Source-linked AI summary
Relay-Selection Improves the Security-Reliability Trade-off in Cognitive Radio Systems
Yulong Zou, Benoit Champagne, Wei-Ping Zhu, Lajos Hanzo
TL;DR
The paper addresses secure and reliable secondary transmission in cognitive radio networks exposed to eavesdropping. It analyzes single- and multi-relay selection under realistic spectrum sensing and compares them with direct and artificial-noise transmission. Relay selection generally improves the security-reliability trade-off, with multi-relay selection performing best and gains increasing with the number of relays.
Problem
Eavesdropping threatens confidentiality in cognitive radio transmissions, requiring analysis of the security-reliability trade-off under realistic spectrum sensing.
Method
The paper proposes single-relay selection using one selected SR and multi-relay selection using multiple simultaneous SRs, then analyzes intercept and outage probabilities.
Results
Relay-selection schemes generally outperform conventional direct transmission and artificial-noise approaches in SRT, while multi-relay selection outperforms single-relay selection.
Takeaways & Limitations
Increasing the number of SRs significantly improves the SRT of both relay-selection approaches, with greater multi-relay gains but higher synchronization complexity.
Abstract
from arXiv · showhide
We consider a cognitive radio (CR) network consisting of a secondary transmitter (ST), a secondary destination (SD) and multiple secondary relays (SRs) in the presence of an eavesdropper. We rely on careful relay selection for protecting the ST-SD transmission against the eavesdropper with the aid of both single-relay and multi-relay selection. To be specific, only the "best" SR is chosen in the single-relay selection for assisting the secondary transmission, whereas the multi-relay selection invokes multiple SRs for simultaneously forwarding the ST's transmission to the SD. We analyze both the intercept probability and outage probability of the proposed single-relay and multi-relay selection schemes for the secondary transmission relying on realistic spectrum sensing. We also evaluate the performance of classic direct transmission and artificial noise based methods for the purpose of comparison with the proposed relay selection schemes. It is shown that as the intercept probability requirement is relaxed, the outage performance of the direct transmission, the artificial noise based and the relay selection schemes improves, and vice versa. This implies a trade-off between the security and reliability of the secondary transmission in the presence of eavesdropping attacks, which is referred to as the security-reliability trade-off (SRT). Furthermore, we demonstrate that the SRTs of the single-relay and multi-relay selection schemes are generally better than that of classic direct transmission. Moreover, as the number of SRs increases, the SRTs of the proposed single-relay and multi-relay selection approaches significantly improve. Finally, our numerical results show that as expected, the multi-relay selection scheme achieves a better SRT performance than the single-relay selection.
I. INTRODUCTION
Cognitive radio networks face eavesdropping threats, motivating physical-layer security analysis of relay-assisted secondary transmissions under realistic spectrum sensing. The paper proposes single- and multi-relay selection schemes and compares their security-reliability trade-offs with direct and artificial-noise transmission.
- I. INTRODUCTION: Cognitive radio networks are vulnerable to diverse attacks, including eavesdropping that threatens the confidentiality of secondary transmissions.Security threats can arise during spectrum sensing, spectrum sharing, spectrum mobility, and spectrum management.
- I. INTRODUCTION: Physical-layer security protects authorized users against eavesdropping by exploiting wireless-channel characteristics, while fading can degrade achievable secrecy.Prior work investigated MIMO, cooperative relaying, and beamforming to combat fading effects.
- I. INTRODUCTION: The paper studies the security-reliability trade-off of a cognitive relay network with an ST, SD, multiple SRs, and an unauthorized attacker under realistic spectrum sensing.Security and reliability are characterized through intercept probability and outage probability, respectively.
- I. INTRODUCTION: Single-relay selection forwards the ST transmission through one selected SR, whereas multi-relay selection uses multiple SRs simultaneously.The proposed schemes protect secondary transmissions against eavesdropping attacks.
- I. INTRODUCTION: The paper derives closed-form intercept- and outage-probability expressions for both relay-selection schemes over Rayleigh fading and numerically compares them with direct and artificial-noise methods.The comparison is conducted in the presence of realistic spectrum sensing.
- I. INTRODUCTION: Improved spectrum-sensing reliability or reduced false-alarm probability improves the SRTs of both relay-selection schemes, which generally outperform direct and artificial-noise transmission.The system model includes opportunistic access to licensed spectrum based on spectrum sensing, with imperfect detection producing primary-secondary interference.
C. Single-Relay Selection
Single-relay selection uses decode-and-forward relaying and chooses one successfully decoding relay based on its instantaneous SR-SD capacity. This selection enhances the legitimate channel without requiring eavesdropper CSI, while its security benefit arises primarily through improved outage performance.
- System and decoding: The scheme uses N secondary relays with decode-and-forward relaying over two adjacent time slots.If no relay decodes successfully, all relays remain silent; otherwise, one relay forwards the decoded signal.
- Relay selection: The best relay is the successfully decoding SR with the highest instantaneous capacity to SD.Selection is performed within the non-empty decoding set.
- Relay selection: Only SR-SD CSI |hid|2 is required for selection, without eavesdropper CSI knowledge.The selected relay forwards its decoded signal to SD and E at power Ps.
- Performance comparison: SRS improves the legitimate channel because its capacity depends on the maximum of the SR-SD channel gains rather than the single direct-link gain.With i.i.d. gains and sufficiently many relays, the maximum is most likely much higher than |hsd|2.
- Performance comparison: If relay and direct-link gains are non-identically distributed, direct transmission can outperform SRS when the direct-link mean is much higher.The paper notes that practical relay placement typically supports SRS improvement because relays lie between ST and SD.
- Security-reliability trade-off: For i.i.d. eavesdropper-side gains, selection does not affect the wiretap-channel distribution, so its main benefit is reduced outage probability and the resulting SRT improvement.The main and wiretap channels are independent, preventing selection based on SD CSI from substantially changing eavesdropper capacity.
D. Multi-Relay Selection
Multi-relay selection simultaneously forwards the decoded signal through all successfully decoding relays using normalized weights. It improves legitimate capacity and can outperform single-relay selection in SRT performance, but requires complex synchronization.
- Scheme definition: MRS uses every relay that successfully decodes xs to simultaneously transmit it to SD.If the decoding set is empty, no relay forwards the signal.
- Scheme definition: The relay weights are normalized so the total transmit power across the selected relays remains Ps.The weight vector has unit norm and dimension equal to the number of successfully decoding relays.
- Capacity comparison: The MRS legitimate-channel gain is based on the sum of selected-relay gains, yielding higher legitimate transmission capacity than SRS.SRS instead relies on the maximum SR-SD gain.
- Security-reliability trade-off: MRS and SRS have more or less the same wiretap-channel capacity because relay weights are optimized from the main-channel CSI while the two channels are independent.The weight design requires SR-SD CSI Hd but not eavesdropper CSI He.
- Security-reliability trade-off: At a fixed outage requirement, MRS achieves better intercept performance by using its capacity gain to support a higher data rate and reduce IP.The paper links higher rate to lower intercept probability through the IP definition.
- Implementation trade-off: MRS requires high-complexity symbol-level synchronization, and synchronization imperfections may make its performance worse than SRS.SRS does not require the same complex synchronization process.
- Channel model: The channel model throughout the paper is Rayleigh fading, making the squared fading magnitudes exponentially distributed random variables.This assumption applies to channels such as ST-SD, ST-SR, and SR-SD.
III. SRT ANALYSIS OVER RAYLEIGH FADING CHANNELS
The SRT analysis evaluates direct transmission, SRS, and MRS over Rayleigh fading channels. Security is measured by intercept probability, while reliability is measured by outage probability under detected spectrum holes.
- Scope: The analysis covers direct transmission, SRS, and MRS over Rayleigh fading channels.The fading model is used for the proposed and benchmark schemes.
- Metrics: Intercept probability quantifies security and outage probability quantifies reliability.Both metrics are evaluated for the eavesdropper and destination, respectively.
- Spectrum sensing: Because the ST transmits only after detecting a spectrum hole, OP and IP are calculated conditionally on the sensing outcome that the spectrum is unoccupied.The analysis incorporates realistic spectrum sensing through this conditioning.
A. Direct Transmission
Direct transmission serves as the benchmark for calculating outage and intercept probabilities after a spectrum hole is detected. Its outage accounts for both truly vacant and occupied-spectrum cases under sensing uncertainty.
- Outage probability: Direct-transmission outage is evaluated after the spectrum is deemed unoccupied by the sensing decision.The outage event is expressed through the SD channel capacity falling below the data rate.
- Outage probability: When the spectrum is occupied, primary-user interference appears in the SD capacity expression.When it is unoccupied, the capacity depends on the ST-SD fading gain and transmit SNR.
- Sensing notation: The analysis defines π0 and π1 as the conditional probabilities of vacant and occupied spectrum, and Δ as 2^R−1.Pd and Pf denote the detection probability and false-alarm probability, respectively.
- Outage probability: The outage probability combines the cases where the licensed spectrum is actually unoccupied and occupied despite the sensing decision.The two cases are weighted by the corresponding conditional probabilities.
- Intercept probability: An intercept event occurs when the ST-E channel capacity exceeds the data rate R.The resulting IP is conditioned on the sensing decision that the spectrum is unoccupied.
- Intercept probability: The direct-transmission IP expressions use independent exponential distributions for the ST-E and primary-user-to-eavesdropper channel gains.Their means determine the corresponding probability terms.
B. Single-Relay Selection
The single-relay selection analysis derives outage and intercept-probability expressions for transmission through the best selected relay under spectrum-sensing conditions. The formulation accounts for cases where no relay is selected and uses closed-form channel probabilities for the relevant fading events.
- Outage probability: The outage probability is formulated using the capacity from the “best” secondary relay to the destination.When no relay is selected, the forwarding capacity is set to zero.
- Outage probability: The outage analysis rewrites the selection expression using the threshold Λ = 2^(2R−1)/γ_s and independent exponential channel gains.The channel-gain assumptions enable closed-form probability calculations.
- Intercept probability: The intercept probability is formulated from the capacity between the best selected relay and the eavesdropper.If no relay is selected, this relay-to-eavesdropper capacity is zero.
- Intercept probability: Closed-form probabilities for selected-relay and eavesdropper channel events are used to express the single-relay intercept probability.The derivation includes conditional events involving spectrum-sensing-related interference terms.
C. Multi-Relay Selection
The multi-relay selection analysis derives outage and intercept-probability results under Rayleigh fading and assumes identically distributed relay-to-destination channel gains. The outage expression is obtained in closed form, whereas the general intercept-probability expression requires numerical evaluation.
- Outage probability: The multi-relay outage analysis considers simultaneous forwarding by multiple selected relays over Rayleigh fading channels.The derivation uses relay-channel selection events and conditional channel probabilities.
- Channel assumptions: The relay-to-destination channel gains are assumed to be independent and identically distributed with a common mean.This assumption is stated to be statistically valid when relays are uniformly distributed over a geographical area.
- Outage probability: A closed-form outage-probability expression is obtained for the proposed multi-relay selection scheme using incomplete Gamma functions.The incomplete Gamma function appears when substituting the derived probability expressions into the outage formula.
- Intercept probability: The multi-relay intercept probability is analyzed through channel-event probabilities analogous to those used for the single-relay scheme.Several component probabilities are available in closed form, but one conditional probability is difficult to solve generally.
- Intercept probability: A general closed-form intercept-probability expression for multi-relay selection is challenging, so numerical results are obtained using computer simulations.This is the stated analytical limitation of the multi-relay intercept-probability analysis.
IV. NUMERICAL RESULTS AND DISCUSSIONS
The numerical results establish a security-reliability trade-off across transmission schemes: accepting higher outage probability improves intercept probability. Relay selection generally outperforms direct and artificial-noise approaches, with multi-relay selection benefiting further from more relays but requiring synchronization.
- Security-reliability trade-off: Higher tolerated outage probability improves intercept probability for direct transmission, artificial-noise, SRS, and MRS schemes.This demonstrates the security-reliability trade-off for cognitive-radio transmissions.
- Scheme comparisons: SRS and MRS generally outperform direct transmission and artificial-noise methods in security-reliability trade-off performance.MRS also achieves better SRT performance than SRS.
- Scheme comparisons: MRS outperforms SRS for both P0 = 0.2 and P0 = 0.8.As P0 increases from 0.2 to 0.8, both schemes improve their SRT because outage reduction is more significant than the accompanying intercept-probability increase.
- Spectrum sensing: Improving spectrum sensing reliability from (Pd, Pf) = (0.9, 0.1) to (Pd, Pf) = (0.99, 0.01) improves the SRTs of both SRS and MRS.The reported explanation is more accurate detection of unoccupied spectrum and reduced mutual interference.
- Relay population: Increasing the number of SRs from N = 2 to N = 8 significantly improves the SRTs of SRS and MRS, with MRS gaining a more notable advantage over SRS.The MRS improvement comes at the cost of high-complexity symbol-level synchronization among multiple SRs.
V. CONCLUSIONS
The paper evaluates relay-assisted cognitive-radio transmissions using intercept probability and outage probability to characterize the security-reliability trade-off. Both relay-selection schemes improve with more reliable spectrum sensing and more relays, with multi-relay selection generally outperforming the alternatives.
- The SRS and MRS schemes characterize security and reliability through intercept probability and outage probability under realistic spectrum sensing.
- As spectrum sensing reliability increases, the SRTs of both SRS and MRS improve.
- SRS and MRS generally outperform conventional direct transmission and artificial-noise-based approaches in SRT performance.
- MRS achieves better SRT performance than SRS.
- As the number of SRs increases, both SRS and MRS SRT performance improves significantly.
APPENDIX A DERIVATION OF (45) AND (46)
Appendix A derives the relevant probabilities by conditioning on relay-selection events and applying exponential-fading distributions. The derivation expands subset-based expressions, performs integrations, and completes the proofs of equations (45) and (46).
- The derivation rewrites the left sides of equations (45) and (46) using xi and y for channel-power variables.
- It applies independent exponential distributions for the channel-power variables to evaluate the resulting probabilities.
- Set-cardinality and non-empty-subset expansions are used to further expand the probability expressions.
- The appendix obtains the target results by substitution and integration, completing the proofs of equations (45) and (46).
- The best-relay case is handled with the law of total probability over the possible selected relays.
APPENDIX C PROOF OF (53) AND (54)
Appendix C derives equations (53) and (54) by rewriting probability terms in terms of random variables X and Y, deriving their distributions, and integrating under independence.
- The probability terms are rewritten as events involving X and Y, including Pr(X < Λ) and Pr(X < γpΛY + Λ).
- The derivation uses i.i.d. SR-SD fading coefficients to obtain the probability density function of X.
- Y is defined from the primary-destination channel power and is assigned an exponential probability density function.
- Independence of X and Y enables evaluation of Pr(X < γpΛY + Λ) through substitution into the integral expression.
p , |Dn|)]
The appendix concludes the derivation after applying the indicated substitution, completing the proofs of equations (53) and (54).
- The substitution completes the proofs of equations (53) and (54).