Code to accompany the paper "Dynamic fractional frequency reuse in OFDMA systems" by Kate Harrison and Gireeja Ranade.
Matlab M Objective-C
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
Figures
Helpers
data
README.md
create_data_for_specific_user_location.m
create_data_for_various_user_locations.m
demo.m
determine_user_choices.m
figure_3__comparison_in_two_cell_toy.m
figure_5__performance_of_power_minimization_algorithm.m
figure_6a_7ab__supportable_number_of_users.m
figure_6b__comparison_of_DFFR_and_reuse_half.m
figure_8__effect_of_subbands.m Adding sample Figure files for reference. Also completing the Figure … Jun 5, 2013
get_simulation_parameter.m
main_program.m
min_fcn.m
paper.pdf
run_me_first.m

README.md

Code for "Dynamic Fractional Frequency Reuse in OFDMA Systems" by Kate Harrison and Gireeja Ranade

Context

This code accmpanies the paper "Dynamic Fractional Frequency Reuse in OFDMA Systems" by Kate Harrison and Gireeja Ranade. The paper can be found here.

Some of the comments in the code and in particular the filenames may assume that you have read the paper above. They may also assume that you have read Stolyar and Viswanathan's paper "Self-organizing dynamic fractional frequency reuse in OFDMA systems".

Code organization

General simulation parameters are stored in get_simulation_parameter.m. While these values are frequently used throughout the code, there is no guarantee that changes will propagate as desired. Please double-check the source code if you want to be sure that it's working as intended. This file also provides definitions for each of the parameters (which are frequently included in other files but not systematically).

Similarly, changing the length of arrays (e.g. p_array, d_array, BS_power) may (or may not) have unintended results. The code is not as general as it should be and is therefore unfortunately a bit fragile.

The files named figure_*.m will generate each of the figures in the paper.

Some figures (e.g. Figure 3 and Figure 5) are stand-alone and require no pre-computation of data. Others require you to pre-compute some data using other files (e.g. create_data_*.m) before generating the figures. Whenever possible, I have tried to direct the user to the appropriate files in the header text of the figure_*.m file.

The create_data_*.m files rely on main_program.m. This is where the main implementation is housed. The create_data_*.m files are simply wrappers which execute main_program.m with the appropriate parameters. If you are extending this code, you will likely want to write your own wrapper and then call main_program.m. In this case, the create_data_*.m files will help you understand what the parameters of main_program.m are.

Directories

  • Miscellaneous useful files are kept in the Helpers/ directory to reduce clutter and confusion.
  • All generated data is stored in the data/ directory.
  • All generated figures are stored in the Figures/ directory.
  • Since some of the computations can take a long time, partial data is stored in the case of an interruption. There are many such files (over 1200 if you generate the data required for all of the figures) and they are stored in the partial_data/ directory.

All of these directories will be automatically added to your path when you run 'run_me_first.m'.

Code support and maintenance

I (Kate) do not intend to maintain this code any longer. I may be able to provide limited support so please email me if you have any questions or if you find any issues.

Copyright information

This code is freely available to anyone who wants to use it and is provided without guarantee. If you build off of it or find it useful, please consider citing our code or our paper (or both).

Author information

Kate Harrison

Gireeja Ranade