Restricted Neyman-Pearson approach based spectrum sensing in cognitive radio systems
Author(s)
Advisor
Gezici, SinanDate
2012Publisher
Bilkent University
Language
English
Type
ThesisItem Usage Stats
146
views
views
40
downloads
downloads
Abstract
Over the past decade, the demand for wireless technologies has increased enormously,
which leads to a perceived scarcity of the frequency spectrum. Meanwhile,
static allocation of the frequency spectrum leads to under-utilization of
the spectral resources. Therefore, dynamic spectrum access has become a necessity.
Cognitive radio has emerged as a key technology to solve the conflicts
between spectrum scarcity and spectrum under-utilization. It is an intelligent
wireless communication system that is aware of its operating environment and
can adjust its parameters in order to allow unlicensed (secondary) users to access
and communicate over the frequency bands assigned to licensed (primary)
users when they are inactive. Therefore, cognitive radio requires reliable spectrum
sensing techniques in order to avoid interference to primary users. In this
thesis, the spectrum sensing problem in cognitive radio is studied. Specifically,
the restricted Neyman-Pearson (NP) approach, which maximizes the average detection
probability under the constraints on the minimum detection and false
alarm probabilities, is applied to the spectrum sensing problem in cognitive radio
systems in the presence of uncertainty in the prior probability distribution of
primary users’ signals. First, we study this problem in the presence of Gaussian
noise and assume that primary users’ signals are Gaussian. Then, the problem
is reconsidered for non-Gaussian noise channels. Simulation results are obtained
in order to compare the performance of the restricted NP approach with the
existing methods such as the generalized likelihood ratio test (GLRT) and energy
detection. The restricted NP approach outperforms energy detection in all
cases. It is also shown that the restricted NP approach can provide important
advantages over the GLRT in terms of the worst-case detection probability, and
sometimes in terms of the average detection probability depending on the situation
in the presence of imperfect prior information for Gaussian mixture noise
channels.