Implementation of the Modified Monte Carlo Technique Using Importance Sampling on the Block Oriented System Simulator
AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH SCHOOL OF ENGINEERING
Pagination or Media Count:
The purpose of this thesis was to implement the Modified Monte Carlo technique using Importance Sampling on the Block Oriented System Simulator BOSS. Computer simulation techniques of communications systems were reviewed. Next, conventional Monte Carlo techniques and Modified Monte Carlo techniques using Importance Sampling were reviewed. Models of Binary Phase Shift Keying BPSK systems using both Monte Carlo techniques were implemented and simulated. Reasons for the model using Importance Sampling not working correctly are postulated. The Monte Carlo technique is a method of ensuring the an inherently infinite procedure, such as determining system bit error rate BER, can be determined within an appropriate accuracy and a confidence range after a set number of samples. Conventional Monte Carlo requires a certain number of samples to be generated to determine a certain BER. This number of samples results in an estimated BER in the range of 0.5 to 2.0 of the true BER. The number of samples required using conventional Monte Carlo techniques can result in unacceptable simulation times for low probability events. Importance Sampling is a method of reducing the number of samples required to determine an estimated BER with the same accuracy and confidence as conventional Monte Carlo.
- Computer Programming and Software
- Command, Control and Communications Systems