Project

General

Profile

OsmoTRX » History » Version 56

wirelesss, 12/21/2016 04:56 PM

1 41 sylvain
{{>toc}}
2 1 ttsou
3 41 sylvain
h1. [[OsmoTRX]]
4 1 ttsou
5
6 41 sylvain
[[OsmoTRX]] is a software-defined radio transceiver that implements the Layer 1 physical layer of a BTS comprising the following 3GPP specifications:
7
* TS 05.01 "Physical layer on the radio path"
8
* TS 05.02 "Multiplexing and Multiple Access on the Radio Path"
9
* TS 05.04 "Modulation"
10
* TS 05.10 "Radio subsystem synchronization"
11 1 ttsou
12 49 neels
[[OsmoTRX]] is based on the transceiver code from the [[OsmoBTS:OpenBTS]] project, but setup to operate independently with the purpose of using with non-OpenBTS software and projects, while still maintaining backwards compatibility with [[OsmoBTS:OpenBTS]]. Currently there are numerous features contained in [[OsmoTRX:]] that extend the functionality of the [[OsmoBTS:OpenBTS]] transceiver. These features include enhanced support for various embedded platforms - notably ARM - and dual channel diversity support for the Fairwaves [[umtrx:]].
13 41 sylvain
14 46 laforge
h2. OsmoTRX in the Osmocom GSM architecture
15
16
{{graphviz_link()
17
digraph G {
18
    rankdir = LR;
19
    SDR -> OsmoTRX [label="Raw Samples"];
20
    OsmoTRX -> OsmoBTS [label="bursts over UDP"];
21
    OsmoBTS -> OsmoNITB [label="Abis/IP"];
22
    OsmoBTS -> OsmoPCU [label="pcu_sock"];
23
    OsmoPCU -> OsmoSGSN [label="Gb/IP"];
24
    OsmoTRX [color=red];
25
}
26
}}
27 41 sylvain
28
h2. Features
29
30
31
*Intel SSE Support*
32 6 ttsou
* SSE3
33
* SSE4.1
34 20 ttsou
35 41 sylvain
On Intel processors, [[OsmoTRX]] makes heavy use of the Streaming SIMD Extensions (SSE) instruction set. Accelerated operations include pulse shape filtering, resampling, sequence correlation, and many other signal processing operations. SSE3 is the minimum requirement for accelerated use.
36 1 ttsou
37 20 ttsou
SSE3 is present in the majority of Intel processors since later versions of the Pentium 4 architecture and is also present on low power Atom processors. Support is automatically detected at build time. For additional performance information, please see the performance and benchmarks section.
38 29 ttsou
39 41 sylvain
*ARM Support*
40 1 ttsou
* NEON
41
* NEON-VFPv4
42 20 ttsou
43 41 sylvain
[[OsmoTRX]] runs on a variety of ARM processors with and without NEON coprocessors. Like SSE on Intel processors, NEON provides acceleration with SIMD vectorized instructions.
44 20 ttsou
45 1 ttsou
Tested popular architectures include ARM11 (Raspberry Pi), Cortex-A8 (!BeagleBoard), and Cortex-A15 (!ArndaleBoard). Loosely speaking, these platforms are representative of low cost embedded devices, mid-level handsets, and high-end smartphones respectively. Similarly, in order, these platforms include no NEON coprocessor, standard NEON, and NEON-VFPv4. The latter NEON variation, VFPv4, provides additional fused-multiply-accumulate (FMA) instructions useful for many DSP operations.
46
47 26 ttsou
NEON support must be enabled by the user at build time. For additional information, please see the configuration and performance and benchmarks sections.
48 37 ttsou
49 41 sylvain
*Dual Channel (UmTRX and B210)*
50 7 ttsou
51 1 ttsou
Two dual channel modes are available: standard dual channel mode and diversity. In standard dual channel mode, each RF
52 28 ttsou
path of the dual channel device supports a different ARFCN. Each path operates independently a
53 1 ttsou
nd operates similarly to two separate devices. GSM channel capacity in this mode is doubled. This option can be configured at run time from the command line.
54
55 41 sylvain
*Dual Channel Diversity (UmTRX, experimental)*
56 1 ttsou
57 28 ttsou
Diversity mode is similar to the standard dual channel mode except each antenna supports both ARFCN channels. In this case, the receiver sample bandwidth is widened to handle both ARFCN's and subsequently converted and demultiplexed into separate sample streams. Each GSM receive path is fed dual signals, where antenna selection diversity is performed by taking the stronger signal on a burst-by-burst basis. This diversity setup improves uplink reception performance in multipath fading environments.
58 16 ttsou
59 28 ttsou
Limitations are increased CPU utilization and that ARFCN spacing is restricted (currently at 400 kHz) by the receiver sampling bandwidth. Setting the ARFCN spacing beyond the sampling limit will disable the diversity path and operate in standard dual channel mode. This options can be configured at run time from the command line.
60 20 ttsou
61 41 sylvain
*Uplink Burst Detection*
62 39 ttsou
63 49 neels
[[OsmoTRX]] utilizes an updated receive burst detection algorithm that provides greater sensitivity and reliability than the original [[OsmoBTS:OpenBTS]] approach, which relied on energy detection for the initial stage of burst acquisition.
64 39 ttsou
65 1 ttsou
The limitation of the previous approach was that it was slow to adapt to highly transient power levels and false burst detection in challenging situations such as receiver saturation, which may occur in close range lab testing. The other issue was that a high degree of level tuning was often necessary to operate reliably.
66
67
The current receiver code addressed those limitations for improved performance in a wider variety of environments.
68
69 41 sylvain
*Low Phase Error Modulator*
70 16 ttsou
71
The default GSM downlink signal is configured for low distortion using a linearized GMSK modulator. The implementation is based on a two pulse Laurent approximation of continuous phase modulated (CPM) signals. The baseband output signal measures with very low phase error and is capable of passing industry spectrum mask requirements. Please note that actual performance will depend strongly on the particular device in use.
72 1 ttsou
73 41 sylvain
Theoretical details can be found in the report on "GMSK":http://tsou.cc/gsm/report_gmsk.pdf. Octave / Matlab code for "pulse generation":http://tsou.cc/gsm/laurent.m is also available.
74 1 ttsou
75
This option can be enabled or disabled at run time from the command line.
76 28 ttsou
77
Very Low Phase Error (Ettus Research N200)
78 16 ttsou
79 42 laforge
!http://tsou.cc/gsm/osmo-trx-phase75.gif!
80 1 ttsou
81 21 ttsou
Spectrum Mask (Ettus Research N200)
82 1 ttsou
83 42 laforge
!http://tsou.cc/gsm/osmo-trx-spectrum75.gif!
84 1 ttsou
85 41 sylvain
h2. RF Hardware support
86 1 ttsou
87
88 50 neels
Multiple RF devices are currently supported. These include USRP family products from Ettus Research, and the [[UmTRX:]] from Fairwaves.
89 41 sylvain
90
||*Fairwaves*||*Notes*||
91 20 ttsou
||UmTRX||Dual channel||
92
93
All Ettus Research devices are supported.
94 1 ttsou
95 41 sylvain
||*Ettus Research*||*Notes*||
96 1 ttsou
||USRP1||Requires legacy libusrp driver and clocking modification||
97
||USRP2||10 MHz external reference required||
98
||B100||
99
||B110||
100
||B200||GPSDO or 10 MHz external reference recommended||
101
||B210||Dual channel, 10 MHz external reference recommended||
102
||N200||
103 20 ttsou
||N210||
104 1 ttsou
||E100||
105
||E110||
106
107 41 sylvain
h2. Embedded Platform Support
108 1 ttsou
109 41 sylvain
110
[[OsmoTRX]] has been tested on the multiple embedded platforms representing a wide range of device types. Low cost ARM devices are generally limited by memory and I/O as much CPU utilization.
111
112 1 ttsou
Running a full or near full ARFCN configuration (7 simultaneous TCH channels with Combination V) may require running the GSM stack remotely, which can be configured at runtime on the command line. This limitation appears to be scheduling related more so than lack of CPU resources, and may be resolved at a later time.
113
114 43 laforge
|_.Platform|_.SoC*|_.Processor|_.SIMD/FPU|_.Testing Notes|
115
|ArndaleBoard|Samsung Exynos 5250|ARM Cortex-A15|NEON-VFPv4|7 TCH|
116
|BeagleBoard-xM|Texas Instruments OMAP3|ARM Cortex-A8|NEON|7 TCH, remote [[osmobts:]] stack|
117
|Ettus E100|Texas Instruments OMAP3|ARM Cortex-A8|NEON|7 TCH, remote [[osmobts:]] stack|
118
|Raspberry Pi|Broadcom BCM2835|ARM11|VFP|2 TCH, remote [[osmobts:]] stack|
119
|Shuttle PC|NA|Intel Atom D2550|SSE3|Dual channel, 15 TCH|
120 1 ttsou
121
All embedded plaforms were tested with low-phase error modulator disabled. Use of the more accurate modulator on embedded platforms has not been extensively tested.
122 19 ttsou
123 41 sylvain
h2. Mailing List
124 22 ttsou
125 41 sylvain
126 49 neels
For development purposes, [[OsmoTRX:]] is discussed on both [[OsmoBTS:OpenBTS]] and [[OpenBSC:]] mailing lists at openbts-discuss@lists.sourceforge.net and openbsc@lists.osmocom.org respectively.
127 41 sylvain
128 1 ttsou
Please direct questions and bug reports to the list appropriate for the GSM stack being used.
129 41 sylvain
130 47 laforge
Subscription information is available at "and [http://lists.osmocom.org/mailman/listinfo/openbsc/":https://lists.sourceforge.net/lists/listinfo/openbts-discuss].  Please make sure to read our [[cellular-infrastructure:MailingListRules]] before posting.
131 1 ttsou
132 41 sylvain
h2. GPRS support
133 1 ttsou
134
135 44 laforge
[[OsmoTRX]] supports GPRS through [[osmobts:]].
136 1 ttsou
137 49 neels
For GPRS support with [[OsmoBTS:OpenBTS]], please use the transceiver supplied with [[OsmoBTS:OpenBTS]].
138 41 sylvain
139
140
h2. Source code
141
142
143 1 ttsou
The source code is available from git.osmocom.org (module osmo-trx).
144 18 ttsou
145
Public read-only access is available via
146 41 sylvain
<pre>
147 19 ttsou
$ git clone git://git.osmocom.org/osmo-trx
148 41 sylvain
</pre>
149 1 ttsou
You can browse it via cgit: http://cgit.osmocom.org/cgit/osmo-trx/
150
151 48 neels
h2. Dependencies
152 1 ttsou
153 48 neels
Install libusb-1.0 and libbost dev packages. On debian 8.4:
154 1 ttsou
155 48 neels
<pre>
156
sudo apt-get install --no-install-recommends libusb-1.0-0-dev libboost-dev
157
</pre>
158 41 sylvain
159 53 neels
h3. UHD
160 1 ttsou
161 48 neels
Unless using USRP1, you will need the Universal Hardware Driver (UHD),
162
which is available from Ettus Research or Fairwaves; the UHD implementation
163
must match your hardware:
164
165
* Ettus Research UHD for USRP devices
166 51 neels
* Fairwaves UHD with [[UmTRX:]]
167 48 neels
* USRP1 does not use the UHD driver, it is supported through the legacy libusrp driver provided in GNU Radio 3.4.2.
168
169 55 wirelesss
h3. UHD for Debian
170 48 neels
171 52 neels
When you are reading this, Debian packages for UHD may be sufficient for running osmo-trx and osmo-bts-trx.
172 48 neels
here are some of the packages that need to be installed:
173
174
<pre>
175 54 neels
sudo apt-get install libuhd-dev uhd-host
176 48 neels
</pre>
177 1 ttsou
178 55 wirelesss
*Troubleshooting:*
179
 
180 52 neels
At the time of writing this (2016-12), for Debian 8 aka jessie you need to use the jessie-backports packages:
181
182
<pre>
183
sudo -s
184
echo "deb http://ftp.de.debian.org/debian jessie-backports main" > /etc/apt/sources.list.d/uhd.list
185
apt-get update
186
apt-get -t jessie-backports install libuhd-dev uhd-host
187
</pre>
188
189
It may also be possible to use the pothos PPA instead:
190 48 neels
191
<pre>
192
sudo add-apt-repository ppa:guruofquality/pothos
193
sudo apt-get update
194
sudo apt install libboost-dev uhd
195
</pre>
196
197 53 neels
h3. Firmware
198 48 neels
199
You also need to download the firmware using a script provided by the UHD package.
200
Instructions suggest running the script as root, but this way is less dangerous:
201
202
<pre>
203
sudo mkdir /usr/share/uhd
204
sudo chown $USER: /usr/share/uhd
205
/usr/lib/uhd/utils/uhd_images_downloader.py
206
</pre>
207
208 53 neels
h3. Group
209 48 neels
210
You may need to add yourself to the usrp group:
211
212
<pre>
213
sudo gpasswd -a $USER usrp
214
# and re-login to acquire the group
215
</pre>
216
217 53 neels
h3. Verify
218 48 neels
219
run uhd_find_devices to make sure b200 is available:
220
221
<pre>
222
$ uhd_find_devices 
223
linux; GNU C++ version 4.9.1; Boost_105500; UHD_003.007.003-0-unknown
224
225
--------------------------------------------------
226
-- UHD Device 0
227
--------------------------------------------------
228
Device Address:
229
    type: b200
230
    name: MyB210
231
    serial: 1C0FFEE
232
    product: B210
233
</pre>
234
235
h2. Configuration and Build
236
237 41 sylvain
First, run autoreconf to remake the build system files.
238 1 ttsou
<pre>
239 18 ttsou
$ autoreconf -i
240 41 sylvain
...
241 18 ttsou
</pre>
242 41 sylvain
243 18 ttsou
*Intel Platforms (All)*
244 1 ttsou
245 41 sylvain
Intel SSE support is automatically detected on Intel x86 platforms. No user intervention is necessary. The general configuration defaults to the low phase error modulator. Atom users may wish to use the low-CPU utilization modulator, which can be later enabled from the command line at runtime.
246 18 ttsou
<pre>
247 1 ttsou
$ ./configure
248
...
249 19 ttsou
checking whether mmx is supported... yes
250 18 ttsou
checking whether sse is supported... yes
251
checking whether sse2 is supported... yes
252
checking whether sse3 is supported... yes
253
checking whether ssse3 is supported... yes
254
checking whether sse4.1 is supported... yes
255
checking whether sse4.2 is supported... yes
256 41 sylvain
...
257 18 ttsou
</pre>
258 41 sylvain
259 18 ttsou
*ARM Platforms with NEON*
260 41 sylvain
261
Many popular ARM development boards fall under this category including BeagleBoard, PandaBoard, and Ettus E100 USRP. This option will disable the low phase error modulator, which can be re-enabled at runtime. NEON support must be manually enabled.
262 24 ttsou
<pre>
263 41 sylvain
$ ./configure --with-neon
264 1 ttsou
</pre>
265 41 sylvain
266 1 ttsou
*ARM Platforms with NEON-VFPv4*
267 41 sylvain
268
Currently very few development platforms support this instruction set, which is seen mainly in high end smartphones and tablets. Available development boards are ArndaleBoard and ODROID-XU. This option will disable the low phase error modulator, which can be re-enabled at runtime. NEON-VFPv4 support must be manually enabled.
269 1 ttsou
<pre>
270 41 sylvain
$ ./configure --with-neon-vfpv4
271 1 ttsou
</pre>
272 41 sylvain
273 1 ttsou
*ARM Platforms without NEON*
274 41 sylvain
275 1 ttsou
This configuration mainly targets the Raspberry Pi. ARM platforms without NEON vector units are almost always very slow processors, and generally not very suitable for running [[OsmoTRX]]. Running [[OsmoTRX]] on a Raspberry Pi, however, is possible along with limited TCH (voice) channel support. Currently this configuration requires minor code changes.
276
277
Coming soon...
278 41 sylvain
279 1 ttsou
*Build and Install*
280 16 ttsou
281
After configuration, installation is simple.
282 41 sylvain
283 16 ttsou
<pre>
284
$ make
285 41 sylvain
$ sudo make install
286 16 ttsou
</pre>
287
288 41 sylvain
h2. Running
289 16 ttsou
290 56 wirelesss
Normally simply start osmo-trx.
291 41 sylvain
292 56 wirelesss
<pre>
293
$ osmo-trx
294
linux; GNU C++ version 5.3.1 20151219; Boost_105800; UHD_003.009.002-0-unknown
295
296
opening configuration table from path :memory:
297
Config Settings
298
   Log Level............... NOTICE
299
   Device args............. 
300
   TRX Base Port........... 5700
301
   TRX Address............. 127.0.0.1
302
   Channels................ 1
303
   Tx Samples-per-Symbol... 4
304
   Rx Samples-per-Symbol... 1
305
   EDGE support............ Disabled
306
   Reference............... Internal
307
   C0 Filler Table......... Disabled
308
   Multi-Carrier........... Disabled
309
   Diversity............... Disabled
310
   Tuning offset........... 0
311
   RSSI to dBm offset...... 0
312
   Swap channels........... 0
313
314
-- Detected Device: B200
315
-- Loading FPGA image: /usr/share/uhd/images/usrp_b200_fpga.bin... done
316
-- Operating over USB 2.
317
-- Detecting internal GPSDO.... No GPSDO found
318
-- Initialize CODEC control...
319
-- Initialize Radio control...
320
-- Performing register loopback test... pass
321
-- Performing CODEC loopback test... pass
322
-- Asking for clock rate 16.000000 MHz... 
323
-- Actually got clock rate 16.000000 MHz.
324
-- Performing timer loopback test... pass
325
-- Setting master clock rate selection to 'automatic'.
326
-- Asking for clock rate 26.000000 MHz... 
327
-- Actually got clock rate 26.000000 MHz.
328
-- Performing timer loopback test... pass
329
-- Setting B200 4/1 Tx/Rx SPS
330
-- Transceiver active with 1 channel(s)
331
</pre>
332
333 41 sylvain
[[OsmoTRX]] can be configured with a variety of options on the command line. In most cases, the default settings will suffice. Notable options include UHD device argument passing, which is often useful for using network based devices with firewalls, and external 10 MHz reference support.
334
335 16 ttsou
<pre>
336
$ osmo-trx -h
337
linux; GNU C++ version 4.8.1 20130603 (Red Hat 4.8.1-1); Boost_105300; UHD_003.005.004-140-gfb32ed16
338
339
Options:
340 1 ttsou
  -h    This text
341 16 ttsou
  -a    UHD device args
342
  -l    Logging level (EMERG, ALERT, CRT, ERR, WARNING, NOTICE, INFO, DEBUG)
343 1 ttsou
  -i    IP address of GSM core
344
  -p    Base port number
345 16 ttsou
  -d    Enable dual channel diversity receiver
346
  -x    Enable external 10 MHz reference
347 38 ttsou
  -s    Samples-per-symbol (1 or 4)
348
  -c    Number of ARFCN channels (default=1)
349 16 ttsou
  -f    Enable C0 filler table
350 41 sylvain
  -o    Set baseband frequency offset (default=auto)
351 16 ttsou
</pre>
352 41 sylvain
353 1 ttsou
<pre>
354 16 ttsou
$ osmo-trx -a "addr=192.168.10.2"
355
linux; GNU C++ version 4.8.1 20130603 (Red Hat 4.8.1-1); Boost_105300; UHD_003.004.000-b14cde5
356
357
Config Settings
358 1 ttsou
   Log Level............... INFO
359 16 ttsou
   Device args............. addr=192.168.10.2
360 1 ttsou
   TRX Base Port........... 5700
361 16 ttsou
   TRX Address............. 127.0.0.1
362
   Channels................ 1
363
   Samples-per-Symbol...... 4
364
   External Reference...... Disabled
365
   Diversity............... Disabled
366 41 sylvain
367 13 ttsou
-- Opening a [[UmTRX]] device...
368 38 ttsou
-- Current recv frame size: 1472 bytes
369 41 sylvain
-- Current send frame size: 1472 bytes
370 38 ttsou
-- Setting [[UmTRX]] 4 SPS
371 41 sylvain
-- Transceiver active with 1 channel(s)
372 38 ttsou
</pre>
373 1 ttsou
374 49 neels
h2. [[OsmoTRX]] with [[OsmoBTS:OpenBTS]]
375 38 ttsou
376
377 49 neels
[[OsmoTRX]] is fully compatible with [[OsmoBTS:OpenBTS]] for voice and SMS services. Due to differences in handing of GPRS, [[OsmoTRX]] does not support GPRS when used with [[OsmoBTS:OpenBTS]], however, GPRS with the Osmocom stack is supported.
378 41 sylvain
379 49 neels
For use with [[OsmoBTS:OpenBTS]], enable the filler table option "Enable C0 filler table", which enables [[OsmoBTS:OpenBTS]] style idle bursts and retransmissions.
380 41 sylvain
381
<pre>
382 1 ttsou
$ osmo-trx -f
383 41 sylvain
</pre>
384 17 ttsou
385 49 neels
The [[OsmoTRX]] transceiver should be started before running [[OsmoBTS:OpenBTS]]. No symbolic link to './transceiver' should exist in the [[OsmoBTS:OpenBTS]] directory. This prevents [[OsmoBTS:OpenBTS]] from starting its own transceiver instance.
386 35 ttsou
387 1 ttsou
388 41 sylvain
h2. Benchmarks
389 1 ttsou
390 35 ttsou
391 49 neels
A variety of performance benchmarks are available for various code optimizations. These include floating point - integer conversions, convolution, and convolutional decoding. Note that convolutional decoding does not take place in [[OsmoTRX]], but one stop higher in the Layer 1 stack - either in [[osmobts:]] or [[OsmoBTS:OpenBTS]] core.
392 35 ttsou
393 41 sylvain
*Repository*
394
395
Currently the trx-bench repository holds the test files and contains the same NEON and SSE code as [[OsmoTRX]]. The test code may be merged into [[OsmoTRX]] at a later time, but, for now, it exists as a separate repository. NEON configure options are the same as [[OsmoTRX]].
396
397
<pre>
398 35 ttsou
$ git clone https://github.com/ttsou/trx-bench.git
399
400
$ cd trx-bench
401
$ autoreconf -i
402
$ ./configure [--with-neon] [--with-neon-vfp4]
403 1 ttsou
$ make
404
$ src/conv_test
405 35 ttsou
$ src/convert_test
406
$ src/convolve_test
407 41 sylvain
</pre>
408 35 ttsou
409
The convolutional decoding test includes command options including experimental support for benchmarking with multiple threads.
410
411 41 sylvain
<pre>
412 35 ttsou
$ ./conv_test -h
413
Options:
414
  -h    This text
415 1 ttsou
  -i    Number of iterations
416
  -j    Number of threads for benchmark (1 to 32)
417 13 ttsou
  -b    Run benchmark tests
418
  -a    Run validity checks
419
  -e    Run bit-error-rate tests
420 41 sylvain
</pre>
421 10 ttsou
422 1 ttsou
Selected benchmark results are provided below. All tests are run on a single core only.
423
424 41 sylvain
*Intel Haswell (i7 4770K 3.5 GHz)*
425 1 ttsou
426 41 sylvain
<pre>
427 1 ttsou
--- Floating point to integer conversions
428
-- Testing 40000 iterations of 3120 values
429
- Measuring conversion time
430
- Elapsed time base...                  0.065508 secs
431
- Validating SIMD conversion results... PASS
432 3 ttsou
- Measuring conversion time
433
- Elapsed time SIMD ...                 0.011424 secs
434
- Speedup...                            5.734244
435 41 sylvain
</pre>
436 1 ttsou
437 41 sylvain
<pre>
438 3 ttsou
[+] Testing: GSM TCH/AFS 7.95 (recursive, flushed, punctured)
439
[.] Input length  : ret = 165  exp = 165 -> OK
440
[.] Output length : ret = 448  exp = 448 -> OK
441
[.] Pre computed vector checks:
442
[..] Encoding: OK
443
[..] Decoding base: 
444
[..] Decoding SIMD: 
445 1 ttsou
[..] Code N 3
446
[..] Code K 7
447
OK
448
[.] Random vector checks:
449
[.] Testing baseline:
450 17 ttsou
[..] Encoding / Decoding 10000 cycles:
451
[.] Elapsed time........................ 1.435066 secs
452
[.] Rate................................ 3.121808 Mbps
453
[.] Testing SIMD:
454
[..] Encoding / Decoding 10000 cycles:
455
[.] Elapsed time........................ 0.073524 secs
456
[.] Rate................................ 60.932485 Mbps
457
[.] Speedup............................. 19.518334
458 41 sylvain
</pre>
459 17 ttsou
460 41 sylvain
*Intel Atom (D2500 1.86 GHz)*
461
<pre>
462 17 ttsou
--- Floating point to integer conversions
463
-- Testing 40000 iterations of 3120 values
464
- Measuring conversion time
465
- Elapsed time base...                 1.147449 secs
466 1 ttsou
- Validating SSE conversion results... PASS
467 17 ttsou
- Measuring conversion time
468 1 ttsou
- Elapsed time SSE ...                 0.347838 secs
469
- Quotient...                          3.298803
470 41 sylvain
</pre>
471 17 ttsou
472 41 sylvain
<pre>
473 1 ttsou
[+] Testing: GSM TCH/AFS 7.95 (recursive, flushed, punctured)
474
[.] Input length  : ret = 165  exp = 165 -> OK
475
[.] Output length : ret = 448  exp = 448 -> OK
476
[.] Pre computed vector checks:
477 17 ttsou
[..] Encoding: OK
478
[..] Decoding base: 
479
[..] Decoding SIMD: 
480
[..] Code N 3
481
[..] Code K 7
482
OK
483
[.] Random vector checks:
484 19 ttsou
[.] Testing baseline:
485
[..] Encoding / Decoding 10000 cycles:
486
[.] Elapsed time........................ 11.822688 secs
487 17 ttsou
[.] Rate................................ 0.378932 Mbps
488
[.] Testing SIMD:
489
[..] Encoding / Decoding 10000 cycles:
490
[.] Elapsed time........................ 0.550423 secs
491
[.] Rate................................ 8.139195 Mbps
492
[.] Speedup............................. 21.479277
493 41 sylvain
</pre>
494 17 ttsou
495 41 sylvain
*!ArndaleBoard (ARM Cortex-A15 1.7 GHz)*
496 17 ttsou
497
Please note that the Viterbi implementations on ARM is largely C based with speedup generated primarily through algorithm changes. In comparison, vector optimization on Intel platforms with SSE is currently much more aggressive, which explains the disparity on decoding performance.
498
499 41 sylvain
<pre>
500 17 ttsou
--- Floating point to integer conversions
501
-- Testing 40000 iterations of 3120 values
502
- Measuring conversion time
503
- Elapsed time base...                 0.384097 secs
504
- Validating SSE conversion results... PASS
505
- Measuring conversion time
506
- Elapsed time SSE ...                 0.100877 secs
507
- Quotient...                          3.807578
508 41 sylvain
</pre>
509 17 ttsou
510 41 sylvain
<pre>
511 17 ttsou
[+] Testing: GSM TCH/AFS 7.95 (recursive, flushed, punctured)
512
[.] Input length  : ret = 165  exp = 165 -> OK
513
[.] Output length : ret = 448  exp = 448 -> OK
514
[.] Pre computed vector checks:
515
[..] Encoding: OK
516
[..] Decoding base: 
517
[..] Decoding SIMD: 
518
[..] Code N 3
519
[..] Code K 7
520
OK
521
[.] Random vector checks:
522
[.] Testing baseline:
523
[..] Encoding / Decoding 10000 cycles:
524
[.] Elapsed time........................ 5.371288 secs
525
[.] Rate................................ 0.834064 Mbps
526 3 ttsou
[.] Testing SIMD:
527
[..] Encoding / Decoding 10000 cycles:
528
[.] Elapsed time........................ 1.016621 secs
529
[.] Rate................................ 4.406755 Mbps
530
[.] Speedup............................. 5.283471
531 41 sylvain
</pre>
532 3 ttsou
533 41 sylvain
*!BeagleBoard-xM (ARM Cortex-A8 800 MHz)*
534
<pre>
535 5 ttsou
--- Floating point to integer conversions
536 3 ttsou
-- Testing 40000 iterations of 3120 values
537
- Measuring conversion time
538
- Elapsed time base...                  6.292542 secs
539 4 ttsou
- Validating SIMD conversion results... PASS
540 3 ttsou
- Measuring conversion time
541
- Elapsed time SIMD ...                 0.839081 secs
542
- Quotient...                           7.499326
543 41 sylvain
</pre>
544 1 ttsou
545 41 sylvain
<pre>
546 31 ttsou
[+] Testing: GSM TCH/AFS 7.95 (recursive, flushed, punctured)
547
[.] Input length  : ret = 165  exp = 165 -> OK
548
[.] Output length : ret = 448  exp = 448 -> OK
549
[.] Pre computed vector checks:
550
[..] Encoding: OK
551
[..] Decoding base: 
552
[..] Decoding SIMD: 
553
[..] Code N 3
554
[..] Code K 7
555 1 ttsou
OK
556 32 ttsou
[.] Random vector checks:
557
[.] Testing baseline:
558
[..] Encoding / Decoding 10000 cycles:
559
[.] Elapsed time........................ 21.963257 secs
560
[.] Rate................................ 0.203977 Mbps
561
[.] Testing SIMD:
562
[..] Encoding / Decoding 10000 cycles:
563
[.] Elapsed time........................ 3.083282 secs
564
[.] Rate................................ 1.452997 Mbps
565
[.] Speedup............................. 7.123337
566 41 sylvain
</pre>
567 32 ttsou
568
569 41 sylvain
*Full Results*
570 32 ttsou
571 41 sylvain
"[http://tsou.cc/gsm/shuttle.txt":http://tsou.cc/gsm/haswell.txt]
572 31 ttsou
573 41 sylvain
"[http://tsou.cc/gsm/beagle.txt":http://tsou.cc/gsm/arndale.txt]
574 1 ttsou
575 30 ttsou
576 1 ttsou
577 41 sylvain
h2. Authors
578 1 ttsou
579
580 49 neels
[[OsmoTRX]] is currently developed and maintained by Thomas Tsou with generous support from Fairwaves, the Open Technology Institute, and Ettus Research. The code is derived from the [[OsmoBTS:OpenBTS]] project, which was originally developed by David Burgess and Harvind Samra at Range Networks.
Add picture from clipboard (Maximum size: 48.8 MB)