DSpace at VNU: Efficient re-use of CRs under deep fading for improving cooperative sensing performance

Efficient Re-use of CRs under Deep Fading for Improving Cooperative Sensing Performance Dinh Thi Thai Mai, Nguyen Quoc Tuan University of Engineering and Technology, Vietnam National Un

Efficient Re-use of CRs under Deep Fading for Improving Cooperative Sensing Performance

Dinh Thi Thai Mai, Nguyen Quoc Tuan

University of Engineering and Technology, Vietnam National

University, Hanoi - Vietnam Email: dttmai@vnu.edu.vn, tuannq@vnu.edu.vn

Dinh-Thong Nguyen Faculty of Engineering and Information Technology, University of Technology, Sydney – Australia Email: dinh-thong-nguyen@eng.uts.edu.au

Abstract – It is well known in the research field of cooperative

spectrum sensing that deeply faded local cognitive sensors need to

be eliminated from contributing to the fusion pool at the fusion

center in order to improve global detection reliability Also, most

current works on the topic unrealistically assume that the reporting

channels between the cognitive sensors and the fusion center are

free of loss and fading This paper proposes an innovative

technique to re-use those eliminated sensors by reassigning them to

act as cooperative diversity relays to assist the surviving sensors in

combatting outages due to Rayleigh fading in the reporting

channel

Cooperative spectrum sensing using cognitive radios (CRs)

has become a popular technique in sensing a primary user

under a multipath fading and/or shadowing environment [1]

[2] The sensing is done in two stages: in the sensing stage, the

CRs independently measure and process the signal from the

primary user, and in the reporting stage the CRs independently

report the processed information to a fusion center (FC) which

is usually a cognitive base station, to make the final global

sensing decision as to whether the primary user is present or

absent Most researchers, however, do not take into account

the differences in local transmission quality and reliability

between individual sensors Signal-to-noise ratio (SNR) is a

dominant metric of transmission quality affecting the detection

performance of a CR sensor This is particularly true in the

most common sensing technique using energy detection [3]

[4] When the popular k-out-of-N decision fusion rule is used,

it is easy to see that the inclusion of deeply faded CRs, i.e

with low SNRs, in the decision fusion at the fusion center

diminishes the reliability of the cooperative CR detection of

the primary user In [5] the authors presented a deep fading

scenario under correlative shadow fading and proved that by

discarding the detection contribution from shadowed CRs, the

detection probability of the centralized cooperative CR

network was improved The CRs under shadowing are

therefore wasted In this paper, we are not interested in how to

identify the unreliable or low quality CRs and how many are to

be discarded from the fusion process, but only in how to re-use

these CRs to assist channels between the CRs and the fusion

center With the knowledge of the SNRs of the primary user’s

signal received at indidual CRs, i.e SNRs of the sensing

channels, the FC can use various algorithms to handle this

issue

In most current research publications on cooperative spectrum sensing using CRs, the wireless links between individual CRs and the fusion centre, i.e the reporting

channels, are assumed lossless In reality, when these channels

are of significant distance in suburban macrocells or under shadowing in urban microcells, loss and fading is a significant issue Cooperative diversity relaying has become a popular research topic for combating serious fading problems Once a channel is in deep fade, advanced message coding is no longer effective in improving transmission reliability, and cooperative diversity transmission has proved to dramatically improve the performance of wireless transmission In most practical situations, a wireless channel is non-ergodic and its capacity is

a random variable, thus no transmission rate is reliable In this

case, the outage probability is defined as the probability that

the instantaneous random capacity falls below a given threshold, and capacity versus outage probability is the natural information theoretic performance measure [6] The cooperative diversity relay can process the information it receives from the source and forward the information to the

destination using either amplify-and-forward (AF) or decode-and-forward (DF) protocols

In this paper, we propose an efficient re-use of those CRs under deep fading of the sensing channels by re-assigning them to act as diversity relays to their more healthy peers in the reporting channels, thus improving the global detection reliability of the fusion center

II COOPERATIVESPECTRUMSENSING

A Sensing System Definition

Figure 1 shows a cooperative spectrum sensing CR network in which local CRs detect signals from the primary user, perform local processing, then send their detected signals

{y i , i=1, , N} to the fusion centre (FC) for the latter to make

the final global sensing decision as to whether the primary user

is absent (Η ) or present (0 Η ) In many applications, the CRs 1

report their local {SNR i }and {P Di}to the fusion centre via the reporting wireless channels In this paper, for simplicity of analysis, we assume that reporting channels are subject to Rayleigh fading only, so that we can concentrate on the main theme, that is the principle of re-use of those CRs which are otherwise discarded from the cooperative sensing process by the fusion centre

B Detection and False Alarm Probabilities over Fading

Channels

There are two probabilities that are most commonly used

as performance metrics for spectrum sensing: probability of a

false alarm P F which is the probability that a signal is detected

while it does not exist, and probability of detection P D which is

the probability that the presence of a signal is correctly

detected

Probabilities of false alarm is given in [1] as

( )

m

λ

Γ (1)

where λ is the power detection threshold, Γ(.) and Γ(.,.) are

complete and incomplete gamma functions, respectively

P D in fading conditions is obtained by averaging its value in

the AWGN case over the SNR fading distribution f(γ), while

P F remains the same under all fading conditions since it is

calculated under Η hypothesis, i.e no signal, hence 0

independent of SNR, i.e

1

P =∫∞f γ Q γ G− P dγ (2)

Where γ is the SNR, Q m (.,.) is the generalized Marcum’s Q

function of 2m degree of freedom

The exponential pdf of SNR in a Rayleigh fading channel is

f( )γ 1exp( γ)

= − , (3)

By replacing (3) into (2), we obtain the average probability of

detection of the local CR subject to Rayleigh fading [3] as

1

m

P D Rayleigh

m m

m

m

γ

λ γ γ γ

Γ −

(4)

The behavior of P D versus P F curve represents the receiver

operating characteristics (ROC)

C Cooperative Spectrum Sensing

In this paper, we assume that individual CRs send their P Di,

P Fi , and received signal power SNR i , i.e a de-centralized case

Note that these are {SNR i} of the sensing channels. Let u i =

[0,1], i=1,2,…,n, denote the 1-bit decision from the ith CR If

the individual measurements are mutually independent, then

the k-out-of-n rule is most popular by which hypothesis Η is 1

chosen if k or more individual decisions are 1, and Η is 0

chosen otherwise Under this rule the global average

probabilities are

1

1( ) (1i ) i

i

u j

n

=

−

∑ (5) and

1 1

i

n n

u u

j k u j i

∑

Figure 1 : A decentralized cooperative spectrum sensing network with the Nth

CR sensor eliminated from the fusion process III COOPERATIVE DIVERSITY RELAYING

A Cooperative Relaying System Definition

Figure 2 shows a simple cooperative diversity relay

network using M relaying branches The system definition

below is taken from [8] In the application for CR reuse in this paper, a Source is a reliable CR, the Destination is the FC and

a Relay is an unreliable CR being reused The relays are assumed to operate in the time division mode having two phases: the relay-receive phase and the relay-transmit phase;

each phase or sub-block is of duration T/2 There is no

correlation between the source transmit signal and the relay transmit signal

In the relay-receive phase at discrete time m =1,2,…T/2,

the source transmits the complete message of symbols to both the destination and the relays (broadcast mode) in the AF case,

but only to the relays in the DF case, i.e only (7a) applies

y m = P m h x m s +n m (7a)

sd s sd s sd

y m = P m h x m +n m (7b)

d rd

Phenomenon

{H H, }

Fusion Center

i i

i P P SNR D F i

δ =

u 0 {0,1}

P D1 P D2 … P DN

d sr

d sd

Sensing channel

Reporting channel

In the decode-and-forward (DF) relaying protocol, the

relay detects by fully decoding the entire codeword it receives

from the source in the relay-receive phase, symbol by symbol,

then retransmits the signal, after recoding it, to the destination

during the relay-transmit phase

In the relay-transmit phase at discrete time m=T/2+1,

T/2+2,….T, the relays send their signals to the destination and

the source may or may not send signal to the destination

depending on the relaying protocol used (multiple access

mode) The received signal at the destination is

1

M

i

=

∑ (8)

Figure 2: Diagram of an M-relay cooperative diversity relaying network

where x, y, n, and P are the normalized transmit signal, i.e

E x = , the corresponding received signal, the additive

noise which is modeled as a circularly symmetric complex

Gaussian random variable with zero mean and variance σ2 at

the receiver, i.e n ~ N (0,σ2 ), and the transmit power,

respectively The parameters’ double subscript ij is to mean

being associated with the channel link from i to j h ij is the

channel gains (or loss) from node i to node j, being subject to

frequency nonselective Rayleigh fading, and is modeled as

independent, circularly symmetric, complex Gaussian random

variables with zero mean and variance µ ij It is well known that

under Rayleigh fading,h ij 2 is exponentially distributed

We define the instantaneous signal-to-noise ratio (SNR) in

the received signal as

2

2 2

ij i

ij ij

ij

σ

= = (9)

where γijAWGN is the SNR of the unfaded AWGN channel

For convenience, and to be consistent with many papers on the

subject, in this paper the same as in [8], we simply use SNR to

mean γAWGN.

Under Rayleigh fading, SNR in (9) is an independent

exponential random variable with expected (average) value

ij ij i2 ij

ij

P SNR

μ

σ

= = (10) The outage probability, i j ( )

ou

h t t h

P μ of the wireless link between

two points i and j having instantaneous channel gain h ij for a

given outage information rate threshold, R th is defined as

out

h th ij th h th

P SNR R = h <μ =F μ (11) where the channel gain threshold is defined as

( 1)

2M Rth 1

th

SNR

and M is the diversity order

B Outage Probability of a Single Wireless Link

Since the channel gain has the exponential distribution as

in (3) with mean µ ij, the outage probability of the direct link between the source and the destination (without relaying), is simply

( ) ( ) 1

th sd sd

out

sd th h th

μ μ

μ = μ = − (12)

C Outage Probability of Cooperative DF RelayingNetwork

It is not the subject of this paper to study various cooperative diversity relaying protocols We choose to use the

selection DF (SDF) relaying protocol as an example to

demonstrate the benefit of sensors’ re-use in cooperative

spectrum sensing in a deep fading environment In SDF

relaying protocol, when the relay is not able to decode the source message, i.e the source-relay link is in outage, the source repeats its transmission to the destination on the direct link The maximum average information rate in this case, i.e

first part of (13) is that of repetition coding The information

rate of a selection DF relay network can be expressed as below [7]

1log 1 2( )

2 1

2

sd sr th

sd rd sr th

ISDF

⎪⎪

= ⎨

⎪⎩

(13)

In SDF relaying protocol, it is assumed that a reused CR has to

have the knowledge of the CSI, i.e γ sr, between itself and the

CR that it assists Its outage probability under exponential

fading condition is therefore [8]

2

out

SDF th SDF th

P μ = h ≤μ

sd th sr th

sr th sd rd th

2

th th

sd sr

th

th th sr

sd rd

sd rd

sd rd

μ

μ

−

(14)

D Algorithm for Re-use of CRs in Cooperative Sensing under

Deep Fading

In cooperative spectrum sensing, a CR in deep fading will

be eliminated from the fusion process by the fusion center if its

SNR is lower than a given threshold [4] [5] determined by the

FC Because the reporting channels between the CRs and the

FC are subjected to fading and hence link outages, the reported

parameters received at the FC are usually deteriorated The

effective probability of detection received from the ith CR by

the FC is

P De( )i =P Di{1−P out( )}i (15)

where P out (i) is the outage probability of the channel between

the ith CR by the FC

Figure 3: A simulation scenario of Cooperative Spectrum Sensing under a

multipath fading and shadowing environment

We propose a method that utilizes eliminated CRs by

assigning them to act as cooperative diversity relays for the

surviving CRs that still stay in the fusion pool In Figure 1, we

illustrate an example in which CR N suffers from deep fading

and its reported information is discarded by the fusion center

The fusion centre then reassigns CR N to act as a relay for CR1

to improve the transmission reliability of the CR1-to-FC

reporting channel, thus forming a 3-terminal relay network

with the surviving CR1 as the source (S), the eliminated CR N as

the relay (R) and the FC as the destination (D) It is expected

that the outage probability in (15) is lower with the diversity gain by the use of the relay compared to that without the relay, hence improving the local effective probability of detection received by the fusion center

E Algorithm

• Step 1: Local CRs calculate the SNR {SNR i ;i=1,2, N }

from their respective raw measured data {y i}and calculate

the probability of detection {P Di}using (4)

• Step 2: The fusion center receives the processed information in Step 1 from the local CRs Note that having received individual SNRs (of the sensing channels)

from the CRs, the FC can evaluate the individual P Ds

instead of the CRs However, such evaluated P D s will be

unreliable due to fading in the reporting channels

• Step 3: The FC ranks the CRs in the descending order of their SNR, so that SNR1>SNR2> >SNRN and the CRs are indexed in this same order

• Step 4: The FC sets a SNR threshold, SNRth, below which

the corresponding probability of detection, P i , is eliminated from contributing to the sensing fusion The

M<N surviving CRs and the (R=N-M) eliminated CRs

are:

SNR 1 >SNR 2 > >SNR M-R >SNR M-R+1 > >SNR M-1 > SNR M > SNR th >SNR M+1 > >SNR M+R

• Step 5: Assume that we use all R discarded CRs as relays Assign CR M+1 to be the relay for CR M , CR M+2 for CR

M-1 and CR M+R for CR M-R-1

• Step 6: Calculate outage probability of the direct link of the surviving CRs, i.e P SD out( )i (i=1,2, ,M), using (12)

• Step 7: Calculate P SDF out ( )i (i=M-R+1, M) of the resulting

R cooperative diversity relay networks using SDF

protocol in (14)

• Step 8: Calculate the local effective probability of

detection {P De (i)}(i=1,2, ,M) from (15)

• Step 9: Calculate the global probability of detection, Q D, using (6) for with (in Step 7) and without (in Step 6) cooperative diversity relaying

• Step 10: Plot the ROC with and without using relays

IV RESULTS AND CONCLUSIONS

In Figure 3, we illustrate a fading scenario of the sensing

channels from a TV station to the CRs some of which suffer

from deep fading such as shadowing from trees and buildings, giving very low SNRs and consequently their local reported contributions are discarded by the fusion centre Again, the exact physical location of the CRs is not an issue in this paper;

we assume for simplicity of the simulation that all CRs are approximately at equal distances to the cognitive base station

(CRBS) and that the CRs are distributed at equal distances

from one another

As in [8], we simulate a fading scenario of the sensing

channels in which CRs 7,8,11,12 are subject to log-normal

fading and the rest are under Rayleigh fading The SNRs in

the raw signals measured at the 12 CRs are {SNR i}=[10, 9, 3,

7, 8, 9, -3, -6, 12, 6, 0.2, 1] dB, and the threshold set by the

FC is SNR th=0.5 dB Thus the three CRs 8, 7 and 11 are

excluded from the decision fusion and reassigned to act as

relays for CRs 10, 3 and 12, respectively, in the reporting

channels The resulting 3-node cooperative diversity relay

networks are assumed to be subjected to exponential fading

with realization mean parameters (µ sd , µ sr , µ rd) = (0.01, 1,

0.01); (0.01, 0.1111, 0.01) ; (0.01, 0.04, 0.01)

(a)

(b) Figure 4: Sensing performance of the cooperative spectrum sensing network

under fading with and without re-use of poor CRs as diversity relays,

(a) Outage threshold µ th =0.001 and (b) Outage threshold µ th = 0.005

Figures 4a and 4b show the simulation results of the receiver operating characteristic (ROC) curves of the cooperative

sensing network in three operating environments: transparent reporting channels (no fading reporting channels without

re-use of discarded CRs as diversity relays, and Rayleigh fading

reporting channels with re-use of discarded CRs as diversity

relays All ROC curves are for the case of three re-used CRs and the AND fusion rule is used for simplicity The outage

channel gain threshold µ th is much lower, i.e much healthier

SNR, in Figure 4a than in Figure 4b giving much higher

probability of detection, and therefore the benefit of re-use of deeply faded CRs is not as pronounced in Figure 4a compared

to Figure 4b However, in both cases the benefit of the proposed re-use of poor sensing CRs as diversity relays is very convincing

The idea proposed in this paper is innovative and many issues need to be investigated such as how to optimally

determine the SNR threshold SNR th and how to optimally reassign the eliminated CRs to assist the surviving CRs in order to preserve the total power or to maximize the final probability of detection These issues are being investigated by our group

ACKNOWLEDGEMENTS This work was supported by a research grant from QG.10.44 -TRIGB and QC 09.28 Projects of the University of Engineering and Technology, Vietnam National University Hanoi

REFERENCES [1] A Ghasemi and E.S Sousa, “Collaborative Spectrum Sensing for

Opportunistic Access in Fading Environments,” Proc IEEE 1 st Symposium on Dynamic Spectral Access Networks (DySPAN’05), pp 131-136, Baltimore,

November 2005

[2] A Ghasemi and E.S Sousa, “Opportunistic Spectrum Access in Fading

Channels Through Collaborative Sensing,” Journal of Communications, vol

2, No 2, pp 71-82 , March 2007

[3] F.F Digham, M.S Alouini and M.K Simon, “On the energy detection of

unknown signals over fading channels,” Proc IEEE Int Conf on Coms

ICC’03, pp.3575-3579, May 2003

[4] Yi Zheng et al “Cooperative Spectrum Sensing Based on SNR Comparison in Fusion Center for Cognitive Radio,” Proc International

Conference on Advanced Computer Control, ICACC’2009, pp 212-216, 22-24

Jan 2009

[5] DTT Mai et al “Improving Cooperative Spectrum Sensing under Correlated Log-normal Shadowing,” Proc International Conference Cyber

2010, October,2010

[6] L.H Ozarow et al., “Information theoretic considerations for cellular

mobile radio,” IEEE Trans on Vehicular Technology, vol 43, no 2,

pp.359-377, May 1994

[7] J.N Laneman et al., “Cooperative Diversity in Wireless Networks:

Efficient Protocols and Outage Behavior,” IEEE Trans Inform Theory, 50

(12), pp 3062-3080, Dec 2004

[8] D.T Nguyen et al “Outage Probability Analysis of Cooperative Diversity

DF Relaying under Rayleigh Fading,” Submitted to ATC’11 International

Conference, Aug 2011