READ ME File For 'Dataset'

Dataset DOI: 10.5258/SOTON/D1519

ReadMe Author: Haochen LIU, University of Southampton

This dataset supports the publication:
AUTHORS: Haochen Liu, Siyao Lu, Mohammed El-Hajjar, Lie-Liang Yang
TITLE: Machine Learning Assisted Adaptive Index Modulation for mmWave Communications
JOURNAL: IEEE Open Journal of the Communications Society


This dataset contains: Figure  2, 6, 7, 8, 10, 12, 13, 14, 15, 16, 17 and 18 of the aforementioned paper. Each folder is named according to its content, where the curves of each figure are stored in mat files.  To regenerate the results, please use the Matlab.


The figures are as follows:

  - Figure-2: Contains the dataset of Figure 2. Achievable data rate versus the SNR for a system employing a transmitter with 64 elements and ULA, URPA, UCPA and UCYA and a receiver employing ULA with 32 elements.
   
  - Figure-6: Contains the dataset of Figure 6. An illustration of the channel gains over a time-frequency plane.
  
  - Figure-7: Contains the dataset of Figure 7. Time-varying channel magnitude of a subcarrier

  - Figure-8: Contains the dataset of Figure 8. BER performance of OFDM-CSIM systems

  - Figure-10: Contains the dataset of Figure 10. The training sets for MODE1, MODE2 and MODE3.

  - Figure-12: Contains the dataset of Figure 12. Data rate achieved by a 64 * 16 MIMO system with the ULA at both transmitter and receiver, when both transmitter and receiver employ 4 RF chains.
 
  - Figure-13: Contains the dataset of Figure 13. Achievable data rate versus SNR performance by adapting hybrid beamforming for ULA, URPA, UCPA and UCYA.
 
  - Figure-14: Contains the dataset of Figure 14. BER versus SNR performance of the knn assisted adaptive modulations with different k values.

  - Figure-15: Contains the dataset of Figure 15. Performance comparison of the knn assisted adaptive modulations with different k values.

  - Figure-16: Contains the dataset of Figure 16. Probabilities for the system to be operated with MODE1, MODE2 and MODE3. The parameters of propagation medium is summarized in  Table III, and the elevation angle and azimuth angle for both the arrival and departure rays are assumed to obey the Laplace distribution with the angular spread of 7.5.

  - Figure-17: Contains the dataset of Figure 17. Throughput comparison of the conventional and learning-assisted adaptive modulation.

  - Figure-18: Contains the dataset of Figure 18. BER performance of the conventional and learning-assisted adaptive modulations, as well as of the individual MODEs.


Date of data collection: 06, 2019 ~ 06, 2020

Information about geographic location of data collection: University of Southampton

Licence: CC BY

Related projects:
The financial support of the Engineering and Physical Sciences Research Council (EPSRC) and the Royal Academy of Engineering  EP/P034284/1.


Date that the file was created: 08, 2020