public string Classify(double lastTimestamp)
{
string predicted_class = null;
//attempt to generate a feature vector
if (Extractor.GenerateFeatureVector(lastTimestamp))
{
Instance newinstance = new Instance(instances.numAttributes());
newinstance.Dataset = instances;
for (int i = 0; (i < this.features.Count); i++)
newinstance.setValue(instances.attribute(i), Extractor.Features[this.featureIndex[i]]);
double predicted = classifier.classifyInstance(newinstance);
predicted_class = newinstance.dataset().classAttribute().value_Renamed((int)predicted);
}
return predicted_class;
}
/// <summary> Convert a single instance over. The converted instance is
/// added to the end of the output queue.
///
/// </summary>
/// <param name="instance">the instance to convert
/// </param>
private void convertInstance(Instance instance)
{
Instance inst = null;
if (instance is SparseInstance)
{
double[] vals = new double[instance.numValues()];
int[] indices = new int[instance.numValues()];
int num = 0;
for (int j = 0; j < instance.numValues(); j++)
{
if (instance.isMissingSparse(j) && (getInputFormat().classIndex() != instance.index(j)) && (instance.attributeSparse(j).Nominal || instance.attributeSparse(j).Numeric))
{
if (m_ModesAndMeans[instance.index(j)] != 0.0)
{
vals[num] = m_ModesAndMeans[instance.index(j)];
indices[num] = instance.index(j);
num++;
}
}
else
{
vals[num] = instance.valueSparse(j);
indices[num] = instance.index(j);
num++;
}
}
if (num == instance.numValues())
{
inst = new SparseInstance(instance.weight(), vals, indices, instance.numAttributes());
}
else
{
double[] tempVals = new double[num];
int[] tempInd = new int[num];
Array.Copy(vals, 0, tempVals, 0, num);
Array.Copy(indices, 0, tempInd, 0, num);
inst = new SparseInstance(instance.weight(), tempVals, tempInd, instance.numAttributes());
}
}
else
{
double[] vals = new double[getInputFormat().numAttributes()];
for (int j = 0; j < instance.numAttributes(); j++)
{
if (instance.isMissing(j) && (getInputFormat().classIndex() != j) && (getInputFormat().attribute(j).Nominal || getInputFormat().attribute(j).Numeric))
{
vals[j] = m_ModesAndMeans[j];
}
else
{
vals[j] = instance.value_Renamed(j);
}
}
inst = new Instance(instance.weight(), vals);
}
inst.Dataset = instance.dataset();
push(inst);
}
/// <summary> Convert a single instance over if the class is nominal. The converted
/// instance is added to the end of the output queue.
///
/// </summary>
/// <param name="instance">the instance to convert
/// </param>
private void convertInstance(Instance instance)
{
double[] vals = new double[outputFormatPeek().numAttributes()];
int attSoFar = 0;
for (int j = 0; j < getInputFormat().numAttributes(); j++)
{
weka.core.Attribute att = getInputFormat().attribute(j);
if (!att.Nominal || (j == getInputFormat().classIndex()) || !m_Columns.isInRange(j))
{
vals[attSoFar] = instance.value_Renamed(j);
attSoFar++;
}
else
{
if (att.numValues() <= 2)
{
vals[attSoFar] = instance.value_Renamed(j);
attSoFar++;
}
else
{
if (instance.isMissing(j))
{
for (int k = 0; k < att.numValues(); k++)
{
vals[attSoFar + k] = instance.value_Renamed(j);
}
}
else
{
for (int k = 0; k < att.numValues(); k++)
{
//UPGRADE_WARNING: Data types in Visual C# might be different. Verify the accuracy of narrowing conversions. "ms-help://MS.VSCC.v80/dv_commoner/local/redirect.htm?index='!DefaultContextWindowIndex'&keyword='jlca1042'"
if (k == (int) instance.value_Renamed(j))
{
vals[attSoFar + k] = 1;
}
else
{
vals[attSoFar + k] = 0;
}
}
}
attSoFar += att.numValues();
}
}
}
Instance inst = null;
if (instance is SparseInstance)
{
inst = new SparseInstance(instance.weight(), vals);
}
else
{
inst = new Instance(instance.weight(), vals);
}
copyStringValues(inst, false, instance.dataset(), InputStringIndex, getOutputFormat(), OutputStringIndex);
inst.Dataset = getOutputFormat();
push(inst);
}
/// <summary> Evaluates the classifier on a single instance.
///
/// </summary>
/// <param name="classifier">machine learning classifier
/// </param>
/// <param name="instance">the test instance to be classified
/// </param>
/// <returns> the prediction made by the clasifier
/// </returns>
/// <throws> Exception if model could not be evaluated </throws>
/// <summary> successfully or the data contains string attributes
/// </summary>
public virtual double evaluateModelOnce(Classifier classifier, Instance instance)
{
Instance classMissing = (Instance) instance.copy();
double pred = 0;
classMissing.Dataset = instance.dataset();
classMissing.setClassMissing();
if (m_ClassIsNominal)
{
double[] dist = classifier.distributionForInstance(classMissing);
pred = Utils.maxIndex(dist);
//UPGRADE_WARNING: Data types in Visual C# might be different. Verify the accuracy of narrowing conversions. "ms-help://MS.VSCC.v80/dv_commoner/local/redirect.htm?index='!DefaultContextWindowIndex'&keyword='jlca1042'"
if (dist[(int) pred] <= 0)
{
pred = Instance.missingValue();
}
updateStatsForClassifier(dist, instance);
}
else
{
pred = classifier.classifyInstance(classMissing);
updateStatsForPredictor(pred, instance);
}
return pred;
}
private void readDataTimer_Tick(object sender, EventArgs e)
{
if (okTimer > 3000)
okTimer = -1;
okTimer++;
#if (PocketPC)
#else
if (this.Visible)
this.Visible = false;
#endif
if (isStartedReceiver)
{
count++;
if ((Environment.TickCount - time) >= 1000)
{
ave = sum / (double)count;
//textBoxDist.Text = "R: " + count + " T: " + sum;
aMITesLogger.WriteLogComment("R: " + count + " T: " + sum);
sum = 0;
count = 0;
time = Environment.TickCount;
}
//aMITesDecoder.GetSensorData(this.mitesControllers[0]);
for (int i = 0; (i < this.sensors.TotalReceivers); i++) // FIX SSI
this.mitesDecoders[i].GetSensorData(this.mitesControllers[i]);
for (int i = 1; (i < this.sensors.TotalReceivers); i++)
this.mitesDecoders[0].MergeDataOrderProperly(this.mitesDecoders[i]);
//A training session has started
if (IsTraining==true)
{
if (AutoTraining == true)
{
string current_activity = ((AXML.Label)((AXML.Category)this.annotation.Categories[0]).Labels[autoTrainingIndex]).Name;
if (Extractor.IsTrained(current_activity))
{
this.startActivityTime = Environment.TickCount + Extractor.Configuration.TrainingWaitTime*1000;//Constants.TRAINING_GAP;
autoTrainingIndex++;
//check if auto training completed
if (autoTrainingIndex == ((AXML.Category)this.annotation.Categories[0]).Labels.Count)
{
//this.label5.Text = "TR completed";
//this.label11.Text = "Training Completed";
EndTraining();
this.goodTimer.reset();
this.overallTimer.reset();
}
else
{
((Button)this.categoryButtons[0]).Text = ((AXML.Label)((AXML.Category)this.annotation.Categories[0]).Labels[autoTrainingIndex]).Name;
this.goodTimer.reset();
PlaySound(@"\Windows\Voicbeep", IntPtr.Zero, (int)(PlaySoundFlags.SND_FILENAME | PlaySoundFlags.SND_SYNC));
PlaySound(@"\Windows\Voicbeep", IntPtr.Zero, (int)(PlaySoundFlags.SND_FILENAME | PlaySoundFlags.SND_SYNC));
}
}
else if (this.startActivityTime < Environment.TickCount) // TRAINING_GAP passed
{
// this.label5.Text = "TR in progress";
//this.label11.Text = "Training " + current_activity;
this.goodTimer.start();
double lastTimeStamp = Extractor.StoreMITesWindow();
if (Extractor.GenerateFeatureVector(lastTimeStamp))
{
Extractor.TrainingTime[current_activity] = (int)Extractor.TrainingTime[current_activity] + 200;//Extractor.Configuration.OverlapTime;// get it from configuration
string arffSample = Extractor.toString() + "," + current_activity.Replace(' ', '_');
this.tw.WriteLine(arffSample);
this.label8.Text = Extractor.DiscardedLossRateWindows.ToString();
//this.label10.Text = Extractor.DiscardedConsecutiveLossWindows.ToString();
}
}
else
{
// this.label5.Text = "TR in " + ((int)(this.startActivityTime - Environment.TickCount) / 1000) + " secs";
// this.label11.Text = "Training " + current_activity + " in " + ((int)(this.startActivityTime - Environment.TickCount) / 1000) + " secs";
}
}
else // Manual Training
{
}
}
//Classifying
if (isClassifying==true)
{
double lastTimeStamp = Extractor.StoreMITesWindow();
if (Extractor.GenerateFeatureVector(lastTimeStamp))
{
Instance newinstance = new Instance(instances.numAttributes());
newinstance.Dataset = instances;
for (int i = 0; (i < Extractor.Features.Length); i++)
newinstance.setValue(instances.attribute(i), Extractor.Features[i]);
double predicted = classifier.classifyInstance(newinstance);
string predicted_activity = newinstance.dataset().classAttribute().value_Renamed((int)predicted);
int currentIndex=(int)labelIndex[predicted_activity];
labelCounters[currentIndex] = (int)labelCounters[currentIndex] + 1;
classificationCounter++;
if (classificationCounter == Extractor.Configuration.SmoothWindows)
{
classificationCounter = 0;
int mostCount = 0;
string mostActivity = "";
for (int j=0;(j<labelCounters.Length);j++)
{
if (labelCounters[j] > mostCount)
mostActivity = activityLabels[j];
labelCounters[j] = 0;
}
this.label6.Text = mostActivity;
//this.label11.Text = "Fahd is "+mostActivity;
}
}
}
if (activityCountWindowSize > Extractor.Configuration.QualityWindowSize) //write a line to CSV and initialize
{
DateTime now= DateTime.Now;
DateTime origin = new DateTime(1970, 1, 1, 0, 0, 0, 0);
TimeSpan diff = now.Subtract(origin);
string timestamp = diff.TotalMilliseconds + "," + now.ToString("yyyy'-'MM'-'dd' 'HH':'mm':'ssK");
string master_csv_line = timestamp;
string hr_csv_line = timestamp;
//to restore the date
//DateTime restored = origin.AddMilliseconds(diff.TotalMilliseconds);
if (this.overallTimer.isRunning())
{
foreach (Button button in this.categoryButtons)
master_csv_line += "," + button.Text;
}
else
{
foreach (Button button in this.categoryButtons)
master_csv_line += ",";
}
foreach (Sensor sensor in this.sensors.Sensors)
{
string csv_line1 = timestamp;
string csv_line2 = timestamp;
string csv_line3 = timestamp;
int sensor_id = Convert.ToInt32(sensor.ID);
if (sensor_id > 0) //No HR
{
if (acCounters[sensor_id] > 0)
{
csv_line2 += "," + MITesDataFilterer.MITesPerformanceTracker[sensor_id].LastSamplingRate;
csv_line1 += "," + ((double)(averageX[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00") + ",";
csv_line1 += ((double)(averageY[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00") + ",";
csv_line1 += ((double)(averageZ[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00");
csv_line3 += "," + ((int)(averageRawX[sensor_id] / acCounters[sensor_id])) + ",";
csv_line3 += ((int)(averageRawY[sensor_id] / acCounters[sensor_id])) + ",";
csv_line3 += ((int)(averageRawZ[sensor_id] / acCounters[sensor_id]));
master_csv_line += "," + MITesDataFilterer.MITesPerformanceTracker[sensor_id].LastSamplingRate;
master_csv_line += "," + ((int)(averageRawX[sensor_id] / acCounters[sensor_id])) + ",";
master_csv_line += ((int)(averageRawY[sensor_id] / acCounters[sensor_id])) + ",";
master_csv_line += ((int)(averageRawZ[sensor_id] / acCounters[sensor_id]));
master_csv_line += "," + ((double)(averageX[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00") + ",";
master_csv_line += ((double)(averageY[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00") + ",";
master_csv_line += ((double)(averageZ[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00");
}
else
{
csv_line1 += ",,,,";
csv_line3 += ",,,,";
csv_line2 += ",0";
master_csv_line += ",0,,,,,,";
}
this.activityCountCSVs[sensor_id].WriteLine(csv_line1);
this.samplingCSVs[sensor_id].WriteLine(csv_line2);
this.averagedRaw[sensor_id].WriteLine(csv_line3);
}
averageX[sensor_id] = 0;
averageY[sensor_id] = 0;
averageZ[sensor_id] = 0;
averageRawX[sensor_id] = 0;
averageRawY[sensor_id] = 0;
averageRawZ[sensor_id] = 0;
prevX[sensor_id]=0;
prevY[sensor_id]=0;
prevY[sensor_id]=0;
acCounters[sensor_id] = 0;
}
if (hrCount > 0)
{
this.hrCSV.WriteLine(hr_csv_line+"," + (int)(sumHR / hrCount));
this.masterCSV.WriteLine(master_csv_line + "," + (int)(sumHR / hrCount));
}
else
{
this.hrCSV.WriteLine(hr_csv_line+",");
this.masterCSV.WriteLine(master_csv_line + ",");
}
hrCount = 0;
sumHR = 0;
activityCountWindowSize = 0;
}
activityCountWindowSize += 10; //add 10 milliseconds
//store sum of abs values of consecutive accelerometer readings
for (int i = 0; (i < this.mitesDecoders[0].someMITesDataIndex); i++)
{
if ((this.mitesDecoders[0].someMITesData[i].type != (int)MITesTypes.NOISE) &&
(this.mitesDecoders[0].someMITesData[i].type == (int)MITesTypes.ACCEL))
{
int channel = 0, x = 0, y = 0, z = 0;
channel = (int)this.mitesDecoders[0].someMITesData[i].channel;
x = (int)this.mitesDecoders[0].someMITesData[i].x;
y = (int)this.mitesDecoders[0].someMITesData[i].y;
z = (int)this.mitesDecoders[0].someMITesData[i].z;
// if ((channel != 14) && (channel != 7) && (channel != 8) && (channel != 11) && (channel != 4)
// && (channel != 17) && (channel != 1) && (channel != 0))
// {
// Console.Out.Write("error");
// }
if (channel <= this.sensors.MaximumSensorID) //if junk comes ignore it
{
if ((prevX[channel] > 0) && (prevY[channel] > 0) && (prevZ[channel] > 0) && (x>0) && (y>0) && (z>0))
{
averageX[channel]=averageX[channel]+Math.Abs(prevX[channel]-x);
averageRawX[channel] = averageRawX[channel] + x;
averageY[channel] = averageY[channel] + Math.Abs(prevY[channel] - y);
averageRawY[channel] = averageRawY[channel] + y;
averageZ[channel] = averageZ[channel] + Math.Abs(prevZ[channel] - z);
averageRawZ[channel] = averageRawZ[channel] + z;
acCounters[channel] = acCounters[channel] + 1;
}
prevX[channel] = x;
prevY[channel] = y;
prevZ[channel] = z;
}
}
}
// if (this.isCollectingDetailedData == true)
// aMITesLogger.SaveRawData();
//Collecting Simple plotting data for demos
// for each channel create a file and write to it
//if (this.isCollectingSimpleData == true)
//{
// for (int i = 0; i < aMITesDecoder.someMITesDataIndex; i++)
// {
// if ((aMITesDecoder.someMITesData[i].type != (int)MITesTypes.NOISE) &&
// (aMITesDecoder.someMITesData[i].type == (int)MITesTypes.ACCEL))
// {
// int channel = 0, x = 0, y = 0, z = 0;
// channel = (int)aMITesDecoder.someMITesData[i].channel;
// double unixtimestamp = aMITesDecoder.someMITesData[i].unixTimeStamp;
// if (this.tws[channel] == null)
// {
// xs[channel] = 0;
// ys[channel] = 0;
// zs[channel] = 0;
// this.mitesLastTimeStamps[channel] = unixtimestamp;
// this.mitesSampleCounters[channel] = 0;
// this.tws[channel] = new StreamWriter(this.dataDirectory + "\\MITes_Channel" + channel + ".dat");
// }
// xs[channel] = xs[channel]+ (int)aMITesDecoder.someMITesData[i].x;
// ys[channel] = ys[channel] +(int)aMITesDecoder.someMITesData[i].y;
// zs[channel] = zs[channel] + (int)aMITesDecoder.someMITesData[i].z;
// this.mitesSampleCounters[channel]=this.mitesSampleCounters[channel]+1;
// // if ((unixtimestamp-this.mitesLastTimeStamps[channel])>=1000)
// //{
// this.tws[channel].WriteLine(unixtimestamp + "," +
// (int)(xs[channel] / this.mitesSampleCounters[channel]) + "," +
// (int)(ys[channel] / this.mitesSampleCounters[channel]) + "," +
// (int)(zs[channel] / this.mitesSampleCounters[channel]));
// xs[channel] = 0;
// ys[channel] = 0;
// zs[channel] = 0;
// this.mitesLastTimeStamps[channel]=unixtimestamp;
// System.DateTime dateTime = new System.DateTime(1970, 1, 1, 0, 0, 0, 0);
// dateTime = dateTime.AddMilliseconds(unixtimestamp);
// this.mitesSampleCounters[channel]=0;
// //}
// }
// }
//}
//aMITesLoggerReader.GetSensorData(10, true); // true indicates PLFormat
sum += this.mitesDecoders[0].GetLastByteNum();
aMITesLogger.SaveRawData();
if (flushTimer == 0)
aMITesLogger.FlushBytes();
if (flushTimer > 6000)
flushTimer = -1;
flushTimer++;
aMITesDataFilterer.RemoveZeroNoise();
this.mitesDecoders[0].UpdateSamplingRate(aMITesDataFilterer.CountNonNoise());
if (printSamplingCount > 500)
{
((System.Windows.Forms.Label)this.sensorLabels["SampRate"]).Text = "Samp: " + this.mitesDecoders[0].GetSamplingRate();
//textBoxRate.Text = "Samp: " + aMITesDecoder.GetSamplingRate();
//aMITesLogger.WriteLogComment(textBoxUV.Text);
printSamplingCount = 0;
}
else
printSamplingCount++;
// Check HR values
int hr = aMITesHRAnalyzer.Update();
if (hr > 0)
{
sumHR += hr;
hrCount++;
}
//Compute/get Activity Counts
for (int i = 0; (i < this.sensors.Sensors.Count); i++)
{
int sensor_id = Convert.ToInt32(((SXML.Sensor)this.sensors.Sensors[i]).ID);
if (sensor_id > 0)
((MITesActivityCounter)this.aMITesActivityCounters[sensor_id]).UpdateActivityCounts();
//else if (sensor_id == 0)
// aMITesHRAnalyzer.Update();
}
for (int i = 0; (i < this.sensors.Sensors.Count); i++)
{
int sensor_id = Convert.ToInt32(((SXML.Sensor)this.sensors.Sensors[i]).ID);
if (sensor_id > 0)
((MITesActivityCounter)this.aMITesActivityCounters[sensor_id]).PrintMaxMin();
}
if (((MITesActivityCounter)this.aMITesActivityCounters[this.sensors.FirstAccelerometer]).IsNewEpoch(1000))
{
aMITesHRAnalyzer.ComputeEpoch(30000);
//if (this.tabControl1.TabIndex == 0)
// ReportHR();
for (int i = 0; (i < this.sensors.Sensors.Count); i++)
{
int sensor_id = Convert.ToInt32(((SXML.Sensor)this.sensors.Sensors[i]).ID);
if (sensor_id > 0)
((MITesActivityCounter)this.aMITesActivityCounters[sensor_id]).ComputeEpoch();
}
#if (PocketPC)
if (this.tabControl1.TabIndex == 0)
{
#endif
ReportActivityCounts();
ReportHR();
#if (PocketPC)
}
#endif
if (!isWrittenKey) // Write the key once at the top of the file
{
isWrittenKey = true;
aMITesActivityLogger.StartReportKeyLine();
for (int i = 0; (i < this.sensors.Sensors.Count); i++)
{
int sensor_id = Convert.ToInt32(((SXML.Sensor)this.sensors.Sensors[i]).ID);
if (sensor_id > 0)
aMITesActivityLogger.AddKeyLine(((MITesActivityCounter)this.aMITesActivityCounters[sensor_id]));
}
aMITesActivityLogger.AddKeyLine(aMITesHRAnalyzer);
aMITesActivityLogger.SaveReportKeyLine();
}
aMITesActivityLogger.StartReportLine();
for (int i = 0; (i < this.sensors.Sensors.Count); i++)
{
int sensor_id = Convert.ToInt32(((SXML.Sensor)this.sensors.Sensors[i]).ID);
if (sensor_id > 0)
aMITesActivityLogger.AddReportLine(((MITesActivityCounter)this.aMITesActivityCounters[sensor_id]));
}
aMITesActivityLogger.AddReportLine(aMITesHRAnalyzer);
aMITesActivityLogger.SaveReportLine();
}
// Graph accelerometer data
if (isPlotting)
GraphAccelerometerValues();
stepsToWarning++;
}
}
private void readDataTimer_Tick(object sender, EventArgs e)
{
#region Bluetooth Reconnection Code
#if (PocketPC)
//Start a reconnection thread
foreach (Receiver receiver in this.sensors.Receivers)
{
if (!receiver.Running) //&& (!receiver.Restarting))
{
aMITesActivityLogger.WriteLogComment("Connection broke with " + receiver.ID + ". Restarting");
//reconnectionThreadQuit[receiver.ID] = false;
BluetoothConnector btc=new BluetoothConnector(receiver, this.bluetoothControllers, this.mitesDecoders);
aMITesActivityLogger.WriteLogComment("Initializing reconnection thread");
//ts[receiver.ID] = new Thread(new ThreadStart(btc.Reconnect));
//ts[receiver.ID].Start();
btc.Reconnect();
aMITesActivityLogger.WriteLogComment("Reconnection thread started");
receiver.Restarting = true;
}
else if (receiver.Restarted)
{
aMITesActivityLogger.WriteLogComment("Reconnection completed");
receiver.Restarted = false;
receiver.Restarting = false;
receiver.Running = true;
}
}
#endif
#endregion Bluetooth Reconnection Code
#region Poll All Wockets and MITes and Decode Data
Receiver currentReceiver = null;
try
{
//Poll each CONNECTED receiver channel with the right decoder
foreach (Receiver receiver in this.sensors.Receivers)
{
currentReceiver = receiver;
#if (PocketPC)
if ((receiver.Type == SXML.Constants.RECEIVER_BLUETOOTH) && (receiver.Running == true))
{
if (receiver.Decoder == SXML.Constants.DECODER_MITES)
this.mitesDecoders[receiver.ID].GetSensorData(this.bluetoothControllers[receiver.ID], MITesDecoder.MITES_SENSOR);
else if (receiver.Decoder == SXML.Constants.DECODER_WOCKETS)
this.mitesDecoders[receiver.ID].GetSensorData(this.bluetoothControllers[receiver.ID], MITesDecoder.WOCKETS_SENSOR);
else if (receiver.Decoder == SXML.Constants.DECODER_SPARKFUN)
this.mitesDecoders[receiver.ID].GetSensorData(this.bluetoothControllers[receiver.ID], MITesDecoder.SPARKFUN_SENSOR);
}
else
#endif
if (receiver.Type == SXML.Constants.RECEIVER_USB)
this.mitesDecoders[receiver.ID].GetSensorData(this.mitesControllers[receiver.ID]);
}
if (++ttwcounter == 6000)
{
TextWriter ttw = new StreamWriter("\\Internal Storage\\ts.txt");
ttw.WriteLine(DateTime.Now.ToLongTimeString());
ttw.Close();
ttwcounter = 0;
}
}
//Thrown when there is a Bluetooth failure
//TODO: Make sure no USB failure happening
catch (Exception ex)
{
currentReceiver.Running = false;
//connectionLost = true;
return;
}
//Reset the index of the master decoder and copy all data in order into the master decoder
this.masterDecoder.someMITesDataIndex = 0;
//Only decode running receivers (i.e. with no connection failure)
foreach (Receiver receiver in this.sensors.Receivers)
if (receiver.Running)
this.masterDecoder.MergeDataOrderProperly(this.mitesDecoders[receiver.ID]);
#endregion Poll All Wockets and MITes and Decode Data
#region Poll Builtin Data
#if (PocketPC)
GenericAccelerometerData polledData = null;
if (unprocessedBuiltin == true)
{
this.builtinData.Timestamp = Environment.TickCount;
this.builtinData.Unixtimestamp = UnixTime.GetUnixTime();
polledData = this.builtinData;
}
#endif
#endregion Poll Builtin Data
#region Train in realtime and generate ARFF File
if (IsTraining == true)
{
//We are autotraining each activity after the other
if (AutoTraining == true)
{
//Get current activity to train
string current_activity = ((AXML.Label)((AXML.Category)this.annotation.Categories[0]).Labels[autoTrainingIndex]).Name;
//Check if trained
if (Extractor.IsTrained(current_activity))
{
//store the completed annotation and reset the variables
this.currentRecord.EndDate = DateTime.Now.ToString("yyyy'-'MM'-'dd' 'HH':'mm':'ssK");
this.currentRecord.EndHour = DateTime.Now.Hour;
this.currentRecord.EndMinute = DateTime.Now.Minute;
this.currentRecord.EndSecond = DateTime.Now.Second;
TimeSpan ts = (DateTime.Now - new DateTime(1970, 1, 1, 0, 0, 0));
this.currentRecord.EndUnix = ts.TotalSeconds;
this.annotation.Data.Add(this.currentRecord);
//each time an activity is stopped, rewrite the file on disk, need to backup file to avoid corruption
this.annotation.ToXMLFile();
this.annotation.ToCSVFile();
isExtracting = false;
//Point to the next activity and Calculate the delay to start training
this.startActivityTime = Environment.TickCount + Extractor.Configuration.TrainingWaitTime * 1000;//Constants.TRAINING_GAP;
autoTrainingIndex++;
//If we exceeded the last activity then training is completed
if (autoTrainingIndex == ((AXML.Category)this.annotation.Categories[0]).Labels.Count)
{
this.trainingLabel.Text = "Training Completed";
this.goodTimer.reset();
this.overallTimer.reset();
Thread.Sleep(3000);
EndTraining();
}
else // if there are still activities to train beep twice and reset good timer
{
this.goodTimer.reset();
#if (PocketPC)
PlaySound(@"\Windows\Voicbeep", IntPtr.Zero, (int)(PlaySoundFlags.SND_FILENAME | PlaySoundFlags.SND_SYNC));
PlaySound(@"\Windows\Voicbeep", IntPtr.Zero, (int)(PlaySoundFlags.SND_FILENAME | PlaySoundFlags.SND_SYNC));
#endif
}
}
//if the current activity is not trained and the start time is more that the current time
//then calculate the feature vector
else if (this.startActivityTime < Environment.TickCount) // TRAINING_GAP passed
{
//initialize for extraction
if (isExtracting == false)
{
this.trainingLabel.Text = "Training " + current_activity;
this.goodTimer.start();
//store the current state of the categories
this.currentRecord = new AnnotatedRecord();
this.currentRecord.StartDate = DateTime.Now.ToString("yyyy'-'MM'-'dd' 'HH':'mm':'ssK");
this.currentRecord.StartHour = DateTime.Now.Hour;
this.currentRecord.StartMinute = DateTime.Now.Minute;
this.currentRecord.StartSecond = DateTime.Now.Second;
TimeSpan ts = (DateTime.Now - new DateTime(1970, 1, 1, 0, 0, 0));
this.currentRecord.StartUnix = ts.TotalSeconds;
this.currentRecord.Labels.Add(new AXML.Label(current_activity, "none"));
isExtracting = true;
}
//Extract feature vector from accelerometer data and write an arff line
double lastTimeStamp = Extractor.StoreMITesWindow();
#if (PocketPC)
if ((this.sensors.HasBuiltinSensors) && (polledData != null))
{
//store it in Extractor Buffers as well
lastTimeStamp = Extractor.StoreBuiltinData(polledData);
}
#endif
if (Extractor.GenerateFeatureVector(lastTimeStamp))
{
Extractor.TrainingTime[current_activity] = (int)Extractor.TrainingTime[current_activity] + Extractor.Configuration.OverlapTime;// get it from configuration
string arffSample = Extractor.toString() + "," + current_activity.Replace(' ', '_');
this.tw.WriteLine(arffSample);
this.label8.Text = Extractor.DiscardedLossRateWindows.ToString();
}
}
//if we are waiting for the activity to be trained
else
this.trainingLabel.Text = "Training " + current_activity + " in " + ((int)(this.startActivityTime - Environment.TickCount) / 1000) + " secs";
}
else // Manual Training
{
}
}
#endregion Train in realtime and generate ARFF File
#region Classifying activities
#if (PocketPC)
if (isClassifying == true)
{
double lastTimeStamp = Extractor.StoreMITesWindow();
if ((this.sensors.HasBuiltinSensors) && (polledData != null))
{
//aMITesLoggerPLFormat.SaveRawMITesBuiltinData(polledData);
//store it in Extractor Buffers as well
lastTimeStamp = Extractor.StoreBuiltinData(polledData);
}
if (Extractor.GenerateFeatureVector(lastTimeStamp))
{
Instance newinstance = new Instance(instances.numAttributes());
newinstance.Dataset = instances;
for (int i = 0; (i < Extractor.Features.Length); i++)
newinstance.setValue(instances.attribute(i), Extractor.Features[i]);
double predicted = classifier.classifyInstance(newinstance);
string predicted_activity = newinstance.dataset().classAttribute().value_Renamed((int)predicted);
int currentIndex = (int)labelIndex[predicted_activity];
labelCounters[currentIndex] = (int)labelCounters[currentIndex] + 1;
classificationCounter++;
if (classificationCounter == Extractor.Configuration.SmoothWindows)
{
classificationCounter = 0;
int mostCount = 0;
string mostActivity = "";
for (int j = 0; (j < labelCounters.Length); j++)
{
if (labelCounters[j] > mostCount)
{
mostActivity = activityLabels[j];
mostCount = labelCounters[j];
}
labelCounters[j] = 0;
}
pieChart.SetActivity(mostActivity);
if (this.aList.getEmptyPercent() == 1)
this.aList.reset();
else
this.aList.increment(mostActivity);
if (previousActivity != mostActivity)
{
this.activityTimer.stop();
this.activityTimer.reset();
currentCalories = 0;
}
else
{
if (this.activityTimer.isRunning() == false)
this.activityTimer.start();
}
if (mostActivity == "standing")
{
currentCalories += 1;
totalCalories += 1;
}
else if (mostActivity == "walking")
{
currentCalories += 2;
totalCalories += 2;
}
else if (mostActivity == "brisk-walking")
{
currentCalories += 4;
totalCalories += 4;
}
else
{
currentCalories += 1;
totalCalories += 1;
}
pieChart.SetCalories(totalCalories, currentCalories);
pieChart.Data = this.aList.toPercentHashtable();
pieChart.Invalidate();
previousActivity = mostActivity;
}
}
}
#endif
#endregion Classifying activities
#region Storing CSV data for the grapher (PC Only)
#if (!PocketPC)
if (isCollectingDetailedData == true)
{
if (activityCountWindowSize > Extractor.Configuration.QualityWindowSize) //write a line to CSV and initialize
{
DateTime now = DateTime.Now;
DateTime origin = new DateTime(1970, 1, 1, 0, 0, 0, 0);
TimeSpan diff = now.Subtract(origin);
string timestamp = diff.TotalMilliseconds + "," + now.ToString("yyyy'-'MM'-'dd' 'HH':'mm':'ssK");
string master_csv_line = timestamp;
string hr_csv_line = timestamp;
//to restore the date
//DateTime restored = origin.AddMilliseconds(diff.TotalMilliseconds);
if (this.overallTimer.isRunning())
{
foreach (Button button in this.categoryButtons)
master_csv_line += "," + button.Text;
}
else
{
foreach (Button button in this.categoryButtons)
master_csv_line += ",";
}
foreach (Sensor sensor in this.sensors.Sensors)
{
string csv_line1 = timestamp;
string csv_line2 = timestamp;
string csv_line3 = timestamp;
int sensor_id = Convert.ToInt32(sensor.ID);
if (sensor_id > 0) //No HR
{
if (acCounters[sensor_id] > 0)
{
csv_line2 += "," + MITesDataFilterer.MITesPerformanceTracker[sensor_id].LastSamplingRate;
csv_line1 += "," + ((double)(averageX[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00") + ",";
csv_line1 += ((double)(averageY[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00") + ",";
csv_line1 += ((double)(averageZ[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00");
csv_line3 += "," + ((int)(averageRawX[sensor_id] / acCounters[sensor_id])) + ",";
csv_line3 += ((int)(averageRawY[sensor_id] / acCounters[sensor_id])) + ",";
csv_line3 += ((int)(averageRawZ[sensor_id] / acCounters[sensor_id]));
master_csv_line += "," + MITesDataFilterer.MITesPerformanceTracker[sensor_id].LastSamplingRate;
master_csv_line += "," + ((int)(averageRawX[sensor_id] / acCounters[sensor_id])) + ",";
master_csv_line += ((int)(averageRawY[sensor_id] / acCounters[sensor_id])) + ",";
master_csv_line += ((int)(averageRawZ[sensor_id] / acCounters[sensor_id]));
master_csv_line += "," + ((double)(averageX[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00") + ",";
master_csv_line += ((double)(averageY[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00") + ",";
master_csv_line += ((double)(averageZ[sensor_id] / (double)acCounters[sensor_id])).ToString("00.00");
}
else
{
csv_line1 += ",,,,";
csv_line3 += ",,,,";
csv_line2 += ",0";
master_csv_line += ",0,,,,,,";
}
this.activityCountCSVs[sensor_id].WriteLine(csv_line1);
this.samplingCSVs[sensor_id].WriteLine(csv_line2);
this.averagedRaw[sensor_id].WriteLine(csv_line3);
}
averageX[sensor_id] = 0;
averageY[sensor_id] = 0;
averageZ[sensor_id] = 0;
averageRawX[sensor_id] = 0;
averageRawY[sensor_id] = 0;
averageRawZ[sensor_id] = 0;
prevX[sensor_id] = 0;
prevY[sensor_id] = 0;
prevY[sensor_id] = 0;
acCounters[sensor_id] = 0;
}
if (hrCount > 0)
{
this.hrCSV.WriteLine(hr_csv_line + "," + (int)(sumHR / hrCount));
this.masterCSV.WriteLine(master_csv_line + "," + (int)(sumHR / hrCount));
}
else
{
this.hrCSV.WriteLine(hr_csv_line + ",");
this.masterCSV.WriteLine(master_csv_line + ",");
}
hrCount = 0;
sumHR = 0;
activityCountWindowSize = 0;
}
activityCountWindowSize += 10; //add 10 milliseconds
}
#endif
#endregion Storing CSV data for the grapher (PC Only)
#region Calibration and CSV Calculateion Code
if ((isCalibrating) || (isCollectingDetailedData == true))
{
//store sum of abs values of consecutive accelerometer readings
for (int i = 0; (i < this.masterDecoder.someMITesDataIndex); i++)
{
if ((this.masterDecoder.someMITesData[i].type != (int)MITesTypes.NOISE) &&
(this.masterDecoder.someMITesData[i].type == (int)MITesTypes.ACCEL))
{
int channel = 0, x = 0, y = 0, z = 0;
channel = (int)this.masterDecoder.someMITesData[i].channel;
x = (int)this.masterDecoder.someMITesData[i].x;
y = (int)this.masterDecoder.someMITesData[i].y;
z = (int)this.masterDecoder.someMITesData[i].z;
#region Calibration Calculation
if (isCalibrating)
{
if (this.calSensor == -1)
{
if (this.currentCalibrationSensorIndex < this.sensors.Sensors.Count)
{
this.calX = new int[Constants.CALIBRATION_SAMPLES];
this.calY = new int[Constants.CALIBRATION_SAMPLES];
this.calZ = new int[Constants.CALIBRATION_SAMPLES];
this.calSensor = Convert.ToInt32(((Sensor)this.sensors.Sensors[this.currentCalibrationSensorIndex]).ID);
this.calSensorPosition = Constants.CALIBRATION_FLAT_HORIZONTAL_POSITION;
int receiver_id = Convert.ToInt32(((Sensor)this.sensors.Sensors[this.currentCalibrationSensorIndex]).Receiver);
int[] channels = new int[6];
#if (PockectPC)
if (this.calSensor != PhoneAccelerometers.Constants.BUILT_IN_ACCELEROMETER_CHANNEL_ID)
{
#endif
channels[0] = this.calSensor;
this.mitesControllers[receiver_id].SetChannels(1, channels);
#if (PockectPC)
}
#endif
}
else //all sensors are calibrated
{
TextWriter tw = new StreamWriter("SensorData.xml");
tw.WriteLine(this.sensors.toXML());
tw.Close();
MessageBox.Show("Calibration... Completed");
#if (PocketPC)
Application.Exit();
System.Diagnostics.Process.GetCurrentProcess().Kill();
#else
Environment.Exit(0);
#endif
}
}
if (channel == this.calSensor)
{
if (this.calSensorPosition == Constants.CALIBRATION_FLAT_HORIZONTAL_POSITION)
{
this.calX[this.calCounter] = x;
this.calY[this.calCounter] = y;
this.calCounter++;
}
else //vertical
{
this.calZ[this.calCounter] = z;
this.calCounter++;
}
// if all required samples are collected
if (this.calCounter == Constants.CALIBRATION_SAMPLES)
{
if (this.calSensorPosition == Constants.CALIBRATION_FLAT_HORIZONTAL_POSITION)
{
this.calSensorPosition = Constants.CALIBRATION_SIDE_VERTICAL_POSITION;
double meanX = 0.0, meanY = 0.0;
double stdX = 0.0, stdY = 0.0;
this.label17.Text = "Calibration Completed! Please place the sensor " + ((SXML.Sensor)this.sensors.Sensors[this.currentCalibrationSensorIndex]).ID + " vertical on a flat surface then click start.";
this.pictureBox2.Image = this.verticalMITes;
this.button2.Enabled = true;
for (int j = 0; (j < this.calCounter); j++)
{
meanX += (double)this.calX[j];
meanY += (double)this.calY[j];
}
meanX = meanX / this.calCounter;
meanY = meanY / this.calCounter;
((SXML.Sensor)this.sensors.Sensors[this.currentCalibrationSensorIndex]).XMean = meanX;
((SXML.Sensor)this.sensors.Sensors[this.currentCalibrationSensorIndex]).YMean = meanY;
for (int j = 0; (j < this.calCounter); j++)
{
stdX += Math.Pow(this.calX[j] - meanX, 2);
stdY += Math.Pow(this.calY[j] - meanY, 2);
}
stdX = Math.Sqrt(stdX / (this.calCounter - 1));
stdY = Math.Sqrt(stdY / (this.calCounter - 1));
((SXML.Sensor)this.sensors.Sensors[this.currentCalibrationSensorIndex]).XStd = stdX;
((SXML.Sensor)this.sensors.Sensors[this.currentCalibrationSensorIndex]).YStd = stdY;
}
else
{
if (this.currentCalibrationSensorIndex < this.sensors.Sensors.Count)
{
this.label17.Text = "Calibration Completed! Please place the sensor " + ((SXML.Sensor)this.sensors.Sensors[this.currentCalibrationSensorIndex]).ID + " horizontal on a flat surface then click start.";
this.pictureBox2.Image = this.horizontalMITes;
this.button2.Enabled = true;
}
this.calSensorPosition = Constants.CALIBRATION_FLAT_HORIZONTAL_POSITION;
double meanZ = 0.0;
double stdZ = 0.0;
for (int j = 0; (j < this.calCounter); j++)
meanZ += (double)this.calZ[j];
meanZ = meanZ / this.calCounter;
((SXML.Sensor)this.sensors.Sensors[this.currentCalibrationSensorIndex]).ZMean = meanZ;
for (int j = 0; (j < this.calCounter); j++)
stdZ += Math.Pow(this.calZ[j] - meanZ, 2);
stdZ = Math.Sqrt(stdZ / (this.calCounter - 1));
((SXML.Sensor)this.sensors.Sensors[this.currentCalibrationSensorIndex]).ZStd = stdZ;
this.currentCalibrationSensorIndex++;
this.calSensor = -1;
}
this.isCalibrating = false;
this.calCounter = 0;
}
}
}
#endregion Calibration Calculation
#region CSV values calculation (PC Only)
#if (!PocketPC)
else if (isCollectingDetailedData)
{
if (channel <= this.sensors.MaximumSensorID) //if junk comes ignore it
{
if ((prevX[channel] > 0) && (prevY[channel] > 0) && (prevZ[channel] > 0) && (x > 0) && (y > 0) && (z > 0))
{
averageX[channel] = averageX[channel] + Math.Abs(prevX[channel] - x);
averageRawX[channel] = averageRawX[channel] + x;
averageY[channel] = averageY[channel] + Math.Abs(prevY[channel] - y);
averageRawY[channel] = averageRawY[channel] + y;
averageZ[channel] = averageZ[channel] + Math.Abs(prevZ[channel] - z);
averageRawZ[channel] = averageRawZ[channel] + z;
acCounters[channel] = acCounters[channel] + 1;
}
prevX[channel] = x;
prevY[channel] = y;
prevZ[channel] = z;
}
}
#endif
#endregion CSV values calculation (PC Only)
}
}
}
#endregion Calibration and CSV Calculateion Code
//Remove data with any values =0 or 1022
aMITesDataFilterer.RemoveZeroNoise();
this.masterDecoder.UpdateSamplingRate(aMITesDataFilterer.CountNonNoise());
if (printSamplingCount > 500)
{
//((System.Windows.Forms.Label)this.sensorLabels["SampRate"]).Text = "Samp: " + this.mitesDecoders[0].GetSamplingRate();
((System.Windows.Forms.Label)this.sensorLabels["SampRate"]).Text = "Samp: ttt";//+this.masterDecoder.GetSamplingRate();
//textBoxRate.Text = "Samp: " + aMITesDecoder.GetSamplingRate();
//aMITesLogger.WriteLogComment(textBoxUV.Text);
printSamplingCount = 0;
}
else
printSamplingCount++;
// Check HR values
int hr = aMITesHRAnalyzer.Update();
#if (PocketPC)
#else
if (hr > 0)
{
sumHR += hr;
hrCount++;
}
#endif
//Compute/get Activity Counts
for (int i = 0; (i < this.sensors.Sensors.Count); i++)
{
int sensor_id = Convert.ToInt32(((SXML.Sensor)this.sensors.Sensors[i]).ID);
if (sensor_id > 0)
((MITesActivityCounter)this.aMITesActivityCounters[sensor_id]).UpdateActivityCounts();
//else if (sensor_id == 0)
// aMITesHRAnalyzer.Update();
}
for (int i = 0; (i < this.sensors.Sensors.Count); i++)
{
int sensor_id = Convert.ToInt32(((SXML.Sensor)this.sensors.Sensors[i]).ID);
if (sensor_id > 0)
((MITesActivityCounter)this.aMITesActivityCounters[sensor_id]).PrintMaxMin();
}
if (((MITesActivityCounter)this.aMITesActivityCounters[this.sensors.FirstAccelerometer]).IsNewEpoch(1000))
{
aMITesHRAnalyzer.ComputeEpoch(30000);
//if (this.tabControl1.TabIndex == 0)
// ReportHR();
for (int i = 0; (i < this.sensors.Sensors.Count); i++)
{
int sensor_id = Convert.ToInt32(((SXML.Sensor)this.sensors.Sensors[i]).ID);
if (sensor_id > 0)
((MITesActivityCounter)this.aMITesActivityCounters[sensor_id]).ComputeEpoch();
}
if (!isWrittenKey) // Write the key once at the top of the file
{
isWrittenKey = true;
aMITesActivityLogger.StartReportKeyLine();
for (int i = 0; (i < this.sensors.Sensors.Count); i++)
{
int sensor_id = Convert.ToInt32(((SXML.Sensor)this.sensors.Sensors[i]).ID);
if (sensor_id > 0)
aMITesActivityLogger.AddKeyLine(((MITesActivityCounter)this.aMITesActivityCounters[sensor_id]));
}
aMITesActivityLogger.AddKeyLine(aMITesHRAnalyzer);
aMITesActivityLogger.SaveReportKeyLine();
}
aMITesActivityLogger.StartReportLine();
for (int i = 0; (i < this.sensors.Sensors.Count); i++)
{
int sensor_id = Convert.ToInt32(((SXML.Sensor)this.sensors.Sensors[i]).ID);
if (sensor_id > 0)
aMITesActivityLogger.AddReportLine(((MITesActivityCounter)this.aMITesActivityCounters[sensor_id]));
}
aMITesActivityLogger.AddReportLine(aMITesHRAnalyzer);
aMITesActivityLogger.SaveReportLine();
}
#region Store the sensor data
if (flushTimer == 0)
aMITesLoggerPLFormat.FlushBytes();
if (flushTimer > FLUSH_TIMER_MAX)
flushTimer = -1;
flushTimer++;
#endregion Store the sensor data
// Graph accelerometer data for multiple recievers
#if (PocketPC)
if (isPlotting)
GraphAccelerometerValues(polledData);
#endif
#if (PocketPC)
if (polledData != null)
{
this.builtinData = null;
this.unprocessedBuiltin = false;
}
#endif
}
