Transcript Poster

Prashant Bajpayee Advisor: Dr. Daniel Noneaker
Motivation
•Currently most radio-frequency spectrum is assigned
exclusively to “primary” users (e.g. Broadcasting of T.V.
channels)
•Assigned spectrum is often underutilized in many
locations at specific times
•This “unused ”spectrum is lost opportunity for other
potential communication systems
General Solution
Exploitation of unutilized spectrum on “secondary” user
basis and Cognitive radio is viewed as a novel approach
for efficient utilization
•Assume DS Spread Spectrum binary antipodal signaling is
used by transmitter so that
System Details
• Multiple – channel communication system
N 1


bk
s(t )   (1)  A x( k 1) N i (t  iTc ) 
k 0
 i 0

•During monitoring of traffic channel j its received signal is
andr (t )  n j (t )
independent random
n1(t )' n2are
( t )'    n j ( t )
process
•Two types of channel:
•Good channel with lesser interference
•Bad channel with more noise interference


• Channel monitoring approach used by destination:
– When no data is transmitting, sequentially monitor
channel to determine level of noise in each channel
– When RTS is received,destination should send CTS
specifying channel with least intereference.
Pe Vs n
0
channel=2
channel=4
channel=6
channel=8
c=2;simulation error
Key properties of Cognitive Radio
•This Spectrum monitoring capability
•Dynamic spectrum assignment protocols/rules
Purpose of Research
•Development and simulation of a mathematical model
of channel monitoring function of CR
•Analyze the effect of time-varying interference in each
channel and design trade-offs that result
Acknowledgement
Error Probability(Pe)
c=3;analysis error
c=5;simulation error
c=5;analysis error
-2
10
-3
10
-4
10
•Dr. Xiao-Bang Xu
•Dr. Michael Pursley
•NSF
mTc
( m  n )Tc
c(t ) 
 r (t )c(t ) sin( w t   )dt
i
1
mTc

 x  (t  iT )
i  
i
c
Detection Technique
Generalized Model
• CR alters between frequency monitoring and other
tasks
• Each channel has time-varying interference/noise
• Now each channel is represented as two-state
Markov random process
1
• S k = state of channel 1 at time k, may be 0 or 1
-1

-2
0
5
10
15
n
20
25
30
10
35
Conclusions
•On increasing n, error probability decreases.
•On increasing c, error probability decreases.
Pe Vs n(for diff. efficiency)
10
0
100
200
300
400
500
T-coh
600
700
0
0
10
-1
10
-2
10
0
5
10
15
20
25
n
30
35
40
45
50
Conclusions
•For fixed efficiency and T-coh, there is an optimal n.
•Increasing efficiency, increases error probability.
•Increasing efficiency, decreases n-optimal.
T-coh=100
T-coh=900
T-coh=3600
-2
10
0
5
10
15
20
25
n
30
35
40
45
50
Simulation Details
n = time required to generate each statistics
and qual to number of chi-square random
variables which add up to form a test statistic
n1 = the time radio spends on other tasks
T_coh = measure of rate of change of states
1 
1
•Now each channel can be said good or bad depending
upon its state at that time
t=0
-1
0

Monitor
f2
Monitor
f1
10
n1/n=1
n1/n=9
n1/n=49
1
800
Conclusions
•For large T-coh, increasing channel , error probability decreases.
•For moderate T-coh, there is an optimal number of channel.
Pe Vs n(for diff. T-coh)
n
work
n1 =q*n
-----
n+n1
work
2n+n1
2n+2n1
Ln+Ln1
•All Channels changes their states after each time interval
=n
•New decision statistics are generated after n+n1 time
interval but only for one channel at a time
•Efficiency of system = n1/n+n1
Conclusions
•On increasing n, error probability decreases.
•On increasing c, error probability decreases.
Future work
•Dr. Daniel Noneaker
•Dr. Harlan Russell
Y1 
10
Error probability(Pe)
Advantages
• Highly reliable communications whenever and
wherever needed
• Efficient utilization of the radio spectrum
• Dynamically reuse available spectrum by changing its
parameter
10
Error Probability(Pe)
•Reconfigurable
•CR has the capability of reconfigurability which is
provided by a platform known as Software- defined
radio on which CR is built
Error probability(Pe)
c=2;analysis error
c=3;simulation error
•Monitor channel 1 by forming statistic R 2  X 2  Y 2
1
1
1
( m  n )Tc
Where X 1   r (t )c(t ) cos( wi t  1 )dt
Similarly for all channels we generate chi-square
random variable Rj with degree 2n and pick
channel with smallest value of Rj
10
-1
Decision Statistic

Pe Vs T-coh
0
10
Characteristic of CR
•Intelligent wireless communication system
• Aware of its environment
•Methodology of understanding-by-building to learn
from environment
•Adapt to statistical variances in the input-stimuli
SURE 2005
• c = ratio of noise variances of good and bad channels
• Assume   
1 / T _ coh
    1 / 2 1  1 / e
•Consider a more general Markov model with different
parameter
• Pe = Probability of selecting BAD channel
• Channel = no. of channels in radio.
• Analyze the effect of mutual coupling amongst different
radios on channel-noise distribution of each radio.


•Examine the performance in terms of bit error probability