/*
* This file is part of GNUnet
- * (C) 2013 Christian Grothoff (and other contributing authors)
+ * Copyright (C) 2013 Christian Grothoff (and other contributing authors)
*
* GNUnet is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published
*
* You should have received a copy of the GNU General Public License
* along with GNUnet; see the file COPYING. If not, write to the
- * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
- * Boston, MA 02111-1307, USA.
+ * Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+ * Boston, MA 02110-1301, USA.
*/
-/*
+/**
* @file sensor/plugin_sensor_model_gaussian.c
* @brief Gaussian model for sensor analysis
* @author Omar Tarabai
#define LOG(kind,...) GNUNET_log_from (kind, "sensor-model-gaussian", __VA_ARGS__)
-/*
+/**
* Plugin state information
*/
struct Plugin
{
- /*
+ /**
* Configuration handle
*/
const struct GNUNET_CONFIGURATION_Handle *cfg;
+ /**
+ * Number of initial readings to be used for training only
+ */
+ int training_window;
+
+ /**
+ * Number of standard deviations considered within "normal"
+ */
+ int confidence_interval;
+
+ /**
+ * Increase in weight with each reading
+ */
+ float weight_inc;
+
};
-/*
+/**
+ * State of single model instance
+ */
+struct Model
+{
+
+ /**
+ * Pointer to the plugin state
+ */
+ struct Plugin *plugin;
+
+ /**
+ * Gaussian sums
+ */
+ long double s[3];
+
+ /**
+ * Number of readings so far
+ */
+ int n;
+
+ /**
+ * Weight to be used for the next reading
+ */
+ double w;
+
+};
+
+/**
+ * Update local sums of model with a new value.
+ *
+ * @param model Targe model
+ * @param val New value
+ */
+static void
+update_sums (struct Model *model, double val)
+{
+ int i;
+
+ for (i = 0; i < 3; i++)
+ model->s[i] += model->w * pow (val, (double) i);
+ model->w += model->plugin->weight_inc;
+ model->n++;
+}
+
+
+/**
+ * Feed a new value to a model
+ *
+ * @param cls closure (model state)
+ * @param val value to be fed to the model
+ * @return #GNUNET_YES in case of a detected outlier, #GNUNET_NO otherwise
+ */
+static int
+sensor_gaussian_model_feed (void *cls, double val)
+{
+ struct Model *model = cls;
+ struct Plugin *plugin = model->plugin;
+ long double mean;
+ long double stddev;
+ long double allowed_variance;
+
+ if (model->n < plugin->training_window)
+ {
+ update_sums (model, val);
+ return GNUNET_NO;
+ }
+ if (model->n == plugin->training_window)
+ LOG (GNUNET_ERROR_TYPE_DEBUG, "Gaussian model out of training period.\n");
+ mean = model->s[1] / model->s[0];
+ stddev =
+ (model->s[0] * model->s[2] -
+ model->s[1] * model->s[1]) / (model->s[0] * (model->s[0] - 1));
+ if (stddev < 0) /* Value can be slightly less than 0 due to rounding errors */
+ stddev = 0;
+ stddev = sqrt (stddev);
+ allowed_variance = (plugin->confidence_interval * stddev);
+ if ((val < (mean - allowed_variance)) || (val > (mean + allowed_variance)))
+ return GNUNET_YES;
+ update_sums (model, val);
+ return GNUNET_NO;
+}
+
+
+/**
+ * Destroy a model instance
+ *
+ * @param cls closure (model state)
+ */
+static void
+sensor_gaussian_model_destroy_model (void *cls)
+{
+ struct Model *model = cls;
+
+ GNUNET_free (model);
+}
+
+
+/**
+ * Create a model instance
+ *
+ * @param cls closure (plugin state)
+ * @return model state to be used for later calls
+ */
+static void *
+sensor_gaussian_model_create_model (void *cls)
+{
+ struct Plugin *plugin = cls;
+ struct Model *model;
+
+ model = GNUNET_new (struct Model);
+
+ model->plugin = plugin;
+ model->w = 1;
+ return model;
+}
+
+
+/**
* Entry point for the plugin.
*
* @param cls The struct GNUNET_CONFIGURATION_Handle.
static struct Plugin plugin;
const struct GNUNET_CONFIGURATION_Handle *cfg = cls;
struct GNUNET_SENSOR_ModelFunctions *api;
+ unsigned long long num;
if (NULL != plugin.cfg)
return NULL; /* can only initialize once! */
memset (&plugin, 0, sizeof (struct Plugin));
plugin.cfg = cfg;
+ if (GNUNET_OK !=
+ GNUNET_CONFIGURATION_get_value_number (cfg, "sensor-model-gaussian",
+ "TRAINING_WINDOW", &num))
+ {
+ LOG (GNUNET_ERROR_TYPE_ERROR,
+ _("Missing `TRAINING_WINDOW' value in configuration.\n"));
+ return NULL;
+ }
+ if (num < 1)
+ {
+ LOG (GNUNET_ERROR_TYPE_WARNING,
+ "Minimum training window invalid (<1), setting to 1.\n");
+ plugin.training_window = 1;
+ }
+ else
+ {
+ plugin.training_window = (int) num;
+ }
+ if (GNUNET_OK !=
+ GNUNET_CONFIGURATION_get_value_number (cfg, "sensor-model-gaussian",
+ "CONFIDENCE_INTERVAL", &num))
+ {
+ LOG (GNUNET_ERROR_TYPE_ERROR,
+ _("Missing `CONFIDENCE_INTERVAL' value in configuration.\n"));
+ return NULL;
+ }
+ if (GNUNET_OK !=
+ GNUNET_CONFIGURATION_get_value_float (cfg, "sensor-model-gaussian",
+ "WEIGHT_INC", &plugin.weight_inc))
+ {
+ LOG (GNUNET_ERROR_TYPE_ERROR,
+ _("Missing `WEIGHT_INC' value in configuration.\n"));
+ return NULL;
+ }
+ plugin.confidence_interval = (int) num;
api = GNUNET_new (struct GNUNET_SENSOR_ModelFunctions);
+
api->cls = &plugin;
- LOG(GNUNET_ERROR_TYPE_DEBUG, "Guassian model plugin is running\n");
+ api->create_model = &sensor_gaussian_model_create_model;
+ api->destroy_model = &sensor_gaussian_model_destroy_model;
+ api->feed_model = &sensor_gaussian_model_feed;
+ LOG (GNUNET_ERROR_TYPE_DEBUG, "Gaussian model plugin is running.\n");
return api;
}
-/*
+
+/**
* Exit point from the plugin.
*
* @param cls The plugin context (as returned by "init")
projects / oweals / gnunet.git / blobdiff