Project

General

Profile

OsmoTRX » History » Version 39

ttsou, 02/19/2016 10:47 PM

1 34 ipse
[[PageOutline]]
2 1 ttsou
= OsmoTRX =
3 1 ttsou
4 1 ttsou
OsmoTRX is a software-defined radio transceiver that implements the Layer 1 physical layer of a BTS comprising the following 3GPP specifications:
5 1 ttsou
 * TS 05.01 "Physical layer on the radio path"
6 1 ttsou
 * TS 05.02 "Multiplexing and Multiple Access on the Radio Path"
7 1 ttsou
 * TS 05.04 "Modulation"
8 1 ttsou
 * TS 05.10 "Radio subsystem synchronization"
9 1 ttsou
10 39 ttsou
OsmoTRX is based on the transceiver code from the OpenBTS project, but setup to operate independently with the purpose of using with non-OpenBTS software and projects, while still maintaining backwards compatibility with OpenBTS. Currently there are numerous features contained in OsmoTRX that extend the functionality of the OpenBTS transceiver. These features include enhanced support for various embedded platforms - notably ARM - and dual channel diversity support for the Fairwaves UmTRX.
11 6 ttsou
12 6 ttsou
== Features ==
13 6 ttsou
14 16 ttsou
'''Intel SSE Support'''
15 6 ttsou
* SSE3
16 6 ttsou
* SSE4.1
17 6 ttsou
18 20 ttsou
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.
19 1 ttsou
20 26 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.
21 20 ttsou
22 29 ttsou
'''ARM Support'''
23 6 ttsou
* NEON
24 1 ttsou
* NEON-VFPv4
25 6 ttsou
26 20 ttsou
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.
27 1 ttsou
28 20 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.
29 20 ttsou
30 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.
31 20 ttsou
32 37 ttsou
'''Dual Channel (UmTRX and B210)'''
33 7 ttsou
34 7 ttsou
Two dual channel modes are available: standard dual channel mode and diversity. In standard dual channel mode, each RF
35 37 ttsou
path of the dual channel device supports a different ARFCN. Each path operates independently a
36 28 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.
37 1 ttsou
38 37 ttsou
'''Dual Channel Diversity (UmTRX, experimental)'''
39 1 ttsou
40 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.
41 16 ttsou
42 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.
43 20 ttsou
44 39 ttsou
'''Uplink Burst Detection'''
45 30 ttsou
46 39 ttsou
OsmoTRX utilizes an updated receive burst detection algorithm that provides greater sensitivity and reliability than the original OpenBTS approach, which relied on energy detection for the initial stage of burst acquisition.
47 30 ttsou
48 30 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.
49 30 ttsou
50 39 ttsou
The current receiver code addressed those limitations for improved performance in a wider variety of environments.
51 30 ttsou
52 16 ttsou
'''Low Phase Error Modulator'''
53 16 ttsou
54 1 ttsou
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.
55 1 ttsou
56 1 ttsou
Theoretical details can be found in the report on [http://tsou.cc/gsm/report_gmsk.pdf GMSK]. Octave / Matlab code for [http://tsou.cc/gsm/laurent.m pulse generation] is also available.
57 28 ttsou
58 28 ttsou
This option can be enabled or disabled at run time from the command line.
59 16 ttsou
60 20 ttsou
Very Low Phase Error (Ettus Research N200)
61 1 ttsou
62 1 ttsou
[[Image(http://tsou.cc/gsm/osmo-trx-phase75.gif)]]
63 1 ttsou
64 21 ttsou
Spectrum Mask (Ettus Research N200)
65 1 ttsou
66 1 ttsou
[[Image(http://tsou.cc/gsm/osmo-trx-spectrum75.gif)]]
67 1 ttsou
68 20 ttsou
== RF Hardware support ==
69 1 ttsou
70 20 ttsou
Multiple RF devices are currently supported. These include USRP family products from Ettus Research, and the UmTRX from Fairwaves.
71 1 ttsou
72 20 ttsou
||'''Fairwaves'''||'''Notes'''||
73 20 ttsou
||UmTRX||Dual channel||
74 20 ttsou
75 20 ttsou
All Ettus Research devices are supported.
76 20 ttsou
77 20 ttsou
||'''Ettus Research'''||'''Notes'''||
78 20 ttsou
||USRP1||Requires legacy libusrp driver and clocking modification||
79 20 ttsou
||USRP2||10 MHz external reference required||
80 1 ttsou
||B100||
81 1 ttsou
||B110||
82 39 ttsou
||B200||GPSDO or 10 MHz external reference or recommended||
83 39 ttsou
||B210||Dual channel, 10 MHz external reference or recommended||
84 1 ttsou
||N200||
85 1 ttsou
||N210||
86 1 ttsou
||E100||
87 20 ttsou
||E110||
88 20 ttsou
89 1 ttsou
== Embedded Platform Support ==
90 1 ttsou
91 20 ttsou
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.
92 1 ttsou
93 20 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.
94 20 ttsou
95 20 ttsou
||'''Platform'''||'''SoC'''||'''Processor'''||'''SIMD/FPU'''||'''Testing Notes'''
96 20 ttsou
||!ArndaleBoard||Samsung Exynos 5250||ARM Cortex-A15||NEON-VFPv4||7 TCH||
97 21 ttsou
||!BeagleBoard-xM||Texas Instruments OMAP3||ARM Cortex-A8||NEON||7 TCH, remote OsmoBTS stack||
98 21 ttsou
||Ettus E100||Texas Instruments OMAP3||ARM Cortex-A8||NEON||7 TCH, remote OsmoBTS stack||
99 21 ttsou
||Raspberry Pi||Broadcom BCM2835||ARM11||VFP||2 TCH, remote OsmoBTS stack||
100 1 ttsou
||Shuttle PC||NA||Intel Atom D2550||SSE3||Dual channel, 15 TCH||
101 20 ttsou
102 25 ttsou
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.
103 19 ttsou
104 19 ttsou
== Mailing List ==
105 18 ttsou
106 1 ttsou
For development purposes, OsmoTRX is discussed on both OpenBTS and OpenBSC mailing lists at openbts-discuss@lists.sourceforge.net and openbsc@lists.osmocom.org respectively.
107 1 ttsou
108 22 ttsou
Please direct questions and bug reports to the list appropriate for the GSM stack being used.
109 19 ttsou
110 16 ttsou
Subscription information is available at [https://lists.sourceforge.net/lists/listinfo/openbts-discuss] and [http://lists.osmocom.org/mailman/listinfo/openbsc/].
111 19 ttsou
112 19 ttsou
== GPRS support ==
113 16 ttsou
114 1 ttsou
OsmoTRX supports GPRS through OsmoBTS.
115 16 ttsou
116 1 ttsou
For GPRS support with OpenBTS, please use the transceiver supplied with OpenBTS.
117 1 ttsou
118 1 ttsou
== Source code ==
119 16 ttsou
120 1 ttsou
The source code is available from git.osmocom.org (module osmo-trx).
121 1 ttsou
122 1 ttsou
Public read-only access is available via
123 36 9600
{{{
124 36 9600
$ git clone git://git.osmocom.org/osmo-trx
125 36 9600
}}}
126 1 ttsou
You can browse it via cgit: http://cgit.osmocom.org/cgit/osmo-trx/
127 18 ttsou
128 18 ttsou
== Configuration and Build ==
129 19 ttsou
130 1 ttsou
The only package dependency is the Universal Hardware Driver (UHD), which is available from Ettus Research or Fairwaves depending on the device. Please note that the UHD implementation must match hardware (i.e. Ettus Research UHD for USRP devices and Fairwaves UHD with UmTRX). The one device that does not use the UHD driver is the USRP1, which is supported through the legacy libusrp driver provided in GNU Radio 3.4.2.
131 36 9600
132 36 9600
First, run autoreconf to remake the build system files.
133 36 9600
{{{
134 36 9600
$ autoreconf -i
135 36 9600
...
136 36 9600
}}}
137 1 ttsou
138 18 ttsou
'''Intel Platforms (All)'''
139 18 ttsou
140 18 ttsou
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.
141 18 ttsou
{{{
142 18 ttsou
$ ./configure
143 18 ttsou
...
144 1 ttsou
checking whether mmx is supported... yes
145 18 ttsou
checking whether sse is supported... yes
146 18 ttsou
checking whether sse2 is supported... yes
147 18 ttsou
checking whether sse3 is supported... yes
148 18 ttsou
checking whether ssse3 is supported... yes
149 18 ttsou
checking whether sse4.1 is supported... yes
150 1 ttsou
checking whether sse4.2 is supported... yes
151 19 ttsou
...
152 18 ttsou
}}}
153 18 ttsou
154 18 ttsou
'''ARM Platforms with NEON'''
155 18 ttsou
156 18 ttsou
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.
157 18 ttsou
{{{
158 19 ttsou
$ ./configure --with-neon
159 18 ttsou
}}}
160 18 ttsou
161 18 ttsou
'''ARM Platforms with NEON-VFPv4'''
162 18 ttsou
163 18 ttsou
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.
164 18 ttsou
{{{
165 18 ttsou
$ ./configure --with-neon-vfpv4
166 18 ttsou
}}}
167 18 ttsou
168 18 ttsou
'''ARM Platforms without NEON'''
169 18 ttsou
170 18 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.
171 1 ttsou
172 24 ttsou
Coming soon...
173 18 ttsou
174 16 ttsou
'''Build and Install'''
175 16 ttsou
176 16 ttsou
After configuration, installation is simple.
177 16 ttsou
178 16 ttsou
{{{
179 18 ttsou
$ make
180 19 ttsou
$ sudo make install
181 16 ttsou
}}}
182 16 ttsou
183 16 ttsou
== Running ==
184 16 ttsou
185 16 ttsou
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.
186 16 ttsou
187 16 ttsou
{{{
188 16 ttsou
$ osmo-trx -h
189 16 ttsou
linux; GNU C++ version 4.8.1 20130603 (Red Hat 4.8.1-1); Boost_105300; UHD_003.005.004-140-gfb32ed16
190 16 ttsou
191 16 ttsou
Options:
192 16 ttsou
  -h    This text
193 16 ttsou
  -a    UHD device args
194 16 ttsou
  -l    Logging level (EMERG, ALERT, CRT, ERR, WARNING, NOTICE, INFO, DEBUG)
195 16 ttsou
  -i    IP address of GSM core
196 16 ttsou
  -p    Base port number
197 16 ttsou
  -d    Enable dual channel diversity receiver
198 16 ttsou
  -x    Enable external 10 MHz reference
199 16 ttsou
  -s    Samples-per-symbol (1 or 4)
200 16 ttsou
  -c    Number of ARFCN channels (default=1)
201 38 ttsou
  -f    Enable C0 filler table
202 38 ttsou
  -o    Set baseband frequency offset (default=auto)
203 16 ttsou
}}}
204 16 ttsou
205 16 ttsou
{{{
206 16 ttsou
$ osmo-trx -a "addr=192.168.10.2"
207 16 ttsou
linux; GNU C++ version 4.8.1 20130603 (Red Hat 4.8.1-1); Boost_105300; UHD_003.004.000-b14cde5
208 16 ttsou
209 16 ttsou
Config Settings
210 16 ttsou
   Log Level............... INFO
211 16 ttsou
   Device args............. addr=192.168.10.2
212 16 ttsou
   TRX Base Port........... 5700
213 1 ttsou
   TRX Address............. 127.0.0.1
214 16 ttsou
   Channels................ 1
215 1 ttsou
   Samples-per-Symbol...... 4
216 1 ttsou
   External Reference...... Disabled
217 16 ttsou
   Diversity............... Disabled
218 16 ttsou
219 16 ttsou
-- Opening a UmTRX device...
220 16 ttsou
-- Current recv frame size: 1472 bytes
221 16 ttsou
-- Current send frame size: 1472 bytes
222 16 ttsou
-- Setting UmTRX 4 SPS
223 19 ttsou
-- Transceiver active with 1 channel(s)
224 13 ttsou
}}}
225 38 ttsou
226 38 ttsou
== OsmoTRX with OpenBTS ==
227 38 ttsou
228 39 ttsou
OsmoTRX is fully compatible with OpenBTS for voice and SMS services. Due to differences in handing of GPRS, OsmoTRX does not support GPRS when used with OpenBTS, however, GPRS with the Osmocom stack is supported.
229 38 ttsou
230 38 ttsou
For use with OpenBTS, enable the filler table option "Enable C0 filler table", which enables OpenBTS style idle bursts and retransmissions.
231 38 ttsou
232 38 ttsou
{{{
233 38 ttsou
$ osmo-trx -f
234 38 ttsou
}}}
235 38 ttsou
236 38 ttsou
The OsmoTRX transceiver should be started before running OpenBTS. No symbolic link to './transceiver' should exist in the OpenBTS directory. This prevents OpenBTS from starting its own transceiver instance.
237 19 ttsou
238 19 ttsou
== Benchmarks ==
239 17 ttsou
240 13 ttsou
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 OpenBTS core.
241 13 ttsou
242 35 ttsou
'''Repository'''
243 1 ttsou
244 35 ttsou
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.
245 1 ttsou
246 35 ttsou
{{{
247 35 ttsou
$ git clone https://github.com/ttsou/trx-bench.git
248 35 ttsou
249 35 ttsou
$ cd trx-bench
250 35 ttsou
$ autoreconf -i
251 35 ttsou
$ ./configure [--with-neon] [--with-neon-vfp4]
252 35 ttsou
$ make
253 35 ttsou
$ src/conv_test
254 35 ttsou
$ src/convert_test
255 35 ttsou
$ src/convolve_test
256 35 ttsou
}}}
257 35 ttsou
258 35 ttsou
The convolutional decoding test includes command options including experimental support for benchmarking with multiple threads.
259 35 ttsou
260 35 ttsou
{{{
261 35 ttsou
$ ./conv_test -h
262 35 ttsou
Options:
263 35 ttsou
  -h    This text
264 35 ttsou
  -i    Number of iterations
265 35 ttsou
  -j    Number of threads for benchmark (1 to 32)
266 35 ttsou
  -b    Run benchmark tests
267 35 ttsou
  -a    Run validity checks
268 35 ttsou
  -e    Run bit-error-rate tests
269 35 ttsou
}}}
270 35 ttsou
271 35 ttsou
Selected benchmark results are provided below. All tests are run on a single core only.
272 35 ttsou
273 13 ttsou
'''Intel Haswell (i7 4770K 3.5 GHz)'''
274 13 ttsou
275 13 ttsou
{{{
276 10 ttsou
--- Floating point to integer conversions
277 10 ttsou
-- Testing 40000 iterations of 3120 values
278 1 ttsou
- Measuring conversion time
279 1 ttsou
- Elapsed time base...                  0.065508 secs
280 1 ttsou
- Validating SIMD conversion results... PASS
281 1 ttsou
- Measuring conversion time
282 1 ttsou
- Elapsed time SIMD ...                 0.011424 secs
283 1 ttsou
- Speedup...                            5.734244
284 1 ttsou
}}}
285 1 ttsou
286 1 ttsou
{{{
287 1 ttsou
[+] Testing: GSM TCH/AFS 7.95 (recursive, flushed, punctured)
288 1 ttsou
[.] Input length  : ret = 165  exp = 165 -> OK
289 1 ttsou
[.] Output length : ret = 448  exp = 448 -> OK
290 3 ttsou
[.] Pre computed vector checks:
291 3 ttsou
[..] Encoding: OK
292 3 ttsou
[..] Decoding base: 
293 3 ttsou
[..] Decoding SIMD: 
294 3 ttsou
[..] Code N 3
295 3 ttsou
[..] Code K 7
296 3 ttsou
OK
297 3 ttsou
[.] Random vector checks:
298 3 ttsou
[.] Testing baseline:
299 3 ttsou
[..] Encoding / Decoding 10000 cycles:
300 3 ttsou
[.] Elapsed time........................ 1.435066 secs
301 3 ttsou
[.] Rate................................ 3.121808 Mbps
302 1 ttsou
[.] Testing SIMD:
303 1 ttsou
[..] Encoding / Decoding 10000 cycles:
304 17 ttsou
[.] Elapsed time........................ 0.073524 secs
305 1 ttsou
[.] Rate................................ 60.932485 Mbps
306 1 ttsou
[.] Speedup............................. 19.518334
307 1 ttsou
}}}
308 1 ttsou
309 17 ttsou
'''Intel Atom (D2500 1.86 GHz)'''
310 17 ttsou
{{{
311 17 ttsou
--- Floating point to integer conversions
312 17 ttsou
-- Testing 40000 iterations of 3120 values
313 17 ttsou
- Measuring conversion time
314 17 ttsou
- Elapsed time base...                 1.147449 secs
315 17 ttsou
- Validating SSE conversion results... PASS
316 17 ttsou
- Measuring conversion time
317 17 ttsou
- Elapsed time SSE ...                 0.347838 secs
318 17 ttsou
- Quotient...                          3.298803
319 17 ttsou
}}}
320 17 ttsou
321 17 ttsou
{{{
322 17 ttsou
[+] Testing: GSM TCH/AFS 7.95 (recursive, flushed, punctured)
323 17 ttsou
[.] Input length  : ret = 165  exp = 165 -> OK
324 17 ttsou
[.] Output length : ret = 448  exp = 448 -> OK
325 17 ttsou
[.] Pre computed vector checks:
326 17 ttsou
[..] Encoding: OK
327 17 ttsou
[..] Decoding base: 
328 17 ttsou
[..] Decoding SIMD: 
329 17 ttsou
[..] Code N 3
330 1 ttsou
[..] Code K 7
331 1 ttsou
OK
332 1 ttsou
[.] Random vector checks:
333 17 ttsou
[.] Testing baseline:
334 17 ttsou
[..] Encoding / Decoding 10000 cycles:
335 17 ttsou
[.] Elapsed time........................ 11.822688 secs
336 17 ttsou
[.] Rate................................ 0.378932 Mbps
337 17 ttsou
[.] Testing SIMD:
338 17 ttsou
[..] Encoding / Decoding 10000 cycles:
339 17 ttsou
[.] Elapsed time........................ 0.550423 secs
340 19 ttsou
[.] Rate................................ 8.139195 Mbps
341 19 ttsou
[.] Speedup............................. 21.479277
342 19 ttsou
}}}
343 17 ttsou
344 17 ttsou
'''!ArndaleBoard (ARM Cortex-A15 1.7 GHz)'''
345 17 ttsou
346 17 ttsou
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.
347 17 ttsou
348 17 ttsou
{{{
349 17 ttsou
--- Floating point to integer conversions
350 17 ttsou
-- Testing 40000 iterations of 3120 values
351 17 ttsou
- Measuring conversion time
352 17 ttsou
- Elapsed time base...                 0.384097 secs
353 17 ttsou
- Validating SSE conversion results... PASS
354 17 ttsou
- Measuring conversion time
355 17 ttsou
- Elapsed time SSE ...                 0.100877 secs
356 17 ttsou
- Quotient...                          3.807578
357 17 ttsou
}}}
358 17 ttsou
359 17 ttsou
{{{
360 17 ttsou
[+] Testing: GSM TCH/AFS 7.95 (recursive, flushed, punctured)
361 17 ttsou
[.] Input length  : ret = 165  exp = 165 -> OK
362 17 ttsou
[.] Output length : ret = 448  exp = 448 -> OK
363 17 ttsou
[.] Pre computed vector checks:
364 17 ttsou
[..] Encoding: OK
365 17 ttsou
[..] Decoding base: 
366 17 ttsou
[..] Decoding SIMD: 
367 17 ttsou
[..] Code N 3
368 17 ttsou
[..] Code K 7
369 17 ttsou
OK
370 17 ttsou
[.] Random vector checks:
371 17 ttsou
[.] Testing baseline:
372 17 ttsou
[..] Encoding / Decoding 10000 cycles:
373 17 ttsou
[.] Elapsed time........................ 5.371288 secs
374 17 ttsou
[.] Rate................................ 0.834064 Mbps
375 17 ttsou
[.] Testing SIMD:
376 17 ttsou
[..] Encoding / Decoding 10000 cycles:
377 17 ttsou
[.] Elapsed time........................ 1.016621 secs
378 17 ttsou
[.] Rate................................ 4.406755 Mbps
379 17 ttsou
[.] Speedup............................. 5.283471
380 17 ttsou
}}}
381 17 ttsou
382 3 ttsou
'''!BeagleBoard-xM (ARM Cortex-A8 800 MHz)'''
383 3 ttsou
{{{
384 3 ttsou
--- Floating point to integer conversions
385 3 ttsou
-- Testing 40000 iterations of 3120 values
386 3 ttsou
- Measuring conversion time
387 3 ttsou
- Elapsed time base...                  6.292542 secs
388 3 ttsou
- Validating SIMD conversion results... PASS
389 3 ttsou
- Measuring conversion time
390 3 ttsou
- Elapsed time SIMD ...                 0.839081 secs
391 5 ttsou
- Quotient...                           7.499326
392 3 ttsou
}}}
393 3 ttsou
394 3 ttsou
{{{
395 4 ttsou
[+] Testing: GSM TCH/AFS 7.95 (recursive, flushed, punctured)
396 3 ttsou
[.] Input length  : ret = 165  exp = 165 -> OK
397 3 ttsou
[.] Output length : ret = 448  exp = 448 -> OK
398 3 ttsou
[.] Pre computed vector checks:
399 1 ttsou
[..] Encoding: OK
400 1 ttsou
[..] Decoding base: 
401 1 ttsou
[..] Decoding SIMD: 
402 31 ttsou
[..] Code N 3
403 31 ttsou
[..] Code K 7
404 31 ttsou
OK
405 31 ttsou
[.] Random vector checks:
406 31 ttsou
[.] Testing baseline:
407 31 ttsou
[..] Encoding / Decoding 10000 cycles:
408 31 ttsou
[.] Elapsed time........................ 21.963257 secs
409 31 ttsou
[.] Rate................................ 0.203977 Mbps
410 31 ttsou
[.] Testing SIMD:
411 1 ttsou
[..] Encoding / Decoding 10000 cycles:
412 32 ttsou
[.] Elapsed time........................ 3.083282 secs
413 32 ttsou
[.] Rate................................ 1.452997 Mbps
414 32 ttsou
[.] Speedup............................. 7.123337
415 32 ttsou
}}}
416 32 ttsou
417 32 ttsou
418 32 ttsou
'''Full Results'''
419 32 ttsou
420 32 ttsou
[http://tsou.cc/gsm/haswell.txt]
421 32 ttsou
422 32 ttsou
[http://tsou.cc/gsm/shuttle.txt]
423 32 ttsou
424 32 ttsou
[http://tsou.cc/gsm/arndale.txt]
425 32 ttsou
426 32 ttsou
[http://tsou.cc/gsm/beagle.txt]
427 32 ttsou
428 31 ttsou
429 1 ttsou
== Authors ==
430 1 ttsou
431 30 ttsou
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 OpenBTS project, which was originally developed by David Burgess and Harvind Samra at Range Networks.
Add picture from clipboard (Maximum size: 48.8 MB)