Matlab

Download the following libraries into a folder of your system:

Apache XML-RPC
Apache Jakarta Commons
SDAS Core Libraries
SDAS Client

Since the server at baco computer uses an older version, if you are planing on accessing it, you should download this version of SDAS Core Libraries and Client:

SDAS Core Libraries
SDAS Client

Add all of the previous libraries to your system classpath.
Use the matlab static path. More information at the matlab documentation site.

You should use the last version (at least 1.5) of the Sun Java.
First you have to find out where is your java home located. To avoid errors, download and run this utility. The value returned by the utility is the MATLAB_JAVA value.

Now you have to set the MATLAB_JAVA as a system variable: 

Open the windows System Properties (right - click on My Computer or go to the Control Panel)
Select the tab Advanced
Click on Environment Variables...
In the system variables click New...
The Variable name is: MATLAB_JAVA
In the value field (supposing you have the java_home in the libraries in C:\Program Files\java\jdk1.5.0_04\jre) enter the following value: C:\Program Files\java\jdk1.5.0_04\jre

Supposing you have the java_home in /opt/jdk1.5.0_04/jre/ :

export MATLAB_JAVA="/opt/jdk1.5.0_04/jre/"

NOTE: Each time you change system properties, you have to restart MatLab

Before starting to use the system check if JAVA is well configured in MatLab, type in the matlab console:

version -java

If you get an answer like: Java 1.5.0_04 with Sun Microsystems Inc. Java HotSpot(TM) Server VM, then you’re ready to start. Otherwise check all the previous steps.

import org.sdas.core.client.*;
import org.sdas.core.time.*;
client = SDASClient('baco.ipfn.ist.utl.pt', 8888);

If you get an error check the classpath.

found = client.searchDeclaredEventsByName('S');

found = client.searchDeclaredEventsByName('S');

found = client.searchDeclaredEventsByName('SHOT', 'pt');

found = client.searchDeclaredEventsByUniqueID('SHOT', 'pt');

found = client.searchDeclaredEventsByDescription('SHOT');

found = client.searchDeclaredEventsByDescription('SHOT', 'pt');

for i=1:1:size(found)
    found(i)
end

max = client.searchMaxEventNumber('0x0000')

min = client.searchMinEventNumber('0x0000')

NOTE: You can construct time with a resolution of picosseconds, just add to the example values for millis, micros, nanos and picos
NOTE 2: Date constructors have the months index to 0 (January is 0 and December is 11)

Search events in December 2005:

date_start = Date(2005, 11, 1);
date_end = Date(2005, 11, 31);
tstart = TimeStamp(date_start);
tend = TimeStamp(date_end);
eventsFound = client.searchEventsByEventTimeWindow(tstart, tend);
for i = 1:1:size(eventsFound)
    eventsFound(i)
end

Search events in the 22 December 2005 between 5pm and 6pm:

date_start = Date(2005, 11, 22);
date_end = Date(2005,11,22);
time_start = Time(17, 0, 0);
time_end = Time(18, 0, 0);
tstart = TimeStamp(date_start, time_start);
tend = TimeStamp(date_end, time_end);
eventsFound = client.searchEventsByEventTimeWindow(tstart, tend);
for i = 1:1:size(eventsFound)
    eventsFound(i)
end

parametersFound = client.searchParametersByName('DENS');

parametersFound = client.searchParametersByName('DENS', 'pt');

parametersFound = client.searchParametersByUniqueID('DENS');

parametersFound = client.searchParametersByDescription('current');

parametersFound = client.searchParametersByDescription('corrente', 'pt');

for i = 1:1:size(parametersFound)
    parametersFound(i)
end

This function returns the parameters unique identifiers where the data isn’t null for the selected event:

dataFound = client.searchDataByEvent('0x0000', 17898);
for i = 1:1:size(dataFound)
    dataFound (i)
end

NOTE: The unique identifiers are CASE-SENSITIVE

Data for only one parameter

dataStruct=client.getData('POST.PROCESSED.DENSITY','0x0000', 17898)
dataStruct=dataStruct(1);

Data for several parameters in the same event

dataStruct=client.getMultipleData({'POST.PROCESSED.DENSITY', 'POST.PROCESSED.IPLASMA'},'0x0000', 17898)
dataStructDens=dataStruct(1,1);
dataStructIP=dataStruct(2,1);
dens=dataStructDens.getData(); 
ip=dataStructIP.getData(); 

Data for several parameters in different events

dataStruct=client.getMultipleData({'POST.PROCESSED.DENSITY', 'POST.PROCESSED.IPLASMA'},{'0x0000','0x0000'}, [17898,17899])
dataStructDens=dataStruct(1,1);
dataStructIP=dataStruct(2,1);
dens=dataStructDens.getData(); 
ip=dataStructIP.getData(); 

Data for the same parameter in different events

dataStruct=client.getMultipleData('POST.PROCESSED.DENSITY',{'0x0000','0x0000'}, [17898,17899])
dataStructDens=dataStruct(1,1);
dataStructIP=dataStruct(2,1);
dens=dataStructDens.getData(); 
ip=dataStructIP.getData(); 

Data for the same parameter in different event numbers

dataStruct=client.getMultipleData('POST.PROCESSED.DENSITY', '0x0000', [17898,17899])

This data structure gives you information about:

• start time
• end time
• time of the event
• mime_type
• the parameter unique identifier

The following example shows how to calculate and plot the density at ISTTOK:
1. The data

dataStruct=client.getMultipleData({'CENTRAL.OS9_ADC_VME_I8.IF0CS',
'CENTRAL.OS9_ADC_VME_I8.IF0SN'},'0x0000', 11244);
cosine = dataStruct(1,1).getData;
isine = dataStruct(2,1).getData;

2. Calculate the phase

len = length(cosine)
cosavg = cosine - max(movavg(cosine, 10, 10))/2 - min(movavg(cosine, 10, 10))/2;
sinavg = isine - max(movavg(isine, 10, 10))/2 - min(movavg(isine, 10, 10))/2;
for j = 1:len
  phase(j) = atan2(double(sinavg(j)), double(cosavg(j)));
end

3. Unwrap

unwraped = unwrap(phase);

4. The start time

tstart = dataStruct(1,1).getTStart

5. The end time

tend = dataStruct(1,1).getTEnd

6. The time between samples is:

tbs= (tend.getTimeInMicros - tstart.getTimeInMicros)/len

7. The events

events = dataStruct(1,1).getEvents;

8. The event time (I’m assuming the event I want is at the index 0, but I should check first...)

tevent = events(1,1).getTimeStamp

9. The delay of the start time relative to the event time

delay = tstart.getTimeInMicros - tevent.getTimeInMicros

10. Create the time array

times = delay:tbs:delay+tbs*(len-1);

11. Normalize the data

dens = -7e17 * unwraped;
thold = dens(len-100:len);
density = dens-mean(thold);

11. Plot the chart

plot(times, density)