Power Control in Cognitive Radio Systems Based on Spectrum ...
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the ICC 2007 proceedings.
Power Control in Cognitive Radio Systems Based on Spectrum Sensing Side Information Karama Hamdi, Wei Zhang, and Khaled Ben Letaief, Fellow, IEEE Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology Kowloon, Hong Kong Email: {eekarama, eewzhang, eekhaled}@ust.hk Abstract— Cognitive radio has been recently proposed as a promising technology to improve the spectrum utilization efficiency by intelligently sensing and accessing some vacant bands of licensed users. In this paper, we consider the coexistence between a cognitive radio and a licensed user in order to enhance the spectrum efficiency. We develop an approach to allow the cognitive radio to operate in the presence of the licensed user. In order to minimize the interference to the licensed user, the transmit power of the cognitive radio is controlled by using the side information of spectrum sensing. Numerical results will show that the quality of service for the licensed user can be guaranteed in the presence of the cognitive radio by the proposed approach.
I. I NTRODUCTION The explosive growth in wireless services over the past several years illustrates the huge and growing demand of the business community, consumers and the government for wireless communications. With this growth of communication applications, the spectrum becomes more congested. Even though the Federal Communications Commission (FCC) has expanded some spectral bands, these frequency bands are exclusively assigned to specific users or service providers. Such expansion does not necessarily guarantee that the bands are being used most efficiently all the time. Recent survey has in fact proved that most of the radio frequency spectrum is vastly under-utilized [1], [2]. For example, cellular network bands are overloaded in most parts of the world but amateur radio or paging frequencies are not. Moreover, those rarely used frequency bands are assigned to specific services that cannot be accessed by unlicensed users, even if the transmission of the unlicensed users does not introduce any interference to the licensed service. To deal with the conflicts between spectrum congestion and spectrum under-utilization, cognitive radio has been recently proposed as a smart and agile technology which allows nonlegitimate users to utilize licensed bands opportunistically [3], [4]. By detecting particular spectrum holes and jumping into them rapidly, the cognitive radio can improve the spectrum utilization significantly. To guarantee a high spectrum efficiency while avoiding the interference to the licensed users, the cognitive radio should be able to adapt spectrum conditions flexibly. Hence, some important abilities should be provided by the cognitive radio which include spectrum sensing, dynamic frequency selection and transmit power control [5].
One of the most challenging problems of cognitive radio is the interference which occurs when a cognitive radio accesses a licensed band but fails to notice the presence of the licensed user. To address this problem, the cognitive radio should be designed to co-exist with the licensed user without creating harmful interference. Recently, several interference mitigation techniques have been presented for cognitive radio systems. An orthogonal frequency division multiplexing (OFDM) was considered as a candidate for cognitive radio to avoid the interference by leaving a set of subchannels unused [6]. Thus, it can provide a flexible spectral shape that fills the spectral gaps without interfering with the licensed users. A transform domain communication system (TDCS) was proposed to mitigate the interference by not putting the waveform energy at corrupted spectral locations [7]. A power control rule was presented to allow cognitive radios to adjust their transmit powers in order to guarantee a quality of service to the primary system [8]. To avoid the interference to the licensed users, the transmit power of the cognitive radio should be limited based on the locations of the licensed users [8]. However, it is difficult to locate the licensed users for the cognitive radio in practice because the channels between the cognitive radio and the licensed users are usually unknown. Furthermore, the environment where the system is in operation may have large delay spread and hence the channel model is complicated by fading, shadowing and path loss effects. In [9], the local oscillator (LO) leakage power was exploited to locate the primary receivers. But it is still not easy to apply this in practice because the approach requires a sensor node mounted close to the primary receivers to detect the LO leakage power. In this paper, we present a power control approach in cognitive radio systems based on spectrum sensing side information in order to mitigate the interference to the primary user due to the presence of cognitive radios. This approach consists of two steps. Firstly, the shortest distance between a licensed receiver and a cognitive radio is derived from the spectrum sensing side information. Then, the transmit power of the cognitive radio is determined based on this shortest distance to guarantee a quality of service for the licensed user. Because the worst case is considered in this approach where the cognitive radio is the closest to the licensed user, the proposed power control approach can be applied to the licensed user in any location. The rest of this paper is organized as follows. In Section
1-4244-0353-7/07/$25.00 ©2007 IEEE 5161 Authorized licensed use limited to: National Taiwan University. Downloaded on July 19, 2009 at 10:28 from IEEE Xplore. Restrictions apply.
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the ICC 2007 proceedings.
Rd
d
µ Rp PTx
∆
PRx
ψ
CR
Fig. 1. System model. PTx, PRx and CR denote primary transmitter, primary receiver and cognitive radio, respectively.
II, the system model is given. In Section III, a power control approach is proposed for cognitive radio systems based on spectrum sensing side information. Numerical results are presented and discussed in Section IV. Finally, conclusions are drawn in Section V. II. S YSTEM M ODEL We consider a scenario in which a primary system (licensed user), formed by a transmitter-receiver pair, co-exists in the same area with a secondary user (cognitive radio). The system model of our interest is illustrated in Fig. 1. In the primary system, the primary transmitter communicates with the primary receiver with the transmit power Qp . Some system parameters which are shown in Fig. 1 are explained as follows. The circle around the primary transmitter with the radius Rd (m) represents the decodable region within which the signal-to-noise ratio (SNR) of decodability occurs in the absence of interference to the primary receiver. The circle around the primary transmitter with the radius Rp (m) denotes the protection region within which the primary receiver must be guaranteed successful reception even in the presence of the cognitive radio. ∆ (dB) is the signal attenuation due to the distance Rd . µ is the margin of protection in dB which represents how much interference above the noise floor the primary system can tolerate [8]. In the secondary system, a cognitive radio is considered to work in the same frequency band as the primary system. Before accessing the channel, the cognitive radio acts as a listener to detect from the received signals whether the primary system is in operation. Let d (m) denote the distance between the primary transmitter and the cognitive radio. In practice, it is difficult to obtain the value of d because the signals from the primary transmitter and the channel are both unknown to the cognitive radio. Thus, computing d will be a challenging problem. Another challenging issue is to allow the cognitive radio to access the same spectrum band where the primary user is operating. In such a case, the cognitive radio may interfere with the primary system, thereby, degrading the quality of
service for the primary receiver. To reduce the interference, the transmit power Qc of the cognitive radio should be limited based on the tolerable interference to the primary receiver which directly depends on the distance between the cognitive radio and the primary receiver. However, it is difficult for the cognitive radio to locate the primary receiver which can be at any location inside the protection region. To address this problem, the worst case scenario is investigated in this work where the primary receiver is located on the crossing point between the boundary of the protection region and the line from the primary transmitter to the cognitive radio as shown in Fig. 1. By limiting the transmit power of the cognitive radio for the above worst case, we can guarantee a good quality of service for the primary receiver in operation at any location inside the protection region. In this case, it can be seen from Fig. 1 that the transmit power of the cognitive radio which is allowed to inflict tolerable interference on the primary receiver depends on the SNR loss (µ+ψ) in dB. Note the SNR loss due to the distance d is given by η ψ + ∆ (dB). Then, the transmit power control problem is essentially converted to the problem of evaluating the SNR loss η due to d for a given µ and ∆. We assume that the channel between any two terminals in Fig. 1 experiences flat Rayleigh fading and path loss. The propagation power attenuation is characterized by Q(r) = r−α , where r represents the distance and α denotes the power loss exponent which is usually a constant in the range 2 ∼ 6. Throughout this paper, α = 2 is used which corresponds to the free-space attenuation. III. P OWER C ONTROL BASED ON S PECTRUM S ENSING S IDE I NFORMATION In this section, we present a power control approach in cognitive radio systems based on spectrum sensing side information to efficiently utilize the spectrum by allowing the cognitive radio to co-exist with the primary system. We firstly propose an idea of determining the distance d between the primary transmitter and the cognitive radio from spectrum sensing. Then, we show that the transmit power of the cognitive radio can be controlled based on the distance d in order to guarantee a quality of service to the primary receiver. A. Spectrum Sensing Side Information In order to avoid the harmful interference to the primary (licensed) system, the cognitive radio needs to sense the availability of the spectrum. The goal of spectrum sensing is to decide between the following two hypotheses: H0 : x(t) = n(t) H1 : x(t) = hs(t) + n(t)
0 λ I(Yi ) = (8) 0, otherwise
5163 Authorized licensed use limited to: National Taiwan University. Downloaded on July 19, 2009 at 10:28 from IEEE Xplore. Restrictions apply.
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the ICC 2007 proceedings.
for i = 1, · · · , N , where Yi denotes the energy collected by the cognitive radio in time slot i and N is the total number of time slots. Then, Pm can be estimated as
−1
10
(9)
Probability of missing
N 1 Pˆm = 1 − I(Yi ). N i=1
0
10
Once Pm is determined, d (or η) can be obtained from (7). B. Transmit Power Control for Cognitive Radio When the presence of the primary user is not properly detected during the spectrum sensing process, the overall system performance will be degraded significantly due to the interference from the cognitive radio. In this section, we propose a transmit power control method to address this problem by limiting the interference due to the presence of the cognitive radio while guaranteeing efficient spectrum utilization. To allow the primary receiver to successfully decode the received signals from the primary transmitter in the presence of the the cognitive radio, the signal-to-interference-plus-noise ratio (SINR) of the primary receiver should be guaranteed to be above a threshold of the decodability SNR γd (in dB), i.e., SINR≥ γd . Then, the quality of service for the primary receiver can be evaluated by γd Qp ≥ 10 10 , 2 Qc + σ
Qp
where and denote the received signal power from the primary transmitter and the cognitive radio, respectively. From (10) and those parameters shown in Fig. 1, we can arrive at [8] µ Qc ≤ ∆ + 10 log(10 10 − 1) 10 log 2 σ ψ µ 1 1 + 10α log (10 10 ) α − (10− 10 ) α g(ψ)
(11)
for the constants α and µ. It can be seen from (11) that the value of the allowable Qc depends on the SNR loss ψ. Since the location of the primary receiver is usually unknown for the cognitive radio, it is difficult to get the value of ψ. In this work, we consider the worst case that the primary receiver is located at the closet point to the cognitive radio. In this case, from Fig. 1, we have ψ = η − ∆, dB.
(12)
By substituting (12) into (11), we can see that Qc can be decided by η. Considering η = −10 log(d−α ), we can have Qmax c
−3
10
−4
10
= g (η − ∆) + 10 log(σ 2 ), dB
= g −10 log(d−α ) − ∆ + 10 log(σ 2 ), dB(13)
where Qmax denote the maximum value of Qc in dB and d c has been derived from the spectrum sensing side information in Section III-A. As a result, the transmit power of the cognitive
60
65
70
Fig. 4.
75
80 η (dB)
85
90
95
100
Pm versus η (in dB).
radio which is allowed to guarantee a good quality of service for the primary receiver can be derived by the following steps. Proposed Power Control Algorithm: Step 1: Calculate Pm from (9). Step 2: Derive d or η from (7). from (13). Step 3: Calculate Qmax c
(10)
Qc
−2
10
IV. N UMERICAL R ESULTS In this section, numerical results are presented to demonstrate the potential of the proposed transmit power control method in cognitive radio systems. Assume that the system parameters are as follows: • ∆ = 60 dB; • µ = 1 dB; 2 • Qp /σ = 100 dB; • Pf = 0.01; • α = 2. The channel environment is assumed to have flat Rayleigh fading and path loss. In order to allow the cognitive radio to share the spectrum with the primary system while guaranteeing a good quality of service to the primary receiver characterized by (10), the transmit power of the cognitive radio should be designed judiciously. In the following, we show how our approach works to get the maximum transmit power of the cognitive radio. Because it is difficult to locate the primary receiver for the cognitive radio, we consider the worst case scenario where the primary receiver is the nearest to the cognitive radio, as shown in Fig. 1. Firstly, from (7) we can obtain Pm vs. η as shown in Fig. 4. This shows the proportional relationship between Pm and the SNR loss due to the distance d. vs. η as shown in Fig. 5. It From (13) we can get Qmax c demonstrates that the allowable transmit power of the cognitive radio can be increased when a heavy SNR loss occurs between the cognitive radio and the primary receiver. This is reasonable
5164 Authorized licensed use limited to: National Taiwan University. Downloaded on July 19, 2009 at 10:28 from IEEE Xplore. Restrictions apply.
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the ICC 2007 proceedings.
90
100
90 Maximum secondary power (dBW)
Maximum secondary power (dBW)
80
70
60
50
40
30 60
80
70
60
50
40
65
70
Fig. 5.
75 η (dB)
80
85
30 −3 10
90
Qmax versus η (in dB). c
−2
10
Fig. 6.
because the interference power that the cognitive radio inflicts on the primary receiver is reduced by the large path loss. Finally, from Fig. 4 and Fig. 5, we can establish the and Pm as illustrated in Fig. 6. relationship between Qmax c By calculating Pm from (9), the maximum transmit power Qmax can be determined to guarantee the quality of service c for the licensed user in the presence of the cognitive radio. Because the maximum power of Qc is evaluated according to the worse case scenario where the primary receiver is the nearest to the cognitive radio, our power control approach can be applied to a primary receiver in any location. V. C ONCLUSION We have considered the case of a primary user and a cognitive radio sharing spectrum simultaneously. To limit the interference to the primary user, we have developed a power control approach which intelligently adjusts the transmit power of the cognitive radio while maintaining a quality of service for the primary user. The transmit power is controlled by the spectrum sensing side information, the probability of missing which actually includes the implicit location information of the primary user. Numerical results were presented to show that the proposed approach can guarantee a reliable quality of service for the primary user in any location while enhancing the spectrum utilization greatly.
−1
10 Probability of missing
0
10
Qmax versus Pm . c
[4] T. A. Weiss and F. K. Jondral, “Spectrum pooling: an innovative strategy for the enhancement of spectrum efficiency,” IEEE Commun. Mag., vol. 42, Mar. 2004, pp. S8–14. [5] S. Haykin, “Cognitive radio: brain-empowered wireless communications,” IEEE J. Sel. Areas in Commun., vol. 23, pp. 201–220, Feb. 2005. [6] T. Weiss, J. Hillenbrand, A. Krohn, and F. K. Jondral, “Mutual interference in OFDM-based spectrum pooling systems,” in Proc. IEEE VTC, May 2004, vol. 4, pp. 1873–1877. [7] V. D. Chakravarthy, A. K. Shaw, M. A. Temple, and J. P. Stephens, “Cognitive radio – an adaptive waveform with spectral sharing capability,” in Proc. IEEE WCNC, New Orleans, USA, Mar. 2005, vol. 2, pp. 724– 729. [8] N. Hoven and A. Sahai, “Power scaling for cognitive radio,” in Proc. WCNC, Maui, Hawaii, USA, June 2005, vol. 1, pp. 250–255. [9] B. Wild and K. Ramchandran, “Detecting primary receivers for cognitive radio applications,” in Proc. 1st IEEE Symp. New Frontiers in Dynamic Spectrum Access Networks (DySPAN’05), Baltimore, USA, Nov. 8–11, 2005, pp. 124–130. [10] A. Sahai, N. Hoven, and R. Tandra, “Some fundamental limits on cognitive radio,” in Proc. of Allerton Conf., Monticello, Oct. 2004. [11] A. Ghasemi and E. S. Sousa, “Collaborative spectrum sensing for opportunistic access in fading environments,” in Proc. 1st IEEE Symp. New Frontiers in Dynamic Spectrum Access Networks (DySPAN’05), Baltimore, USA, Nov. 8–11, 2005, pp. 131–136.
ACKNOWLEDGEMENT This work is supported in part by the Hong Kong Research Grant Council under Grant No. N HKUST622/06. R EFERENCES [1] Federal Communications Commission, “Spectrum Policy Task Force,” Rep. ET Docket no. 02-135, Nov. 2002. [2] M. A. McHenry, “NSF spectrum occupancy measurements project summary,” Shared Spectrum Company Report, August, 2005. [Online] Available: http://www.sharedspectrum.com. [3] J. Mitola and G. Q. Maguire, “Cognitive radio: Making software radios more personal,” IEEE Pers. Commun., vol. 6, pp. 13–18, Aug. 1999.
5165 Authorized licensed use limited to: National Taiwan University. Downloaded on July 19, 2009 at 10:28 from IEEE Xplore. Restrictions apply.
Power Control in Cognitive Radio Systems Based on Spectrum Sensing Side Information Karama Hamdi, Wei Zhang, and Khaled Ben Letaief, Fellow, IEEE Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology Kowloon, Hong Kong Email: {eekarama, eewzhang, eekhaled}@ust.hk Abstract— Cognitive radio has been recently proposed as a promising technology to improve the spectrum utilization efficiency by intelligently sensing and accessing some vacant bands of licensed users. In this paper, we consider the coexistence between a cognitive radio and a licensed user in order to enhance the spectrum efficiency. We develop an approach to allow the cognitive radio to operate in the presence of the licensed user. In order to minimize the interference to the licensed user, the transmit power of the cognitive radio is controlled by using the side information of spectrum sensing. Numerical results will show that the quality of service for the licensed user can be guaranteed in the presence of the cognitive radio by the proposed approach.
I. I NTRODUCTION The explosive growth in wireless services over the past several years illustrates the huge and growing demand of the business community, consumers and the government for wireless communications. With this growth of communication applications, the spectrum becomes more congested. Even though the Federal Communications Commission (FCC) has expanded some spectral bands, these frequency bands are exclusively assigned to specific users or service providers. Such expansion does not necessarily guarantee that the bands are being used most efficiently all the time. Recent survey has in fact proved that most of the radio frequency spectrum is vastly under-utilized [1], [2]. For example, cellular network bands are overloaded in most parts of the world but amateur radio or paging frequencies are not. Moreover, those rarely used frequency bands are assigned to specific services that cannot be accessed by unlicensed users, even if the transmission of the unlicensed users does not introduce any interference to the licensed service. To deal with the conflicts between spectrum congestion and spectrum under-utilization, cognitive radio has been recently proposed as a smart and agile technology which allows nonlegitimate users to utilize licensed bands opportunistically [3], [4]. By detecting particular spectrum holes and jumping into them rapidly, the cognitive radio can improve the spectrum utilization significantly. To guarantee a high spectrum efficiency while avoiding the interference to the licensed users, the cognitive radio should be able to adapt spectrum conditions flexibly. Hence, some important abilities should be provided by the cognitive radio which include spectrum sensing, dynamic frequency selection and transmit power control [5].
One of the most challenging problems of cognitive radio is the interference which occurs when a cognitive radio accesses a licensed band but fails to notice the presence of the licensed user. To address this problem, the cognitive radio should be designed to co-exist with the licensed user without creating harmful interference. Recently, several interference mitigation techniques have been presented for cognitive radio systems. An orthogonal frequency division multiplexing (OFDM) was considered as a candidate for cognitive radio to avoid the interference by leaving a set of subchannels unused [6]. Thus, it can provide a flexible spectral shape that fills the spectral gaps without interfering with the licensed users. A transform domain communication system (TDCS) was proposed to mitigate the interference by not putting the waveform energy at corrupted spectral locations [7]. A power control rule was presented to allow cognitive radios to adjust their transmit powers in order to guarantee a quality of service to the primary system [8]. To avoid the interference to the licensed users, the transmit power of the cognitive radio should be limited based on the locations of the licensed users [8]. However, it is difficult to locate the licensed users for the cognitive radio in practice because the channels between the cognitive radio and the licensed users are usually unknown. Furthermore, the environment where the system is in operation may have large delay spread and hence the channel model is complicated by fading, shadowing and path loss effects. In [9], the local oscillator (LO) leakage power was exploited to locate the primary receivers. But it is still not easy to apply this in practice because the approach requires a sensor node mounted close to the primary receivers to detect the LO leakage power. In this paper, we present a power control approach in cognitive radio systems based on spectrum sensing side information in order to mitigate the interference to the primary user due to the presence of cognitive radios. This approach consists of two steps. Firstly, the shortest distance between a licensed receiver and a cognitive radio is derived from the spectrum sensing side information. Then, the transmit power of the cognitive radio is determined based on this shortest distance to guarantee a quality of service for the licensed user. Because the worst case is considered in this approach where the cognitive radio is the closest to the licensed user, the proposed power control approach can be applied to the licensed user in any location. The rest of this paper is organized as follows. In Section
1-4244-0353-7/07/$25.00 ©2007 IEEE 5161 Authorized licensed use limited to: National Taiwan University. Downloaded on July 19, 2009 at 10:28 from IEEE Xplore. Restrictions apply.
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the ICC 2007 proceedings.
Rd
d
µ Rp PTx
∆
PRx
ψ
CR
Fig. 1. System model. PTx, PRx and CR denote primary transmitter, primary receiver and cognitive radio, respectively.
II, the system model is given. In Section III, a power control approach is proposed for cognitive radio systems based on spectrum sensing side information. Numerical results are presented and discussed in Section IV. Finally, conclusions are drawn in Section V. II. S YSTEM M ODEL We consider a scenario in which a primary system (licensed user), formed by a transmitter-receiver pair, co-exists in the same area with a secondary user (cognitive radio). The system model of our interest is illustrated in Fig. 1. In the primary system, the primary transmitter communicates with the primary receiver with the transmit power Qp . Some system parameters which are shown in Fig. 1 are explained as follows. The circle around the primary transmitter with the radius Rd (m) represents the decodable region within which the signal-to-noise ratio (SNR) of decodability occurs in the absence of interference to the primary receiver. The circle around the primary transmitter with the radius Rp (m) denotes the protection region within which the primary receiver must be guaranteed successful reception even in the presence of the cognitive radio. ∆ (dB) is the signal attenuation due to the distance Rd . µ is the margin of protection in dB which represents how much interference above the noise floor the primary system can tolerate [8]. In the secondary system, a cognitive radio is considered to work in the same frequency band as the primary system. Before accessing the channel, the cognitive radio acts as a listener to detect from the received signals whether the primary system is in operation. Let d (m) denote the distance between the primary transmitter and the cognitive radio. In practice, it is difficult to obtain the value of d because the signals from the primary transmitter and the channel are both unknown to the cognitive radio. Thus, computing d will be a challenging problem. Another challenging issue is to allow the cognitive radio to access the same spectrum band where the primary user is operating. In such a case, the cognitive radio may interfere with the primary system, thereby, degrading the quality of
service for the primary receiver. To reduce the interference, the transmit power Qc of the cognitive radio should be limited based on the tolerable interference to the primary receiver which directly depends on the distance between the cognitive radio and the primary receiver. However, it is difficult for the cognitive radio to locate the primary receiver which can be at any location inside the protection region. To address this problem, the worst case scenario is investigated in this work where the primary receiver is located on the crossing point between the boundary of the protection region and the line from the primary transmitter to the cognitive radio as shown in Fig. 1. By limiting the transmit power of the cognitive radio for the above worst case, we can guarantee a good quality of service for the primary receiver in operation at any location inside the protection region. In this case, it can be seen from Fig. 1 that the transmit power of the cognitive radio which is allowed to inflict tolerable interference on the primary receiver depends on the SNR loss (µ+ψ) in dB. Note the SNR loss due to the distance d is given by η ψ + ∆ (dB). Then, the transmit power control problem is essentially converted to the problem of evaluating the SNR loss η due to d for a given µ and ∆. We assume that the channel between any two terminals in Fig. 1 experiences flat Rayleigh fading and path loss. The propagation power attenuation is characterized by Q(r) = r−α , where r represents the distance and α denotes the power loss exponent which is usually a constant in the range 2 ∼ 6. Throughout this paper, α = 2 is used which corresponds to the free-space attenuation. III. P OWER C ONTROL BASED ON S PECTRUM S ENSING S IDE I NFORMATION In this section, we present a power control approach in cognitive radio systems based on spectrum sensing side information to efficiently utilize the spectrum by allowing the cognitive radio to co-exist with the primary system. We firstly propose an idea of determining the distance d between the primary transmitter and the cognitive radio from spectrum sensing. Then, we show that the transmit power of the cognitive radio can be controlled based on the distance d in order to guarantee a quality of service to the primary receiver. A. Spectrum Sensing Side Information In order to avoid the harmful interference to the primary (licensed) system, the cognitive radio needs to sense the availability of the spectrum. The goal of spectrum sensing is to decide between the following two hypotheses: H0 : x(t) = n(t) H1 : x(t) = hs(t) + n(t)
0 λ I(Yi ) = (8) 0, otherwise
5163 Authorized licensed use limited to: National Taiwan University. Downloaded on July 19, 2009 at 10:28 from IEEE Xplore. Restrictions apply.
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the ICC 2007 proceedings.
for i = 1, · · · , N , where Yi denotes the energy collected by the cognitive radio in time slot i and N is the total number of time slots. Then, Pm can be estimated as
−1
10
(9)
Probability of missing
N 1 Pˆm = 1 − I(Yi ). N i=1
0
10
Once Pm is determined, d (or η) can be obtained from (7). B. Transmit Power Control for Cognitive Radio When the presence of the primary user is not properly detected during the spectrum sensing process, the overall system performance will be degraded significantly due to the interference from the cognitive radio. In this section, we propose a transmit power control method to address this problem by limiting the interference due to the presence of the cognitive radio while guaranteeing efficient spectrum utilization. To allow the primary receiver to successfully decode the received signals from the primary transmitter in the presence of the the cognitive radio, the signal-to-interference-plus-noise ratio (SINR) of the primary receiver should be guaranteed to be above a threshold of the decodability SNR γd (in dB), i.e., SINR≥ γd . Then, the quality of service for the primary receiver can be evaluated by γd Qp ≥ 10 10 , 2 Qc + σ
Qp
where and denote the received signal power from the primary transmitter and the cognitive radio, respectively. From (10) and those parameters shown in Fig. 1, we can arrive at [8] µ Qc ≤ ∆ + 10 log(10 10 − 1) 10 log 2 σ ψ µ 1 1 + 10α log (10 10 ) α − (10− 10 ) α g(ψ)
(11)
for the constants α and µ. It can be seen from (11) that the value of the allowable Qc depends on the SNR loss ψ. Since the location of the primary receiver is usually unknown for the cognitive radio, it is difficult to get the value of ψ. In this work, we consider the worst case that the primary receiver is located at the closet point to the cognitive radio. In this case, from Fig. 1, we have ψ = η − ∆, dB.
(12)
By substituting (12) into (11), we can see that Qc can be decided by η. Considering η = −10 log(d−α ), we can have Qmax c
−3
10
−4
10
= g (η − ∆) + 10 log(σ 2 ), dB
= g −10 log(d−α ) − ∆ + 10 log(σ 2 ), dB(13)
where Qmax denote the maximum value of Qc in dB and d c has been derived from the spectrum sensing side information in Section III-A. As a result, the transmit power of the cognitive
60
65
70
Fig. 4.
75
80 η (dB)
85
90
95
100
Pm versus η (in dB).
radio which is allowed to guarantee a good quality of service for the primary receiver can be derived by the following steps. Proposed Power Control Algorithm: Step 1: Calculate Pm from (9). Step 2: Derive d or η from (7). from (13). Step 3: Calculate Qmax c
(10)
Qc
−2
10
IV. N UMERICAL R ESULTS In this section, numerical results are presented to demonstrate the potential of the proposed transmit power control method in cognitive radio systems. Assume that the system parameters are as follows: • ∆ = 60 dB; • µ = 1 dB; 2 • Qp /σ = 100 dB; • Pf = 0.01; • α = 2. The channel environment is assumed to have flat Rayleigh fading and path loss. In order to allow the cognitive radio to share the spectrum with the primary system while guaranteeing a good quality of service to the primary receiver characterized by (10), the transmit power of the cognitive radio should be designed judiciously. In the following, we show how our approach works to get the maximum transmit power of the cognitive radio. Because it is difficult to locate the primary receiver for the cognitive radio, we consider the worst case scenario where the primary receiver is the nearest to the cognitive radio, as shown in Fig. 1. Firstly, from (7) we can obtain Pm vs. η as shown in Fig. 4. This shows the proportional relationship between Pm and the SNR loss due to the distance d. vs. η as shown in Fig. 5. It From (13) we can get Qmax c demonstrates that the allowable transmit power of the cognitive radio can be increased when a heavy SNR loss occurs between the cognitive radio and the primary receiver. This is reasonable
5164 Authorized licensed use limited to: National Taiwan University. Downloaded on July 19, 2009 at 10:28 from IEEE Xplore. Restrictions apply.
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the ICC 2007 proceedings.
90
100
90 Maximum secondary power (dBW)
Maximum secondary power (dBW)
80
70
60
50
40
30 60
80
70
60
50
40
65
70
Fig. 5.
75 η (dB)
80
85
30 −3 10
90
Qmax versus η (in dB). c
−2
10
Fig. 6.
because the interference power that the cognitive radio inflicts on the primary receiver is reduced by the large path loss. Finally, from Fig. 4 and Fig. 5, we can establish the and Pm as illustrated in Fig. 6. relationship between Qmax c By calculating Pm from (9), the maximum transmit power Qmax can be determined to guarantee the quality of service c for the licensed user in the presence of the cognitive radio. Because the maximum power of Qc is evaluated according to the worse case scenario where the primary receiver is the nearest to the cognitive radio, our power control approach can be applied to a primary receiver in any location. V. C ONCLUSION We have considered the case of a primary user and a cognitive radio sharing spectrum simultaneously. To limit the interference to the primary user, we have developed a power control approach which intelligently adjusts the transmit power of the cognitive radio while maintaining a quality of service for the primary user. The transmit power is controlled by the spectrum sensing side information, the probability of missing which actually includes the implicit location information of the primary user. Numerical results were presented to show that the proposed approach can guarantee a reliable quality of service for the primary user in any location while enhancing the spectrum utilization greatly.
−1
10 Probability of missing
0
10
Qmax versus Pm . c
[4] T. A. Weiss and F. K. Jondral, “Spectrum pooling: an innovative strategy for the enhancement of spectrum efficiency,” IEEE Commun. Mag., vol. 42, Mar. 2004, pp. S8–14. [5] S. Haykin, “Cognitive radio: brain-empowered wireless communications,” IEEE J. Sel. Areas in Commun., vol. 23, pp. 201–220, Feb. 2005. [6] T. Weiss, J. Hillenbrand, A. Krohn, and F. K. Jondral, “Mutual interference in OFDM-based spectrum pooling systems,” in Proc. IEEE VTC, May 2004, vol. 4, pp. 1873–1877. [7] V. D. Chakravarthy, A. K. Shaw, M. A. Temple, and J. P. Stephens, “Cognitive radio – an adaptive waveform with spectral sharing capability,” in Proc. IEEE WCNC, New Orleans, USA, Mar. 2005, vol. 2, pp. 724– 729. [8] N. Hoven and A. Sahai, “Power scaling for cognitive radio,” in Proc. WCNC, Maui, Hawaii, USA, June 2005, vol. 1, pp. 250–255. [9] B. Wild and K. Ramchandran, “Detecting primary receivers for cognitive radio applications,” in Proc. 1st IEEE Symp. New Frontiers in Dynamic Spectrum Access Networks (DySPAN’05), Baltimore, USA, Nov. 8–11, 2005, pp. 124–130. [10] A. Sahai, N. Hoven, and R. Tandra, “Some fundamental limits on cognitive radio,” in Proc. of Allerton Conf., Monticello, Oct. 2004. [11] A. Ghasemi and E. S. Sousa, “Collaborative spectrum sensing for opportunistic access in fading environments,” in Proc. 1st IEEE Symp. New Frontiers in Dynamic Spectrum Access Networks (DySPAN’05), Baltimore, USA, Nov. 8–11, 2005, pp. 131–136.
ACKNOWLEDGEMENT This work is supported in part by the Hong Kong Research Grant Council under Grant No. N HKUST622/06. R EFERENCES [1] Federal Communications Commission, “Spectrum Policy Task Force,” Rep. ET Docket no. 02-135, Nov. 2002. [2] M. A. McHenry, “NSF spectrum occupancy measurements project summary,” Shared Spectrum Company Report, August, 2005. [Online] Available: http://www.sharedspectrum.com. [3] J. Mitola and G. Q. Maguire, “Cognitive radio: Making software radios more personal,” IEEE Pers. Commun., vol. 6, pp. 13–18, Aug. 1999.
5165 Authorized licensed use limited to: National Taiwan University. Downloaded on July 19, 2009 at 10:28 from IEEE Xplore. Restrictions apply.
