Skip to content

Latest commit

 

History

History

throughput_rtt_coverage_ho

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
coverage
coverage
 
 
ho/processed
ho/processed
 
 
plots
plots
 
 
rtt
rtt
 
 
scripts
scripts
 
 
tput
tput
 
 
README.md
README.md
 
 

README.md

Coverage - Throughput & RTT tests - Handovers

Dataset Structure

This sub-dataset is divided into 4 sections:

The respective raw data (application + XCAL) are in the aforementioned directories. Additionally, there are processed data under each subdirectory with a subfolder named "processed".

The scripts folder contains scripts starting with imc_dataset_*_process.py to generate the processed data. Scripts starting with plot_*_.py generates figures 1 to 12 in the IMC paper.

Note: The dataset already contains processed data to generate the plots used in the IMC paper, and processing scripts are provided as a starting point for users to extract various KPIs (used in the paper) according to their specific requirements

Using the processed data

In each sub-folder (coverage/tput/rtt/ho) there is a folder called processed. The following explains how to use those processed data:

  • Coverage:

    # Filename: dist_tz_speed_operator.pkl
import pickle
filehandler = open("dist_tz_speed_operator.pkl", "rb")
total_dist_operator, breakup_dist_operator, total_dist_tz_operator, breakup_dist_tz_operator, total_dist_speed_operator, breakup_dist_speed_operator = pickle.load(filehandler)
filehandler.close()

    Variable information and structure:

    Variable Description Structure
    total_dist_operator Operator-wise total distance travelled in miles 
    {
       "operator1":"x",
       "operator2":"y",
       ...
    }
    breakup_dist_operator Operator-wise and technology-wise breakdown
    - of distance travelled in miles    		 
    {
      "operator1":{
        "tech1":x1,
        "tech2":x2,
        "…"
      },
      "…"
    }
    total_dist_tz_operator Operator-wise total distance travelled
    - per timezone in miles    		 
    {
      "operator1":{
        "tz1":x1,
        "tz2":x2,
        ...
      },
      ...
    }
    breakup_dist_tz_operator Operator-wise and technology-wise breakdown
    - of distance travelled per timezone in miles    		 
    {
      "operator1":{
        "tech1":{
          "tz1":x11,
          "tz2":x12,
          ...
        },
        ...
      },
      ...
    }
    total_dist_speed_operator Operator-wise total distance travelled
    - per speed range in miles    		 
    {
      "operator1":{
        "speed-range1":x1,
        "speed-range2":x2,
        ...
      },
      ...
    }
    breakup_dist_speed_operator Operator-wise and technology-wise breakdown
    - of distance travelled per speed range in miles    		 
    {
      "operator1":{
        "tech1":{
          "speed-range1":x11,
          "speed-range2":x12,
          ...
        },
        ...
      },
      ...
    }

  • Throughput data:

    The main_op_link_dict.pkl has a dict which contains data in the following format:

    ├── operator
    ├── downlink or uplink
        ├── tput_speed_tech_dict
        ├── ca_speed_tech_dict
        ├── fiveg_ca_speed_dict
        ├── lte_ca_speed_dict
        ├── tput_tz_tech_dict
        ├── dist_speed_tech_dict
        ├── mcs_speed_dict
        ├── bler_speed_dict
        ├── rsrp_speed_dict
        ├── wl_speed_dict
        ├── overall_mean_list
        ├── overall_std_list
        ├── overall_5g_high_percent
    # Filename: main_op_link_dict.pkl
import pickle
filehandler = open("main_op_link_dict.pkl", "rb")
main_op_link_tput_dict = pickle.load(filehandler)
for op in main_op_link_tput_dict.keys():
   for link in ['dl', 'ul']: 
      tput_speed_tech_dict, ca_speed_tech_dict, fiveg_ca_speed_dict, lte_ca_speed_dict, tput_tz_tech_dict, dist_speed_tech_dict, mcs_speed_dict, bler_speed_dict, rsrp_speed_dict, wl_speed_dict, overall_mean_list, overall_std_list, overall_5g_high_percent = main_op_link_tput_dict[op][link]
filehandler.close()

    Variable information and structure:

    Variable Structure
    tput_speed_tech_dict 
    {
      "tech1":{
        "speed1":[
          "throughput values"
        ],
        ...
      },
      ...
    }
    ca_speed_tech_dict 
    {
      "tech1":{
        "speed1":[
          "total ca values"
        ],
        ...
      },
      ...
    }
    fiveg_ca_speed_dict 
    {
      "tech1":{
        "speed1":[
          5g ca values
        ],
        ...
      },
      ...
    }
    lte_ca_speed_dict 
    {
      "tech1":{
        "speed1":[
          "lte ca values"
        ],
        ...
      },
      ...
    }
    tput_tz_tech_dict 
    {
      "tz1":[
        "throughput values"
      ],
      "tz2":[
        "throughput values"
      ],
      ...
    }
    dist_speed_tech_dict x
    mcs_speed_dict 
    {
      "tech1":{
        "speed1":[
          "mcs values"
        ],
        ...
      },
      ...
    }
    bler_speed_dict 
    {
      "tech1":{
        "speed1":[
          "bler values"
        ],
        ...
      },
      ...
    }
    rsrp_speed_dict 
    {
      "tech1":{
        "speed1":[
          "rsrp values"
        ],
        ...
      },
      ...
    }
    wl_speed_dict 
    {
      "tech1":{
        "speed1":[
          "edge used or not values"
        ],
        ...
      },
      ...
    }
    overall_mean_list [average of each 30-35 seconds throughput test]
    overall_std_list [standard deviation of each 30-35 seconds throughput test]
    overall_5g_high_percent [% of 5G mid/mmWave in each 30-35 seconds throughput test]

  • Handovers:

    There are 3 pickle files which contains data in the following format:

    ├── main_op_link_ho_per_mile_dict
    ├── operator
        ├── downlink/uplink
            ├── data structure
├── main_op_link_ho_duration_dict
    ├── operator
        ├── downlink/uplink
            ├── data structure
├── main_op_link_ho_tput_dict
    ├── operator
        ├── downlink/uplink
            ├── data structure
    # Filenames: main_op_link_ho_per_mile_dict.pkl, main_op_link_ho_duration_dict.pkl, main_op_link_ho_tput_dict.pkl
import pickle
filehandler = open("main_op_link_ho_per_mile_dict.pkl", "rb")
main_op_link_ho_per_mile_dict = pickle.load(filehandler)
filehandler.close()
filehandler = open("main_op_link_ho_duration_dict.pkl", "rb")
main_op_link_ho_duration_dict = pickle.load(filehandler)
filehandler.close()
filehandler = open("main_op_link_ho_tput_dict.pkl", "rb")
main_op_link_ho_tput_dict = pickle.load(filehandler)
filehandler.close()

    Variable information and structure:

    Variable Structure
    main_op_link_ho_per_mile_dict 
    {
      "operator1":{
        "linkype1":[
          [
            "list of latitude longitude pairs"
          ],
          [
            "handover count list"
          ],
          [
            "distance travelled list"
          ],
          [
            "handovers per mile list"
          ]
        ],
        ...
      },
      ...
    }
    main_op_link_ho_duration_dict 
    {
      "operator1":{
        "linktype1":[
          "list of handover duration"
        ],
        ...
      },
      ...
    }
    main_op_link_ho_tput_dict 
    {
      "operator1":{
        "linktype1":[
          {
            "dictionary of different handovers with pre-handover throughput list"
          },
          {
            "dictionary of different handovers with post-ho throughput list"
          },
          {
            "list of post - pre handover throughput values"
          },
          [
            list of t1,
            t2,
            t3,
            t4,
            t5 from the paper
          ]
        ],
        ...
      },
      ...
    }

  • RTT:

    The main_op_link_dict.pkl has a dict which contains data in the following format:

    ├── main_op_rtt_dict
    ├── operator
        ├── data structure
├── main_rtt_5g_dict
    ├── operator
      ├── data structure
├── main_rtt_tech_dict
    ├── operator
          ├── data structure
├── main_rtt_edge_tech_dict
    ├── operator
      ├── data structure
├── op_speed_dict
    ├── operator
      ├── data structure
├── op_rtt_dict
    ├── operator
      ├── data structure
├── op_tech_dict
    ├── operator
      ├── data structure
    # Filename: main_op_link_dict.pkl
import pickle
filehandler = open("main_op_link_dict.pkl", "rb")
main_op_link_rtt_dict = pickle.load(filehandler)
main_op_rtt_dict, main_rtt_5g_dict, main_rtt_tech_dict, main_rtt_edge_tech_dict, op_speed_dict, op_rtt_dict, op_tech_dict = main_op_link_rtt_dict
filehandler.close()

    Variable information and structure:

    Variable Structure
    main_op_rtt_dict 
    {
      "operator1":[
        [
          "latitude",
          "longitude",
          [
            "list of rtt values"
          ],
          "path_to_rtt_file"
        ]
      ],
      ...
    }
    main_rtt_5g_dict 
    {
      "operator1":{
        "percent-high-5g":[
          "list of rtt values"
        ]
      }
    }
    main_rtt_tech_dict 
    {
      "operator1":{
        "tech1":[
          "list of rtt values"
        ],
        ...
      },
      ...
    }
    main_rtt_edge_tech_dict 
    {
      "operator1":{
        "tech1":[
          "list indicating whether an rtt values was from a edge/cloud server test"
        ],
        ...
      },
      ...
    }
    op_rtt_dict 
    {
      "operator1":[
        list of each 200 ms rtt samples
      ],
      ...
    }:
    op_speed_dict 
    {
      "operator1":[
        "list of speed in miles for each rtt sample"
      ],
      ...
    }
    op_tech_dict 
    {
      "operator1":[
        "list of cellular technology for each rtt sample"
      ],
      ...
    }

Steps to Reproduce IMC Paper Plots (Use a Windows system to generate the plots)

  1. Install dependencies with pip:

    python3 -m pip install -r requirements.txt

  2. Run the plotting script

    python3 scripts\plot_coverage.py
python3 scripts\plot_handover_logger_data.py
python3 scripts\plot_tput_rtt.py
python3 scripts\plot_parallel_throughput_tech.py
python3 scripts\plot_ho.py

  3. Examine the resulting figures. You are supposed to see the following figures being plotted, with filenames matching the figure IDs in the paper:

    plots/
├── fig_1
    ├── fig_1*.pdf
├── fig_2
    ├── fig_2*.pdf
├── fig_3a.pdf
├── fig_3b.pdf
├── fig_4.pdf
├── fig_5.pdf
├── fig_6a.pdf
├── fig_6b.pdf
├── fig_6c.pdf
├── fig_6d.pdf
├── fig_7
    ├── *.pdf
├── fig_8.pdf
├── fig_9.pdf
├── fig_10dl.pdf
├── fig_10ul.pdf
├── fig_10c.pdf
├── fig_11a.pdf
├── fig_11b.pdf
└── fig_12
    ├── *.pdf