Datasets
Standard Dataset
Electric field strength of different micro-environments

- Citation Author(s):
- Submitted by:
- Ruijie Pan
- Last updated:
- Wed, 03/05/2025 - 05:53
- DOI:
- 10.21227/wdqr-8110
- Data Format:
- License:
- Categories:
- Keywords:
Abstract
This dataset is used for machine learning. And the data set is collected in different micro-environments. In this project, ExpoM-RF 4 is used to measure the electric field strength. Four different typs of micro-environments are selected which are urban (6 high population density areas in Kuala Lumpur), suburban (7 low population density areas in Cyberjaya), park (3 park areas) and one indoor micro-environment. From the measurement campaigns, three machine learning (ML) techniques are simulated to model the Electric Field Strength in each micro-environment. The ML techniques are Fully connected neural network (FCNN), eXtreme Gradient Boosting (XG Boost), and Linear Regression (LR) to predict the RMS and Maximum radiation exposure.
ExpoM-RF4 is used to measure RF-EMF in micro-environments. Compared to other measurement devices, it is compact and lightweight, making it convenient to carry in a waist pack. This device measures electric field intensity (V/m) of EMF and can automatically record the peak value, the minimum value, and RMS value of each signal. Additionally, its ability to configure custom frequency band lists from 50 MHz to 6 GHz provides exceptional flexibility, ensuring compatibility with future changes in frequency band allocations and regulations.
The measurement campaign starts with selecting a suitable micro-environment. Based on the literature review, several factors influence radiation levels including population density, frequency, range, vegetation coverage, building height and base station density. Considering these influencing factors, four distinct micro-environments are selected which are urban (6 high population density areas in Kuala Lumpur), suburban (7 low population density areas in Cyberjaya), park (3 park areas) and one indoor micro-environment.
For indoor measurement, precautions are taken to minimize interference by avoiding the use of other electronic devices nearby during data collection. For outdoor micro-environments, predefined walking paths are established on a map to ensure consistency in measurement. As suggested by other micro-environment studies, these paths are typically defined as 1-2 km long and/or require a 15–30-minute walk to complete. Google Maps is used to define the path length and walking area, with each micro-environment linked to a corresponding Google Maps route for reference. During data collection, the researcher follows the predetermined path while measuring RF-EMF exposure. The details of each micro-environment are provided in APPENDIX A.
To ensure accurate RF-EMF measurements while using Google Maps for navigation and minimizing interference from mobile phone radiation, each measurement campaign requires three team members. The first person used Google Maps to navigate the predefined walking path. The second person records environmental details including vegetation coverage, building features, number of base stations, and their proximity. To avoid interference, the third person carries only ExpoM-RF4 device in waist pact and is not allowed to bring any electronic equipment, mobile phone or Bluetooth-enabled equipment. This person follows the other two people while assisting in identifying any missing base stations. Those three should keep as much distance as possible to prevent interference from cell phone frequency. To account for the body-shielding effect during the measurements, the team walks to a designated point and then returns along the same path. This ensures that the ExpoM-RF4 device able to measure the RF-EMF exposure from both front and back direction.