Skip to content

Latest commit

 

History

History
39 lines (26 loc) · 3.46 KB

README.md

File metadata and controls

39 lines (26 loc) · 3.46 KB
Raw

Simulator for LoFi User Scheduling for Multiuser MIMO Wireless Systems

(c) 2023 Victoria Palhares, Reinhard Wiesmayr e-mail: [email protected], [email protected]

Important information

If you are using this simulator (or parts of it) for a publication, then you must cite our paper:

A. Gallyas-Sanhueza, G.Marti, V. Palhares, R. Wiesmayr, and C. Studer, "LoFi User Scheduling for Multiuser MIMO Wireless Systems," 2024 IEEE International Conference on Acoustics, Speech and Signal Processing, Seoul, South Korea, 2024.

and clearly mention this in your paper. In the following, by paper, we mean the above paper.

How to start a simulation:

To regenerate the plots in figures 3 and 4 of the paper, simply assign the letter 'a' to the variable par.sim_scenario in the main_sim_code and run the script. It will generate two plots with BER curves. Also note that the list of schedulers to be simulated is set in the par.scheduling_chosen variable in the param_config.m file.

In order to simulate with parameters other than our predefined values, define a new set of parameters in the switch case block of param_config.m file and assign a value to the variable par.sim_scenario in the main_sim_code.m that corresponds to the new set of parameters. Please consider the following guidelines when creating a new set of simulation parameters:

  • Set the number of BS antennas and UEs and other parameters in the param_config.m file

  • Our channel dataset, which has been generated with REMCOM Wireless InSite, can be found in channel/Channels_B_16_U_16_Ch_100.mat. The channel dataset is of size 16*16*100, where the first dimension specifies 16 basestation receive antennas, the second 16 single-antenna user equipments (UEs), and the third dimension represents 100 channel realizations. In the simulated outdoor scenario, we consider 22448 possible UE positions. For every channel realization, we picked 16 UE positions uniformly and independently at random. Note that the number of channel realizations in par.channels has to be set to an integer less than or equal to 100, unless other channels are provided in the channel folder

  • The variable par.scheduling_options denote all the available methods in the simulator while par.scheduling_chosen denote all the methods that will be simulated

  • We perform par.trials transmission trials, where we fix the channel and scheduler and vary the symbols and noise

  • We perform user scheduling once per channel realization

  • Our simulator is fixed to par.timeslots = 2, but our framework could be extended in future work to more timeslots

  • The dynamic range of the receive power from the weakest to the strongest UE can be set in the variable par.P_db as a decibel value

  • The modulation scheme can be changed to BPSK, QPSK, or 64QAM in the variable par.modulation

  • The channel estimation can be disabled by setting the variable par.channel_estimation to 'Perfect'

  • The overall number of processes is given by the number of par.channels*length(par.SNRdB_list)*length(par.scheduling_chosen)

  • You need to modify your par.results_path variable in main_sim_code according to where you want to save your result files

We highly recommend you to execute the code step-by-step (using MATLAB's debug mode) in order to get a detailed understanding of the simulator.

Version history