update codes

Codes/FTT_autocorrelation/Real data/FTT_autocorrelation.py

+from influxdb_client import InfluxDBClient
+import numpy as np
+import matplotlib.pyplot as plt
+from scipy.fft import fft, fftfreq
+from scipy.signal import find_peaks
+
+# Load configuration from config.json
+config_path = r"C:\Allianz\4\0211\1201\config.json"
+with open(config_path, 'r') as config_file:
+    config = json.load(config_file)
+
+url = config["url"]
+token = config["token"]
+org = config["org"]
+bucket = config["bucket"]
+
+# Create an InfluxDB client
+client = InfluxDBClient(url=url, token=token, org=org)
+query_api = client.query_api()
+
+# Query to fetch beaconing data
+query = '''
+from(bucket: "Net")
+  |> range(start: 2023-08-01T00:00:00Z, stop: 2023-08-02T00:00:00Z)
+  |> filter(fn: (r) => r["_measurement"] == "hostnames")
+  |> filter(fn: (r) => r["url_hostname"] == "saml.allianz.com")
+  |> keep(columns: ["_time"])
+'''
+
+# Fetch data from InfluxDB
+result = query_api.query(query)
+
+# Extract timestamps
+timestamps = [record.get_time() for table in result for record in table.records]
+
+# Convert timestamps to seconds since the first timestamp
+timestamps = np.array([(t - timestamps[0]).total_seconds() for t in timestamps])
+
+# Create a time series with gaps for missing data
+# Define a time grid with a fixed sampling rate (e.g., 1 second)
+time_grid = np.arange(0, timestamps[-1] + 1, 1)  # 1-second resolution
+values = np.zeros_like(time_grid, dtype=float)
+
+# Fill in the values where data exists
+for t in timestamps:
+    idx = int(t)  # Convert timestamp to index in the time grid
+    values[idx] = 1  # Mark the presence of data
+
+# Apply Fourier Transform to detect periodicity
+def apply_fourier_transform(time_grid, values):
+    # Compute the Fast Fourier Transform (FFT) using scipy
+    n = len(values)
+    fft_values = fft(values)
+    freqs = fftfreq(n, d=1.0)  # Frequency bins (d=1.0 for 1-second sampling)
+
+    # Plot the frequency spectrum
+    plt.figure(figsize=(10, 6))
+    plt.plot(freqs[:n // 2], np.abs(fft_values[:n // 2]))
+    plt.title("Frequency Spectrum (Fourier Transform)")
+    plt.xlabel("Frequency (Hz)")
+    plt.ylabel("Amplitude")
+    plt.grid()
+    plt.show()
+
+    # Find the dominant frequency
+    dominant_freq = freqs[np.argmax(np.abs(fft_values[:n // 2]))]
+    print(f"Dominant frequency: {dominant_freq} Hz")
+
+# Apply Autocorrelation to detect repeating patterns
+def apply_autocorrelation(values):
+    print(values)
+    # Compute the autocorrelation function
+    autocorr = np.correlate(values, values, mode='full')[len(values) - 1:]
+    autocorr = autocorr / np.max(autocorr)  # Normalize autocorrelation values
+    lags = np.arange(len(autocorr))
+    print(autocorr)
+    print(lags)
+    # Plot the autocorrelation function
+    plt.figure(figsize=(10, 6))
+    plt.plot(lags, autocorr)
+    plt.title("Autocorrelation Function")
+    plt.xlabel("Lag")
+    plt.ylabel("Autocorrelation")
+    plt.grid()
+    plt.show()
+
+    # Find the lag with the highest autocorrelation (excluding lag 0)
+    peaks, _ = find_peaks(autocorr[1:], height=0.5)  # Adjust height threshold as needed
+    if len(peaks) > 0:
+        max_lag = peaks[0] + 1  # Add 1 to account for the exclusion of lag 0
+        print(f"Lag with highest autocorrelation: {max_lag}")
+    else:
+        print("No significant autocorrelation peaks found.")
+
+# Apply Fourier Transform
+apply_fourier_transform(time_grid, values)
+
+# Apply Autocorrelation
+apply_autocorrelation(values)
+
+# Close the InfluxDB client
+client.close()
\ No newline at end of file