be enough to change the settings in the open-bsc.cfg and I wouldn't have to temper the source code, as I said, I'm new to all the GSM stuff and Linux. I had a quick look at the code and just tried to comment out everything in "get_ctype_by_chreq" and always return LCHAN_SDCCH as you suggested. While testing this setup I saw, that The "Immediate assignment" now sends the calling MS to TS 1 (which is configured as SDCCH/8) subslot 0 and starts encryption and sends SI5 and SI6 messages on this timeslot/subslot. Also the called MS gets "Immediate assignment" and is sent to TS 1 subslot 1, encryption starts and so on. Now the called MS sends "Call Confirmed", but shortly after this the channels are released. (The called MS wasn't ringing at all) Am I right, that this is because both MS are yet sticking on the SDCCH and are not going to a Transport channel? (would make sense, because I always return SDCCH when the BSC wants to change channel) I think I have to change the Table " ctype_by_chreq" or to find a way to insert another "Immediate Assignmend" between "encryption" and "call confirmed" So I think I'll have to study the source code for a while, Again thanks for the hint where to start. [Norbert Englberger]

1 0