WF_analysis-L12.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "342f3100-20a9-48d5-abec-59b11884c9dd",
   "metadata": {},
   "source": [
    "### Using NFStream to process the PCAP files to have flow records in CSV file  (active timeout =41 , decomposition wavelet levels = 12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8de14c96-2940-4839-bfe1-f665e6841542",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting NFStream v6.5.4b\n",
      "Found 165 .pcap files. Processing...\n",
      "All files processed.\n"
     ]
    }
   ],
   "source": [
    "import nfstream\n",
    "from nfstream import NFStreamer\n",
    "import sys\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import numpy.typing as npt\n",
    "import pywt\n",
    "import os\n",
    "\n",
    "from nfstream import NFPlugin\n",
    "\n",
    "class WFPlugin(NFPlugin):\n",
    "    \n",
    "    def on_init(self, packet, flow):\n",
    "        assert hasattr(self, \"levels\")\n",
    "        assert hasattr(self, \"active_timeout\")\n",
    "\n",
    "        # Pywt requires vector of length 2**level as input\n",
    "        # set nbins to that number:\n",
    "        self.nbins = 2**self.levels\n",
    "\n",
    "        # Given `a` as max flow length calculate size of each bin:\n",
    "        self.bin_size = self.active_timeout / 2**self.levels * 1000\n",
    "\n",
    "        # Reserve a empty vectors for data i forward and backward direction\n",
    "        flow.udps.forward = np.zeros(self.nbins).tolist()\n",
    "        flow.udps.backward = np.zeros(self.nbins).tolist()\n",
    "\n",
    "        # Save first packet timestamp that will be used to calc index\n",
    "        # of time series bin\n",
    "        flow.udps.first_packet_timestamp = packet.time  # timestamp in ms\n",
    "\n",
    "    def on_update(self, packet, flow):\n",
    "        # Calculate time in ms from first packet\n",
    "        mstime_since_first_packet: int = packet.time - flow.udps.first_packet_timestamp\n",
    "\n",
    "        # Calculate index of bin to put data into\n",
    "        ibin, _ = divmod(mstime_since_first_packet, self.bin_size)\n",
    "\n",
    "        # Put transport_size into timeseries depending on direction\n",
    "        if packet.direction == 0:\n",
    "            # src to dest\n",
    "            flow.udps.forward[int(ibin)] += packet.transport_size\n",
    "        else:\n",
    "            # dst to src\n",
    "            flow.udps.backward[int(ibin)] += packet.transport_size\n",
    "\n",
    "    def on_expire(self, flow):\n",
    "        # Calculate and add forward features\n",
    "        f_p_k, f_u_k, f_sigma_k, f_S_k = self.calculate_wavelet_features(\n",
    "            data=flow.udps.forward, level=self.levels\n",
    "        )\n",
    "        self.add_attrs_from_list(flow, f_p_k, \"f_p_k\")\n",
    "        self.add_attrs_from_list(flow, f_u_k, \"f_u_k\")\n",
    "        self.add_attrs_from_list(flow, f_sigma_k, \"f_sigma_k\")\n",
    "        self.add_attrs_from_list(flow, f_S_k, \"f_S_k\")\n",
    "\n",
    "        # Calculate and add backward features\n",
    "        b_p_k, b_u_k, b_sigma_k, b_S_k = self.calculate_wavelet_features(\n",
    "            data=flow.udps.backward, level=self.levels\n",
    "        )\n",
    "        self.add_attrs_from_list(flow, b_p_k, \"b_p_k\")\n",
    "        self.add_attrs_from_list(flow, b_u_k, \"b_u_k\")\n",
    "        self.add_attrs_from_list(flow, b_sigma_k, \"b_sigma_k\")\n",
    "        self.add_attrs_from_list(flow, b_S_k, \"b_S_k\")\n",
    "\n",
    "        # Delete temporary variables\n",
    "        del flow.udps.forward\n",
    "        del flow.udps.backward\n",
    "        del flow.udps.first_packet_timestamp\n",
    "\n",
    "    @staticmethod\n",
    "    def add_attrs_from_list(flow, values, attr):\n",
    "        \"\"\"Adds new attributes to flow.udps from a list of `values`. Name of\n",
    "        attributes are generated by adding `attr` and `number`\n",
    "        \"\"\"\n",
    "        for i, value in enumerate(values):\n",
    "            setattr(flow.udps, f\"{attr}_{i}\", value)\n",
    "\n",
    "    @staticmethod\n",
    "    def calculate_wavelet_features(\n",
    "        data: npt.NDArray[np.float64], level\n",
    "    ) -> tuple[npt.NDArray, npt.NDArray, npt.NDArray, npt.NDArray]:\n",
    "        \"\"\"Calculate wavelet features based on [1].\n",
    "\n",
    "        Args:\n",
    "            data\n",
    "                Data. Will be padded to 2**level.\n",
    "\n",
    "        Returns:\n",
    "            4 vectors of shape: (level, ), these vectors are:\n",
    "            - `p_k` (Relative wavelet energy vector),\n",
    "            - `u_k` (Absolute mean of detail coefficients),\n",
    "            - `sigma_k` (Std. deviation of detail coefficients),\n",
    "            - `S_k` (Shannon entropy)\n",
    "        \"\"\"\n",
    "        data = np.pad(data, (0, max(0, 2**level - len(data))))\n",
    "        coeffs = pywt.wavedec(data, \"haar\", level=level)\n",
    "\n",
    "        # Extract detail coefficients and approximation coefficients\n",
    "        detail_coeffs = coeffs\n",
    "\n",
    "        # Prepare for feature calculation\n",
    "        E_k = np.array([np.sum(np.square(c)) for c in detail_coeffs])\n",
    "        E_total = np.sum(E_k)\n",
    "\n",
    "        # Relative wavelet energy\n",
    "        p_k = E_k / (E_total + 1e-7)\n",
    "\n",
    "        # Calculate Shannon entropy for each level\n",
    "        S_k = np.array([\n",
    "            -np.sum((c**2 / (np.sum(c**2) + 1e-7)) * np.log((c**2 / (np.sum(c**2) + 1e-7)) + 1e-7))\n",
    "            for c in detail_coeffs\n",
    "        ])\n",
    "\n",
    "        # Absolute mean of coefficients\n",
    "        u_k = np.array([np.mean(np.abs(c)) for c in detail_coeffs])\n",
    "\n",
    "        # Standard deviation of coefficients\n",
    "        sigma_k = np.array([np.std(c) for c in detail_coeffs])\n",
    "\n",
    "        return (\n",
    "            p_k,\n",
    "            u_k,\n",
    "            sigma_k,\n",
    "            S_k,\n",
    "        )\n",
    "\n",
    "def process_pcap_NF(pcap_directory: str):\n",
    "    \n",
    "    # Loop through each file in the directory\n",
    "    pcap_files = [f for f in os.listdir(pcap_directory) if f.endswith('.pcap')]\n",
    "    print(f\"Found {len(pcap_files)} .pcap files. Processing...\")\n",
    "\n",
    "    for filename in pcap_files:\n",
    "        path = os.path.join(pcap_directory, filename)\n",
    "\n",
    "        # Processing pcap file with NFStreamer\n",
    "        streamer = nfstream.NFStreamer(source=path,\n",
    "                                       statistical_analysis=True,\n",
    "                                       splt_analysis=255,\n",
    "                                       idle_timeout=1200,\n",
    "                                       active_timeout=41,\n",
    "                                       accounting_mode=2,\n",
    "                                       udps=[WFPlugin(active_timeout=41, levels=12)],\n",
    "                                       n_dissections=20,\n",
    "                                       n_meters=1,\n",
    "        )\n",
    "        os.makedirs(\"VNAT/WF\", exist_ok=True)\n",
    "        output_folder = os.path.join(os.path.dirname(path), \"WF\")\n",
    "\n",
    "        output_csv = os.path.join(output_folder, os.path.splitext(os.path.basename(path))[0] + \".csv\")\n",
    "\n",
    "        streamer.to_csv(output_csv)\n",
    "\n",
    "\n",
    "if __name__ == '__main__':\n",
    "    pcap_directory =  \"VNAT\"\n",
    "    print(f\"Starting NFStream v{nfstream.__version__}\")\n",
    "    process_pcap_NF(pcap_directory)\n",
    "    print(\"All files processed.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a6914a7-ff77-406e-8ae3-809cdd128980",
   "metadata": {},
   "source": [
    "### Merge the CSV files into one large CSV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7a6172ac-f26d-4453-aad0-4e64cb36c482",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "\n",
    "directory = \"VNAT/WF\"\n",
    "\n",
    "dataframes = []\n",
    "\n",
    "for filename in os.listdir(directory):\n",
    "    if filename.endswith('.csv'):  \n",
    "        df = pd.read_csv(os.path.join(directory, filename),low_memory=False)\n",
    "        \n",
    "        if 'nonvpn' in filename.lower():\n",
    "            df['label'] = 'NoVPN'\n",
    "        else:    \n",
    "            df['label'] = 'VPN'\n",
    "        df['source_file']=filename\n",
    "        dataframes.append(df)\n",
    "\n",
    "df_large = pd.concat(dataframes, ignore_index=True)\n",
    "\n",
    "df_large.to_csv(\"VNAT/WF/flows_12.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5e824246-7f2e-49b7-bd36-4fa3ce02acb7",
   "metadata": {},
   "source": [
    "### Read the large CSV in and assign categories based on the source file names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "bd6e8dc0-0a9a-4817-ba21-b958eaa911b6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/26/m1ty4dhn4kd02qvztlq5pdlh0000gp/T/ipykernel_61256/1849199583.py:3: DtypeWarning: Columns (86,87,88) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  df = pd.read_csv(\"VNAT/WF/flows.csv\")\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "category\n",
       "File Transfer        17966\n",
       "Command & Control    17585\n",
       "Chat                 11679\n",
       "Streaming             3494\n",
       "VoIP                  1223\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.read_csv(\"VNAT/WF/flows_12.csv\",low_memory=False)\n",
    "\n",
    "\n",
    "category_mapping = {\n",
    "    'skype': 'Chat',\n",
    "    'vimeo': 'Streaming',\n",
    "    'netflix': 'Streaming',\n",
    "    'youtube': 'Streaming',\n",
    "    'ssh': 'Command & Control',\n",
    "    'rdp': 'Command & Control',\n",
    "    'sftp': 'File Transfer',\n",
    "    'rsync': 'File Transfer',\n",
    "    'scp': 'File Transfer',\n",
    "    'voip': 'VoIP'\n",
    "}\n",
    "\n",
    "def assign_category(source_file):\n",
    "    for keyword, category in category_mapping.items():\n",
    "        if keyword in source_file.lower():\n",
    "            return category\n",
    "    return 'Unknown'  \n",
    "\n",
    "df['category'] = df['source_file'].apply(assign_category)\n",
    "\n",
    "df['category'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "92676b68-65af-4a44-b47d-15e41f639df6",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv('VNAT/WF/flows_extra_labelled_12.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2ae2b339-f377-4c46-a851-1897f36afa75",
   "metadata": {},
   "source": [
    "###  CDF to see how the samples distribute across flow duration and packet count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "8cbd1e56-c3ce-45d5-a70e-3cad2657ed8c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/26/m1ty4dhn4kd02qvztlq5pdlh0000gp/T/ipykernel_61256/1538991870.py:7: DtypeWarning: Columns (86,87,88) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  filtered_df = pd.read_csv('VNAT/WF/flows_extra_labelled.csv')\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of rows before filtering: 51947\n",
      "Number of rows after filtering: 51947\n",
      "CDF at 41.0 seconds: 1.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CDF at 8000 packets: 0.9931276108341195\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18877\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from statsmodels.distributions.empirical_distribution import ECDF\n",
    "\n",
    "# Load dataset\n",
    "filtered_df = pd.read_csv('VNAT/WF/flows_extra_labelled_12.csv')\n",
    "print(f\"Number of rows before filtering: {df.shape[0]}\")\n",
    "\n",
    "\n",
    "print(f\"Number of rows after filtering: {filtered_df.shape[0]}\")\n",
    "\n",
    "data_duration = filtered_df['bidirectional_duration_ms'].dropna() / 1000  \n",
    "\n",
    "ecdf_duration = ECDF(data_duration)\n",
    "\n",
    "value_duration = 41.0  \n",
    "print(f\"CDF at {value_duration} seconds: {ecdf_duration(value_duration)}\")\n",
    "\n",
    "plt.step(ecdf_duration.x, ecdf_duration.y, label=\"ECDF\")\n",
    "plt.xscale('log') \n",
    "plt.xlabel('Bidirectional Duration (s)')\n",
    "plt.ylabel('ECDF')\n",
    "plt.title('Empirical CDF of Bidirectional Duration')\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()\n",
    "\n",
    "###################################################################\n",
    "\n",
    "data_packets = filtered_df['bidirectional_packets'].dropna() \n",
    "\n",
    "ecdf_packets = ECDF(data_packets)\n",
    "\n",
    "value_packets = 8000  \n",
    "print(f\"CDF at {value_packets} packets: {ecdf_packets(value_packets)}\")\n",
    "\n",
    "plt.step(ecdf_packets.x, ecdf_packets.y, label=\"ECDF\")\n",
    "plt.xscale('log') \n",
    "plt.xlabel('Bidirectional Packets')\n",
    "plt.ylabel('ECDF')\n",
    "plt.title('Empirical CDF of Bidirectional Packets')\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()\n",
    "\n",
    "\n",
    "print(filtered_df['bidirectional_duration_ms'].value_counts().max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d2c775d1-d08c-4c39-b63d-f743175cf36a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    51947.000000\n",
       "mean        11.371808\n",
       "std         17.391648\n",
       "min          0.000000\n",
       "25%          0.000000\n",
       "50%          0.001000\n",
       "75%         36.364500\n",
       "max         40.999000\n",
       "Name: bidirectional_duration_ms, dtype: float64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_duration.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "75059db6-7bf1-4691-a287-93b4710db5cb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count     51947.000000\n",
       "mean        733.502762\n",
       "std       12444.485317\n",
       "min           1.000000\n",
       "25%           2.000000\n",
       "50%           2.000000\n",
       "75%          36.000000\n",
       "max      857900.000000\n",
       "Name: bidirectional_packets, dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_packets.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "50a34474-2b46-42ca-954e-1432fc8afb8e",
   "metadata": {},
   "source": [
    "### CDF with filter out any flows that have less than 20 packets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "db6c0eb5-1110-4b0f-a873-42fb6fa2e64e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\yasah\\AppData\\Local\\Temp\\ipykernel_2028\\1358216243.py:7: DtypeWarning: Columns (86,87,88) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  filtered_df = pd.read_csv('flows_extra_labelled_12.csv')\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of rows before filtering: 51947\n",
      "Number of rows after filtering: 14617\n",
      "CDF at 41.0 seconds: 1.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CDF at 8000 packets: 0.9755763836628583\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from statsmodels.distributions.empirical_distribution import ECDF\n",
    "\n",
    "filtered_df = pd.read_csv('VNAT/WF/flows_extra_labelled_12.csv')\n",
    "print(f\"Number of rows before filtering: {filtered_df.shape[0]}\")\n",
    "\n",
    "filtered_df = filtered_df[filtered_df['bidirectional_packets'] > 19]\n",
    "\n",
    "print(f\"Number of rows after filtering: {filtered_df.shape[0]}\")\n",
    "\n",
    "data_duration = filtered_df['bidirectional_duration_ms'].dropna() / 1000 \n",
    "\n",
    "ecdf_duration = ECDF(data_duration)\n",
    "\n",
    "value_duration = 41.0  \n",
    "print(f\"CDF at {value_duration} seconds: {ecdf_duration(value_duration)}\")\n",
    "\n",
    "plt.step(ecdf_duration.x, ecdf_duration.y, label=\"ECDF\")\n",
    "plt.xscale('log') \n",
    "plt.xlabel('Bidirectional Duration (s)')\n",
    "plt.ylabel('ECDF')\n",
    "plt.title('Empirical CDF of Bidirectional Duration')\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()\n",
    "\n",
    "###################################################################\n",
    "\n",
    "data_packets = filtered_df['bidirectional_packets'].dropna()  \n",
    "\n",
    "ecdf_packets = ECDF(data_packets)\n",
    "\n",
    "value_packets = 8000  \n",
    "print(f\"CDF at {value_packets} packets: {ecdf_packets(value_packets)}\")\n",
    "\n",
    "plt.step(ecdf_packets.x, ecdf_packets.y, label=\"ECDF\")\n",
    "plt.xscale('log') \n",
    "plt.xlabel('Bidirectional Packets')\n",
    "plt.ylabel('ECDF')\n",
    "plt.title('Empirical CDF of Bidirectional Packets')\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1055e6c5-6102-47cf-9148-bf8f7e822c4b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    14617.000000\n",
       "mean        33.893629\n",
       "std         12.570976\n",
       "min          0.014000\n",
       "25%         37.156000\n",
       "50%         39.554000\n",
       "75%         40.285000\n",
       "max         40.999000\n",
       "Name: bidirectional_duration_ms, dtype: float64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_duration.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "971c5422-c961-45c6-950a-40902cda9769",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "count    14617.00\n",
      "mean      2599.00\n",
      "std      23357.13\n",
      "min         20.00\n",
      "25%         56.00\n",
      "50%         63.00\n",
      "75%         73.00\n",
      "max     857900.00\n",
      "Name: bidirectional_packets, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "with pd.option_context('display.float_format', '{:.2f}'.format):\n",
    "    print(data_packets.describe())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c7124e3b-4078-4e2f-95a1-be3b373be9fc",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}