TiHAN-V2X: A Comprehensive Dataset for Dynamic C-V2X Communication within the Indian Context

Citation Author(s):
ANNU
.
Indian Institute of Technology Hyderabad
Phaneendra
Pydimarri
Indian Institute of Technology, Hyderabad
Pachamuthu
Rajalakshmi
Indian Institute of Technology, Hyderabad
Sunnam Venkata
Srikanth
Center for Development of Advanced Computing, Hyderabad
Munjagala
Prasad
Center for Development of Advanced Computing, Hyderabad
Submitted by:
ANNU SAHU
Last updated:
Tue, 10/08/2024 - 04:05
DOI:
10.21227/f2kd-9g03
Data Format:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

The TiHAN-V2X Dataset was collected in Hyderabad, India, across various Vehicle-to-Everything (V2X) communication types, including Vehicle-to-Vehicle (V2V), Vehicle-to-Infrastructure (V2I), Infrastructure-to-Vehicle (I2V), and Vehicle-to-Cloud (V2C). The dataset offers comprehensive data for evaluating communication performance under different environmental and road conditions, including urban, rural, and highway scenarios. The dataset covers various communication parameters such as transmitted and received latitude, longitude, speed, altitude, heading, SNR, RSSI, path loss, effective radiated power (ERP), RX power, and noise power. This data is vital for studying the reliability and efficiency of V2X systems and optimizing protocols for connected vehicles.

The dataset is structured as follows:

  • V2V: 35 fields, 10,253 entries across 7 scenarios (S1-S7), covering both urban and rural areas, highways, and city streets.
  • V2I: 30 fields, 2,957 entries across 3 scenarios, focusing on optimizing V2I communication performance.
  • I2V: 38 fields, 2,106 entries collected in the TiHAN testbed at IIT Hyderabad, including weather data from RSUs and 5G sidelink interface.
  • V2C: 22 fields, 1,971 entries, focusing on cloud communication from vehicles using the User Datagram Protocol (UDP).

This dataset is critical for researchers and engineers focusing on V2X communication, enabling them to analyze communication parameters and performance under real-world conditions and apply AI/ML models to improve connected vehicle technologies.

Instructions: 

Documentation (Data Dictionaries and Structure Documentation):

  1. V2V (Vehicle-to-Vehicle) Communication:

    • Entries: 10,253
    • Fields: 35
    • Scenarios: 7 (S1-S7) covering urban, rural, highways, and city streets.
    • Key Parameters:
      • Time_Epoch: Time of communication event (epoch format)
      • Transmitted Latitude, Longitude: Location of the transmitting vehicle (degrees)
      • Received Latitude, Longitude: Location of the receiving vehicle (degrees)
      • Speed (km/h): Speed of both transmitting and receiving vehicles
      • Altitude (m): Altitude of both vehicles
      • Heading (degrees): Heading direction of both vehicles
      • Distance (m): Distance between the transmitting and receiving vehicles
      • SNR (dB): Signal-to-noise ratio during communication
      • RSSI (dBm): Received Signal Strength Indicator for two antennas
      • Path Loss (dB): Signal attenuation
      • ERP (dBm): Effective Radiative Power
      • RX Power (dBm): Received power at the receiving vehicle
      • Noise Power (dBm): Power level of background noise
      • Packet Error Rate (PER): Rate of packet loss during transmission
  2. V2I (Vehicle-to-Infrastructure) Communication:

    • Entries: 2,957
    • Fields: 30
    • Scenarios: 3 V2I scenarios
    • Key Parameters:
      • Latitude/Longitude (Tx & Rx): Location of the vehicle and RSU (degrees)
      • Heading (degrees): Heading direction of the vehicle
      • Speed (km/h): Speed of the vehicle
      • Altitude (m): Altitude of both the RSU and vehicle
      • Latency (ms): Communication delay between vehicle and RSU
      • Throughput (bits/sec): Data transfer rate
      • Distance (m): Distance between vehicle and RSU
      • Path Loss (dB): Signal path loss
      • ERP (dBm): Effective radiated power
      • RX Power (dBm): Received power at the vehicle
      • Noise Power (dBm): Background noise level
      • SNR (dB): Signal-to-noise ratio
      • Packet Error Rate (PER): Communication error rate
      • Receiver Gain (dB): Gain of the receiving antenna
  3. I2V (Infrastructure-to-Vehicle) Communication:

    • Entries: 2,106
    • Fields: 38
    • Scenarios: 1 (conducted at TiHAN - IIT Hyderabad)
    • Key Parameters:
      • Similar to V2I but includes additional weather data (temperature, pressure, precipitation) as broadcasted by RSUs via the 5G sidelink interface.
      • This dataset supports the analysis of I2V communication performance and its reliability under different weather conditions.
  4. V2C (Vehicle-to-Cloud) Communication:

    • Entries: 1,971
    • Fields: 22
    • Key Parameters:
      • Transmission/Reception Timestamp: Time of data transmission and reception
      • Latitude/Longitude (Tx): Location of the transmitting vehicle
      • Cloud Coordinates: Location of the receiving cloud server
      • Distance to Cloud (m): Distance between vehicle and cloud
      • Latency (ms): Delay in communication between vehicle and cloud
      • Packet Error Rate (PER): Communication error rate
      • Bandwidth (Mbits/sec): Data transmission capacity
      • Download/Upload Speed (Mbits/sec): Speed of data transfer
      • Total Data Transferred (MBytes): Cumulative amount of data transferred between vehicle and cloud
      • UDP: The dataset employs the User Datagram Protocol (UDP) for time-sensitive, efficient transmission between vehicle and cloud.

Instructions for Utilizing the Dataset:

    1. Loading the Data:

      • Each CSV file (V2V, V2I, I2V, V2C) can be loaded into any data analysis platform, such as Python using pandas.
      • Example for loading the V2V data:pythonimport pandas as pd v2v_data = pd.read_csv('FinalV2V_Dataset.csv')
    2. Preprocessing:

      • Unit normalization: Convert latencies, throughput, and power levels to consistent units across datasets if needed.
      • Filtering: Filter by scenarios or specific locations to focus on particular communication patterns.
      • Handling missing data: Check for missing or erroneous values and apply appropriate imputation techniques.
    3. AI/ML Applications:

      • Predictive Models: Use regression models to predict latency or throughput based on distance, SNR, or other parameters.
      • Classification Models: Build classifiers to detect conditions that lead to high packet error rates or communication failures.
      • Optimization: Machine learning can be used to optimize transmission settings (e.g., frequency, power) for improved V2X performance.
      • Autonomous Driving Simulations: Use this data to model real-world V2X communications in smart city or highway driving scenarios, enhancing decision-making for autonomous vehicles.
    4. Analyzing Performance:

      • Latency and Throughput: Use descriptive statistics to evaluate latency and throughput for different communication types.
      • Path Loss and SNR Analysis: Investigate how varying distances, locations, and scenarios impact path loss and SNR.
      • Scenario-based Analysis: Compare V2X performance across the seven V2V scenarios, or evaluate performance under specific urban, rural, and highway conditions.

Size of the Dataset:

    • V2V Dataset: 10,253 entries, 35 fields
    • V2I Dataset: 2,957 entries, 30 fields
    • I2V Dataset: 2,106 entries, 38 fields
    • V2C Dataset: 1,971 entries, 22 fields

    This large dataset provides sufficient data for in-depth machine learning analysis, simulation studies, and real-world evaluation of V2X systems.