This is for XR/AR network traffic classification & Paper reproducability.

The Extended Reality (XR) technologies continue to advance, driven by hardware and software growth in this scope. In addition to usecases e.g. game and education which is run a completely immersive, computer-generated environment by Virtual Reality (VR), there are some requirements which need to keep the user interactive with real world and overlay the digital object onto the real world e.g. repair, maintenance, health care which it is done by Augmented Reality (AR). The AR applications and users are rising with having been matured the VR Head Mounted Display (HMD) and software. The AR is sensitive to the computing resource, so it needs to offload its functionality to improve its flexibility but it makes it sensitive to delay; in addition, its interaction with the real world makes this sensitivity more drastic. The discrimination of the network traffic belonging to AR, as the first step for appropriate resource allocation, is becoming more critical with the growth of its application and remote rendering capability. In this paper, we propose a decision Tree (DT) and Random Forest (RF) model to classify the network traffic into AR, Cloud Gaming (CG) and others. Secondly, the features are examined to detect the AR and CG network traffic accurately. Thirdly, the model is evaluated with existing datasets to show its performance in accuracy, precision, recall, and f-score. Finally, a dataset for AR and CG are collected to used in this research and published to pave the way for other research on the AR network traffic discrimination. This paper is useful for those are working on the network traffic classification specifically AR network traffic classification.

a) IPI (Inter Packet Interval) b) FS (Frame Size) c) IFI (Inter Frame Interval)

                            # Set the number of samples for generating
                                n_sample = 1000
                                
                           # Save the combined dataset to a CSV file (***************** Add your address to Save th CSV files)
                                combined_dataset_LR.to_csv('AR_UL_LR.csv', index=False)
                                combined_dataset_MR.to_csv('AR_UL_MR.csv', index=False)
                                combined_dataset_HR.to_csv('AR_UL_HR.csv', index=False)
                                combined_dataset_AR70.to_csv('AR_DL70.csv', index=False)
                                combined_dataset_AR90.to_csv('AR_DL90.csv', index=False)

The dataset will be generated for Uplink & Downlink sperated as below: ##### Uplink Low Resolution --> 'AR_UL_LR.csv'| Medeum Resolution --> 'AR_UL_MR.csv'| High Resolution --> 'AR_UL_HR.csv'| ##### Downlink Rate is 70Hz --> 'AR_DL70.csv'| Rate is 90Hz --> 'AR_DL90.csv'|

                          # The root Directory (Enter your address)
                              root_directory = r'Directory includes PCAP or PCAPnj Files'

                          # set the extension to find from the rood directory
                              extension = '.pcapng'  # ('.pcap' or '.pcapng')

The csv files will be stored in the same folder. The next cell in the Python code merge the CSV files as DS.csv and then add direction (UL or DL) and preprocess the csv file. The final output is DS3.csv.

                           # The root Directory (Enter your address)
                              root_directory = r'Directory includes PCAP or PCAPnj Files'

                           # set the extension to find from the rood directory
                              extension = '.csv'  # ('.pcap' or '.pcapng')

The csv files will be stored in the same folder. The next cell in the Python code merge the CSV files as DS.csv and then add direction (UL or DL) and preprocess the csv file. The final output is DS3.csv.

Evaluation is done with Accuracy, precision, recall, f-score, and confusion matrix

                           # Load datasets
                          ### Enter your Dataset file address with csv format
                              ar_data = pd.read_csv(r'AR.csv')
                              cg_data = pd.read_csv(r'CG.csv')
                              others_data = pd.read_csv(r'others.csv')

In Cell numbr 6 only set the test dataset you can test the DT model

                          # Load the test dataset (Mentioned in Table (IV) of the Paper)
                              AR_Test = pd.read_csv(r'AR-Test.csv')
                              CG_Test = pd.read_csv(r'CG_Test.csv')
                              Other_Test = pd.read_csv(r'Other_Test.csv')

Evaluation is done with Accuracy, precision, recall, f-score, and confusion matrix

                           # Load datasets
                          ### Enter your Dataset file address with csv format
                              ar_data = pd.read_csv(r'AR.csv')
                              cg_data = pd.read_csv(r'CG.csv')
                              others_data = pd.read_csv(r'others.csv')

In Cell numbr 5 only set the test dataset you can test the DT model

                          # Load the test dataset (Mentioned in Table (IV) of the Paper)
                              AR_Test = pd.read_csv(r'AR-Test.csv')
                              CG_Test = pd.read_csv(r'CG_Test.csv')
                              Other_Test = pd.read_csv(r'Other_Test.csv')

It is available in https://kaggle.com/datasets/a906acd0ce4c8ee03048bf10c06573547ddca5a5c775ba592306bd04038f3a56

This document help the reader of the paper to be able to reproduce the model and use it.