Analysis of DDoS Detection in SDN Electric Self Driving Vehicles

Performance Analysis of DDoS Detection in SDN Based Electric Self Driving Vehicles

Implementation plan:
——————–

Step 1: Initially, we constructed an SDN enabled vehicular network with 200 Vehicles(increase upto 500), 100 RSU, 20 Switches,20 and 1 Controller.

Step 2: Then, we inject DDoS attack traffic in the simulation using user defined dataset.(CiCDDoS2019, CIVEV2023 , CICIoT2024 and Custom dataset)

Step 3: Next, we collect network data points during the attack based on dataset such as timestamp, flow ID, src/dst IP, packet size, packets/sec, send/receive times, delay and throughput

Step 4: Next, we preprocess the collected data by cleaning, labeling (normal vs attack), removing duplicates, and normalizing features.

Step 5: Next, we extract features such as flow duration, packet/byte rates, inter-arrival times, entropy, and statistical aggregates (mean, std, min, max).

Step 6: Next, we train the data using Naive Bayes, Passive-Aggressive and Hoeffding Tree ML models with RFE and PCA to mitigate attacks .

Step 7: Finally, we plot and analyze the following performance metrics:

7.1: Number of Epochs vs Accuracy (%)
7.2: Number of Epochs vs Precision (%)
7.3: Number of Epochs vs F1 Score(%)
7.4: Number of Epochs vs Recall (%)
7.5: False Negative Rate vs False Positive Rate
7.6: Time Vs Detection Latency (ms)
7.7: Time Vs Processing Throughput (Mbps)
7.8: Time Vs CPU Usage (%)
7.9: Time Vs Memory Usage (MB)
7.10: Time Vs Network Overhead (pkts/sec)
7.11: Classification Speed
7.12: Confusion Matrix

Software Requirements:
———————-

1. Development Tool: OMNeT++ 4.6 or above with Veins
2. Operating System: Windows 10 (64-bit) or above

Note:
—–

1) If the proposed plan does not fully align with your requirements, please provide all necessary details—including steps, parameters, models, and expected outcomes—in advance. Kindly ensure that any missing configurations or specifications are clearly outlined in the plan before confirming.

2) If there’s no built-in solution for what the project needs, we can always turn to reference models, customize our own, different math models or write the code ourselves to fulfil the process.

3) If the plan satisfies your requirement, Please confirm with us.

4) Project based on Simulation only.

5) If you have any dataset to change,kindly provide us before implementing it.

6) We only use 500 vehicles as maximum for better performance system compatibility,if we use more then 500 it makes the system unstable during simulation and also result visualization can’t be analyzed properly.

Related Pages