Accelerating and Improving Simulation Performance in Communication Systems Modeling through Parallel Computing and Clustering


  • Kacper Kapusniak University College London
  • Tanmay Nautiyal The University of Hong Kong
  • Ryan Grammenos University College London



MATLAB, Parallel Computing, Cluster Computing, Signal Processing for 5G, Monte Carlo Simulations


Existing  5G  communication  systems  suffer  from  two  major  problems: the need  for  better  spectrum  efficiency  and  lesser  adjacent  channel  interference.  Thus,  development  of  novel  waveform  techniques  to  overcome  these  problems  is  a  major  topic  of  research  amongst  scholars  and  it  requires  carrying  out  Monte  Carlo  simulations  in  MATLAB©  (by MathWorks) to  measure  the  Bit  Error  Rate  (BER)  of  these  communication  models.  As  most  of  these  simulations  require  millions  of  computations,  they  take  a  significantly  long  time  to  run  (for  example,  days)  as  they  run  on  single-core  machines  and  carry  out  the  computations  serially.  The  main  objective  of  this  research  is  to  reimplement  current  scripts  using  various  parallel  computing  techniques  in  MATLAB  to  study  which  one  is  the  best suited  for  this  particular  type  of  simulations  while  also  scaling  these  scripts  onto  a  multi-core  cluster  to  further  improve  the  execution  time.


Download data is not yet available.


Metrics Loading ...

Author Biographies

Kacper Kapusniak, University College London

Kacper Kapusniak is an undergraduate student at University College London (UCL). He was the recipient of the Undergraduate Research Opportunities Scheme (UROS) studentship. Kacper's research interests lie in the areas of machine learning and parallel computing.

Tanmay Nautiyal, The University of Hong Kong

Tanmay Nautiyal is an undergraduate computer science student at the University of Hong Kong. He was the recipient of the prestigious Laidlaw Research Scholarship, and his research interests lie primarily in parallel computing and machine learning.

Ryan Grammenos, University College London

Dr Ryan Grammenos is a Lecturer (Teaching) in the department of Electronic and Electrical Engineering at University College London (UCL), United Kingdom. He received a first-class honours BEng degree in Electronic Engineering from Cardiff University in 2007 and an MEng degree in Electrical and Electronic Engineering from the University of Nice-Sophia Antipolis, France, in 2008. He graduated with an Engineering Doctorate from UCL in 2013 in Communications focusing on the mathematical modelling and hardware realisation of novel communication transceivers. Ryan's research interests lie in the area of signal processing for communications, software defined radio and the Internet of Things. Ryan is also a strong supporter of innovation and entrepreneurship and was previously responsible for facilitating the technology exchange hub ( within the Faculty of Engineering at UCL. He is a Senior Fellow of the Higher Education Academy (SFHEA), committed to continuous professional development and eager in promoting STEM education.



How to Cite

Kapusniak, K., Nautiyal, T., & Grammenos, R. (2021). Accelerating and Improving Simulation Performance in Communication Systems Modeling through Parallel Computing and Clustering. Journal of Student Research, 9(2).



Research Articles