Error running

Skip to first unread message


Oct 28, 2013, 4:00:57 PM10/28/13

I am trying to run for the first time. It says 'running OK', but then gives some error, as listed below. Also, can you point out how to 'verify' the output? I mean if the code does generate a transmitted file, then how can I verify its contents?

Thanks a lot,

aman@aman-VirtualBox:~/Downloads/aloe-2.0-rc1/app_dl$ sudo ../build/rtdal_lnx/runcf ./ ./config

ALOE++-1.0-rc1  Copyright (C) 2013
This program comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions. See license.txt

Loading VOLK library...
Using Volk machine: ssse3_32
-- Pipeline Scheduling Selected --
Time slot: 10000 us
Platform: 1 cores
Timer: Single

List of commands:
<l> Load waveform
<i> Set INIT
<r> Set RUN
<p> Set PAUSE
<t> Set STEP
<s> Stop waveform
<m> Set waveform mode
<e> View execution time

<Ctr+C> Exit

>> l
Computing mapping for 56 modules...
Platform model. 1 processsors: C[0]=10000, 
Done. 0.62 GOPTS
Loading 56 modules........................................................OK!

>> i
Initiating 56 modules........................................................OK!

>> r


-------- LTE Setup ---------
 BW: 6 PRB
 CFI: 1
 Ports: 1
 CP mode: Normal
 Cell ID: 0
 PDCCH #0: 2 CE, 18 REGs

 --- PDSCH #0: --- 
 Num RBG: 6 RBG
 RBG_Mask: 0x3f
 TBS size: 1096 bits
 MCS: 9
 Avg bitrate: 0.99 Mbps
 Avg SE: 1.35 bits/symbol
 Modulation: QPSK

Module resmapp returned error from work() function. Trying a clean stop
Waveform ./ was cleanly removed from the system.


Ismael Gomez

Oct 28, 2013, 5:02:25 PM10/28/13

I see in your command line the word "VirtualBox". Does it mean you are using a virtualbox to run aloe? That can be problematic, because aloe can not run in real-time. If you just want to do some tests, try increasing the time slot length in the "config" file, maybe to 100 ms. 

The writes the output signal to the file "output.bin", created in the directory from where you call the runcf program. You can see this in the sink module in the waveform description file ( 

You can verify the signal using matlab or the application. 

Thank you for sharing your experience with aloe, don't hesitate to ask if you find new problems.

Good luck, 

2013/10/28 aman <>

You received this message because you are subscribed to the Google Groups "ALOE" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
To post to this group, send email to
To view this discussion on the web visit
For more options, visit


Oct 28, 2013, 5:32:52 PM10/28/13

Thanks for the prompt reply!
I will install Linux separately on my machine. Meanwhile, I tried running the tx file with 100ms time slot length, it did not give any error, nor generate any output. After displaying the LTE Setup parameters and PDSCH #0, it just stayed there like forever. 

Can you also elaborate how I can verify the output in Matlab? Also, I was looking at the source file for equalizer, is there a way I can work on the Matlab equivalent first?

Thanks again,

Vuk Marojevic

Oct 28, 2013, 6:30:54 PM10/28/13
Hello Aman,

(1) if you execute the LTE transmitter that writes the output to a file (, you can read and process that file in Matlab. You can add noise, for instance, to the samples stored in the file and run the OSLD LTE receiver ( and observe its performance.

(2) Yo can find several OSLD modules and partial LTE processing chains implemented in Matlab. Take a look at the Matlab folder and see what is available and how it works.
I encourage developing and testing the new equalizer in Matlab. Once it works, copy the algorithm the ALOE template and modify as necessary to adapt to the C language. Use the instructions given on the web:

Once you have successfully compiled and linked your DSP module, you'll get three targets:
- standalone executable, 
- Matlab MEX-file, and 
- ALOE module. 

You can test the MEX-file against your reference Matlab function. If this works, integrate it in the ALOE processing chain (.app).

(3) Now you can go back to (1), apply the same noise to the transmitter output file (or use the old file modified with Matlab) and run your receiver featuring the new equalizer and see if it performs better than the old.

You'll find all information on GitHub. If you have problems or questions, don't hesitate to ask.

Reply all
Reply to author
0 new messages