model.xml

<project name='project_01' pubsub='auto' threads='1' use-tagged-token='true'> 
<description> 
	This example is a basic demonstration of using the Cepstrum Transform (Cepstrum)
	algorithm. A cepstrum results from taking the inverse Fourier transform of the
	logarithm of the estimated spectrum of a signal. This application is often used
	in speech analysis, specifically with voice detection. Two variations are supported:
	a complex cepstrum and a real cepstrum. 

	This project contains a single continuous query consisting of the following: 
	1.	A source window that receives the data to be analyzed 
	2. 	A calculate window that uses the cepstrum algorithm to compute the cepstrum
</description>
<contqueries>
  <contquery name='cq_01' include-singletons='true' trace='w_calculate'>
   <windows>
    <window-source name='w_source' insert-only='true' autogen-key='true' index='pi_EMPTY'> 
    <description> 
	This source window receives the included example input data file via a file socket 
	connector. The input data stream is placed into two fields for each observation:
	a key called datetime, and a signal x.
    </description>
     <schema>
      <fields>
       <field name='datetime'  type='int64' key='true'/>
       <field name='x' type='double'/>
      </fields>
     </schema>
     <connectors>
      <connector class='fs' name='publisher'>
       <properties>
        <property name='type'>pub</property>
        <property name='fstype'>csv</property>
        <property name='fsname'>input.csv</property>
        <property name='transactional'>true</property>
        <property name='blocksize'>1</property>
        <property name='blocksize'>1</property>
       </properties>
      </connector>
     </connectors>
    </window-source>
    <window-calculate name="w_calculate" algorithm="Cepstrum" index='pi_EMPTY'> 
    <description> 
	This calculate window receives data events from the source window. It computes and
	publishes the computed cepstrum of the data. For this example, the following parameters
	govern the Cepstrum algorithm: 

	windowLength:		Specifies the length of the sliding window. The value you specify 
				must be greater than the value you specify for overlap. 
	windowType:		Specifies the type of the window. Options: 1=Bartlett, 2=Bohman,
				3=Chebyshev, 4=Gaussian, 5=Kaiser, 6=Parzen, 7=Rectangular, 
				10=Tukey, 11=Bartlett-Hann, 12=Blackman-Harris, 13=Blackman, 
				14=Hamming, 15=Hanning, 16=Flat Top. 
	windowParam:		Some window types require an additional parameter. If not required
				 for the window type selected, this value is ignored. 
	overlap:		Overlap between consecutive windows. Must be strictly less than 
				the value of windowLength.
	complexCepstrum:	When set to 1, the complex cepstrum is calculated. When set to 0 or
				unspecified, the real cepstrum is calculated. 

	Additionally the following input-map and output-map properties define this calculate window: 
	
	input:			Name of analysis variable in the input stream. 
	timeId:			Name of the time ID variable in the input stream. 

	timeIdOut:		Name of time ID variable in output stream. 
	cepstrumListOut:	List of cepstrum variables in the output stream. Cepstrum variable
				values are always real numbers. The algorithm writes windowLength
				values for each output event.  
	
	The resulting output is written to the file result.out via a file socket connector. 
    </description>
      <schema>
          <fields>
          <field name='datetime' type='int64' key='true'/>
          <field name='y' type='array(dbl)'/>
        </fields>
      </schema>
        <parameters>
          <properties>
           <property name="windowLength">128</property>
           <property name="windowType">1</property>
           <property name="windowParam">-1.0</property>
           <property name="overlap">0</property>
           <property name="complexCepstrum">1</property>
          </properties>
        </parameters>
        <input-map>
          <properties>
           <property name="input">x</property>
           <property name="timeId">datetime</property>
          </properties>
        </input-map>
        <output-map>
          <properties>
           <property name="timeIdOut">datetime</property>
           <property name="cepstrumListOut">y[1-128]</property>
          </properties>
        </output-map>
     <connectors>
      <connector class='fs' name='sub'>
       <properties>
        <property name='type'>sub</property>
        <property name='fstype'>csv</property>
        <property name='fsname'>result.out</property>
        <property name='snapshot'>true</property> 
        <property name='header'>true</property>
       </properties>
      </connector>
     </connectors>
    </window-calculate>
   </windows>
   <edges>
    <edge source='w_source' target='w_calculate' role='data'/>
   </edges>
  </contquery>
</contqueries>
</project>