Please use this identifier to cite or link to this item: https://doi.org/10.1109/PIMRC.2011.6140023
Title: On discretizing the exponential on-off primary radio activities in simulations
Authors: Wang, Z. 
Chew, Y.H.
Yuen, C.
Keywords: Cognitive radio
Geometric distribution
Opportunistic spectrum access
Traffic modeling
Issue Date: 2011
Source: Wang, Z.,Chew, Y.H.,Yuen, C. (2011). On discretizing the exponential on-off primary radio activities in simulations. IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC : 556-560. ScholarBank@NUS Repository. https://doi.org/10.1109/PIMRC.2011.6140023
Abstract: In opportunistic spectrum access (OSA), secondary radios (SRs) are allowed to access the channel whenever primary radios (PRs) are not transmitting. The PR's spectrum activities are normally assumed to be a 2-state continuous Markov on-off process which can be modeled as two independent exponential random variables. On the other hand, some simulations prefer using a 2-state discrete Markov chain. As SRs periodically sense the channel to determine its availability before transmitting, maintaining the same false alarm and missed detection probabilities is necessary but still insufficient when discretizing the continuous PR on-off process. Earlier work has shown that type II missed detection error, which is a function of spectrum sensing period, occurs because PR, which is sensed to be inactive in the last sensing duration, can become active before the next spectrum sensing is performed. Hence, it is insufficient to simply perform simulation at the time stamp of spectrum sensing period as information about the type II missed detection error will be lost. This paper looks into the context of spectrum sensing on how we should select the time stamp of the simulation, i.e., the transition probabilities of the 2-state discrete Markov chain, so as to approximate the 2-state continuous Markov process without losing any valuable information. We first derive the expression for type II missed detection error based on the discrete PR on-off model. Then we present the method to decide the time stamp of the simulation so that type II missed detection probability is kept within a given error bound. The same approach can be generalized to other probability distributions of the spectrum activities. © 2011 IEEE.
Source Title: IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC
URI: http://scholarbank.nus.edu.sg/handle/10635/71205
ISBN: 9781457713484
DOI: 10.1109/PIMRC.2011.6140023
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

2
checked on Dec 13, 2017

Page view(s)

7
checked on Dec 9, 2017

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.