Skip to content

R scripts to generate operator infrastructure based on point processes

Notifications You must be signed in to change notification settings

MartinPJorge/mec-generator

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

45 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

PPP scripts instructions

All the scripts files related to Poisson Point Processes are located under the /pp directory.

Every script must be executed as follows:

cd pp/
Rscript --vanilla script.R arg1 arg2 ...

Below it is shown a table summarizing the scripts' usage:

name description usage
along-rural-road.R It generates a CSV with antennas along the A3 rural area road. The CSV with the rural area antennas must be generated. along-rural-road.R path/to/out.csv
madrid-center-femto-cells.R It generates a CSV with the femto-cells to be located in the Madrid skycraper madrid-center-femto-cells.R path/to/femto-cells.csv
gen-people-lambda.R It generates the people intensity function and store it as a matrix and two vectors that are the latitude and longitude samples. gen-people-lambda.R regionId longitudeSample latitudeSample matrixOUT lonsOUT latsOUT
macro-cells-factors.R Taking as input the people intensity matrix (and the corresponding longitude and latitude axis), it determines the parameters needed to obtain on average a certain amount of AAUs per km^2. See macro-cells-factors note for output details. macro-cells-factors.R peopleMat longitudeSamp latitudeSamp AAUs outSquareFactors
macro-cells-generator.R It generates a CSV file with macro-cell antennas locations based in the associated factors per each km^2. It is based on the people intensity matrix and the factors obtained with macro-cells-factors.R to obtain the desired average AAUs with repulsion. macro-cells-generator.R peopleMat longitudeSamp latitudeSamp squareFactors macroCellsCSV
mec-locations.R It generates the locations of MEC PoPs based on the antennas coordinates and the radio technology used. Several methods for the selection of such locations are available. See mec-locations note for more details. mec-locations.R regionID numLatSamples numLonSamples mecParamsJSON numMECs\<0 radioTech(FDD30kHz2sSPS, FDD120kHz7sSPS, TDD120kHz7sSPS) method(basicm1, basicm2, dig-outm1, dig-outm2, basicm1m2, dig-outm1m2) macro-cells.csv femto-cells.csv\NULL road-cells.csv\NULL MEClocations.csv intMatM1.csv\NULL intMatM2.csv\NULL
plot-mec-locs.R It plots the locations of the MEC PoPs with the M1 and M2 based intensity functions as heat-map along the locations. plot-mec-locs.R regionID latSamples lonSamples MEC_LOCATIONS_CSV MEC_M1_DIR\NULL MEC_M2_DIR\NULL OUT_PLOT_M1 OUT_PLOT_M2

Dependencies

The PPP scripts are written in R and they use the following packages:

SDMTools
cluster
ggmap
graphics
latex2exp
metR
pracma
rjson
spatstat
stringi

Script notes

Macro-cells factors generation

  • factor / avgAAUs: if people intensity function is multiplied by this at each km^2 square of the region, we get on average the AAUs passed by argument
  • sideSquares^2: the amount of cells in which the km^2 is divided. The repulsion radius of generated macro-cells will be 1 / (2*sideSquares)
  • smaller: determines if the square area is 1km^2 or less. If set to TRUE, then it is a limiting rectangle of the region

MEC locations

For the location of the MEC PoPs one can specify the NR radio technology used by he antennas:

carrier configuration frequency slot length uplink scheduling RTT
FDD 30 kHz 2 symbols SPS 0.78 ms
FDD 120 kHz 7 symbols SPS 0.66 ms
TDD 120 kHz 7 symbols SPS 0.78 ms

Note: both first and last configurations give same results.

As well several methods for the election of the MEC PoPs are available. The basic method is based on the reduction of the intensity function around the antennas once they are covered, whilst the dig-out is based on setting to zero the surroundings of the selected MEC PoP as well as the reduction around the covered antennas. The methods offer the possibility of selecting only locations at the M1, M2 ring or both. If both are selected, then it tries to minimize the number of MEC PoPs:

basicm1, basicm2, dig-outm1, dig-outm2, basicm1m2, dig-outm1m2

Automatic scripts

Under the /script directory there are some scripts that perform automatically specific tasks. Be sure you are inside the /script directory when you execute them.

script description usage
macro-cells-generation.sh It performs the necessary steps to generate macro-cells in the region contained in the REGION_ID variable inside the script. The generated macro-cells are located in a CSV under the /data/antennas/REGION_ID/macro-cells-TIMESTAMP/ ./macro-cells-generation.sh

Cite our work

If you use repo in your work, please consider citing our article:

@ARTICLE{8667076,
  author={J. {Martín-Pérez} and L. {Cominardi} and C. J. {Bernardos} and A. {de la Oliva} and A. {Azcorra}},
  journal={IEEE Transactions on Broadcasting},
  title={Modeling Mobile Edge Computing Deployments for Low Latency Multimedia Services},
  year={2019},
  volume={65},
  number={2},
  pages={464-474},
  doi={10.1109/TBC.2019.2901406}
}
J. Martín-Pérez, L. Cominardi, C. J. Bernardos, A. de la Oliva and A. Azcorra, "Modeling Mobile Edge Computing Deployments for Low Latency Multimedia Services," in IEEE Transactions on Broadcasting, vol. 65, no. 2, pp. 464-474, June 2019, doi: 10.1109/TBC.2019.2901406.

About

R scripts to generate operator infrastructure based on point processes

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published