Abstract
In this investigation, we present a spectrum sensing
algorithm for distributed cognitive applications in the presence
of time and frequency double-selective channels (TF-DSFCs).
Firstly, a dynamic discrete state-space (DSM) model is established
to characterize the sensing process. With this new paradigm,
spectrum sensing is then realized by acquiring primary states
and time-dependent multipath response jointly. To accomplish
this, an efficient estimation algorithm is designed by resorting
to a sequential importance sampling technique, in which the
time-varying non-Gaussian multipath channel and primary state
are recursively recovered. As demonstrated by simulations, the
proposed algorithm may effectively enhance sensing performance
by exploiting the blindly recovered channel response.
algorithm for distributed cognitive applications in the presence
of time and frequency double-selective channels (TF-DSFCs).
Firstly, a dynamic discrete state-space (DSM) model is established
to characterize the sensing process. With this new paradigm,
spectrum sensing is then realized by acquiring primary states
and time-dependent multipath response jointly. To accomplish
this, an efficient estimation algorithm is designed by resorting
to a sequential importance sampling technique, in which the
time-varying non-Gaussian multipath channel and primary state
are recursively recovered. As demonstrated by simulations, the
proposed algorithm may effectively enhance sensing performance
by exploiting the blindly recovered channel response.
Original language | English |
---|---|
Title of host publication | Global Communications Conference (GLOBECOM), 2014 IEEE |
Publisher | IEEE |
Pages | 3051-3056 |
Number of pages | 6 |
DOIs | |
Publication status | Published - Dec 2014 |
Event | Global Communications Conference (GLOBECOM) - Texas, Austin, United States Duration: 8 Dec 2014 → 12 Dec 2014 |
Conference
Conference | Global Communications Conference (GLOBECOM) |
---|---|
Country/Territory | United States |
City | Austin |
Period | 8/12/2014 → 12/12/2014 |