Spectrum sensing analysis method in STAR-RIS auxiliary cognitive radio system

Abstract

The invention discloses a spectrum sensing analysis method in an STAR-RIS (Short Time Assisted Recognition-Rare Information System) assisted cognitive radio system, which is characterized in that the spectrum sensing analysis method in the STAR-RIS assisted cognitive radio system is provided with a spectrum sensing module, a spectrum sensing module and a spectrum sensing module, and the spectrum sensing module and the spectrum sensing module are used for sensing and analyzing the spectrum sensing in the STAR-RIS assisted cognitive radio system. A multi-antenna base station is considered, a system model is established, statistical channel information is used for optimizing the phase of STAR-RIS, and a theoretical expression of the correct detection probability of the cognitive radio system is deduced; through simulation verification, the performance analysis method provided by the invention is effective for evaluating the system performance. According to the method, the spectrum sensing of the STAR-RIS assisted cognitive radio network is researched, the performance characteristics of a multi-antenna base station are considered, the phase of the STAR-RIS is optimized, statistical analysis of the correct detection probability is obtained, and effective guidance is provided for performance evaluation of systems of the same type.

Description

Technical field:

[0001] The present invention belongs to the field of wireless mobile communications and relates to a performance analysis method of a mobile communication system, in particular to a spectrum perception analysis method in a STAR-RIS assisted cognitive radio system. Background technology:

[0002] With the exponential growth of wireless communication traffic, the upcoming sixth-generation mobile communications is expected to deliver data rates exceeding 1 Tbps, serving more users and meeting diverse business needs. Cognitive radio is an effective technology for improving the spectral efficiency of communication systems through spectrum sharing. It allows unauthorized secondary users to access licensed frequency bands when primary users are idle, thus overcoming the problem of scarce spectrum resources. Cognitive radio is widely used in fields such as intelligent transportation, the Internet of Things, and emergency communications, effectively improving the reliability and efficiency of communications.

[0003] Given that RIS can deliver incoming wireless signals to the desired location by facilitating beneficial control of the signal path, research has explored RIS-assisted cognitive radio systems. However, to utilize RIS-assisted cognitive systems, the primary and secondary networks must be located on the same side of the RIS. This geographical restriction may not always be met in practical communication systems, especially for mobile users in wireless communication systems. STAR-RIS, a complementary technology, can be used to address this issue. Its basic principle is to utilize a set of programmable reflection and transmission units that can dynamically adjust the incident signal accordingly, thereby flexibly adjusting the signal propagation path, enhancing signal coverage and quality, and providing additional degrees of freedom for beamforming design.

[0004] While there has been extensive research on spectrum sensing in cognitive systems using RIS, there has been insufficient research on the sensing performance of STAR-RIS, and there is also a lack of performance analysis of multi-antenna base stations. Therefore, this paper investigates the performance of spectrum sensing in cognitive radio networks assisted by STAR-RIS. Summary of the invention:

[0005] The present invention aims to solve the problems existing in the above-mentioned existing technologies and provides a spectrum sensing analysis method in a STAR-RIS assisted cognitive radio system, which can more accurately analyze the detection correctness probability of the STAR-RIS assisted cognitive radio system.

[0006] The technical solution adopted by the present invention is: a spectrum sensing analysis method in a STAR-RIS assisted cognitive radio system, comprising the following steps:

[0007] Step S1: Establish a system model for spectrum sensing in a STAR-RIS-assisted cognitive radio system. Consider a cognitive network consisting of a base station PT equipped with N antennas, a single-antenna primary user P, and two single-antenna secondary users T and R. The network uses a STAR-RIS-assisted communication consisting of M elements. PT sends a primary user signal to the primary user P. The secondary users perform spectrum sensing with the assistance of STAR-RIS to detect whether the primary user is communicating.

[0008] Step S2: Channel from base station PT to STAR-RIS STAR-RIS channels to secondary users T and R k∈{t,r} is modeled as a Rice channel, with the LoS component Written where θ PT,d ,θ R,a , are the departure angle from PT to STAR-RIS, the arrival angle from PT to STAR-RIS, the departure angle from STAR-RIS to the secondary user, and v N (·),u M (·),q k,M (·) represents the corresponding array response vector, and the signal received by the secondary user is expressed as

[0009]

[0010] in is the beamforming vector, n0 is the noise at the receiving end, which has a mean of 0 and a variance of Gaussian distribution, p s is the transmission signal power of the base station PT, s0 is the transmission signal sent to the primary user, is the transmission and reflection factor matrix of STAR-RIS, and the perceived signal-to-noise ratio at the secondary user is in

[0011] Step S3: Optimize the phase of STAR-RIS based on the statistical channel information, and use the triangle inequality to obtain the optimal phase:

[0012] Step S4: Based on the probability distribution function and the optimal phase, the Gauss-Laguerre formula is used to obtain an approximate expression for the probability of correct detection:

[0013]

[0014] where t 1,i is the Laguerre polynomial L n The i-th root of (x), weight N prepresents the polynomial order, τ th is the detection threshold, L k , is the noncentral chi-square distribution parameter, N s is the sampling number, I α (·) is a first-order Bessel function.

[0015] Furthermore, the spectrum perception analysis method in the STAR-RIS assisted cognitive radio system of claim 1 is characterized in that: the design method of the optimal phase in step S3 is as follows: the perceived signal to interference and noise ratio is expressed as

[0016]

[0017] where ξ h ,η h ,ξ g,k ,η g,k is the parameter of the Rice channel, since only is phase-dependent, so maximizing the signal-to-noise ratio through STAR-RIS phase design is equivalent to

[0018]

[0019] Using the triangle inequality, the conditions for it to hold are

[0020]

[0021] This gives the expression for the optimal phase.

[0022] Furthermore, for the probability density function derivation in step S4, let According to the central limit theorem, Z k Approximately follows a Gaussian distribution, It obeys the non-central chi-square distribution with 2 degrees of freedom, and the probability density function is

[0023]

[0024] Therefore, the probability of correct detection is expressed as

[0025]

[0026] By substituting the variables in the above formula and using Gauss-Laguerre numerical integration for approximation, we can obtain an approximate formula for the probability of correct detection.

[0027] The present invention has the following beneficial effects: the present invention studies spectrum sensing of STAR-RIS-assisted cognitive radio networks, considers the performance characteristics of multi-antenna base stations, optimizes the phase of STAR-RIS, obtains statistical analysis of the probability of correct detection, and provides effective guidance for performance evaluation of similar systems. Description of the drawings:

[0028] Figure 1 This is a diagram of the analysis steps for spectrum sensing of the STAR-RIS assisted cognitive radio system of the present invention.

[0029] Figure 2 2 is a model diagram of the system in an embodiment of the present invention.

[0030] Figure 3 This is a comparison chart between the simulation and theory of the correct detection probability of users under different numbers of STAR-RIS elements implemented by the present invention.

[0031] Figure 4 This is a comparison chart between the simulation and theory of the correct detection probability of users under different sampling times implemented by the present invention. Specific implementation method:

[0032] The present invention will be further described below with reference to the accompanying drawings.

[0033] 1. System Model

[0034] A system model of spectrum sensing in STAR-RIS-assisted cognitive radio networks is established. Consider a cognitive network consisting of a base station PT equipped with N antennas, a primary user P with a single antenna, and two secondary users T and R with single antennas. The network uses a STAR-RIS-assisted communication consisting of M elements. STAR-RIS adopts the ES protocol. Assuming that all elements have the same amplitude coefficient, the transmission and reflection factor matrices are: in is the phase shift of the mth reflection or transmission element, is the amplitude factor.

[0035] The channel from the base station PT to STAR-RIS and the channel from STAR-RIS to secondary users T and R are both modeled as Rice channels, expressed as

[0036]

[0037] In order to simplify the expressions derived later, let where d PR ,d k are the distances from PT to STAR-RIS, STAR-RIS to secondary users T and R, respectively; α is the exponent of path loss, and κ h ,κ g,k are the Ricean factors of the PT to STAR-RIS and STAR-RIS to secondary user channels, respectively. Represents the Non-Line of Sight (NLoS) component, expressed as

[0038]

[0039] where θ PT,d ,θ R,a , They represent the departure angle from PT to STAR-RIS, the arrival angle from PT to STAR-RIS, the departure angle from STAR-RIS to the secondary user, and v N (·),u M (·),q k,M (·) represents the corresponding array response vector. The signal received by the secondary user can be expressed as

[0040]

[0041] γ k represents the perceived signal-to-noise ratio at the secondary user, which can be written as

[0042]

[0043] in,

[0044] 2. STAR-RIS Phase Design

[0045] The average value of the perceptual signal-to-noise ratio can be reformulated as:

[0046]

[0047] where f k,n is f k Since only the first term is phase-dependent, maximizing the signal-to-noise ratio through STAR-RIS phase design is equivalent to

[0048]

[0049] The above optimization problem can be equivalent to

[0050]

[0051] Using the triangle inequality, we can get

[0052]

[0053] The conditions for the above formula to be valid are

[0054]

[0055] Therefore the optimal phase can be set to

[0056]

[0057] 3. Statistical Analysis of Correct Detection Probability

[0058] The false alarm probability is defined as the secondary user misjudging the primary user's transmission signal, expressed as P f =Pr(T k >τ th |H0), where τ th is the detection threshold. In the case of H0, the test power is expressed as where N s is the number of sampling times. Since n0 is a complex Gaussian random variable, T k Subject to the degree of freedom of 2N s The chi-square distribution of

[0059]

[0060] The detection threshold can be set by giving a lower false alarm probability The method is to obtain

[0061]

[0062] The detection accuracy probability is defined as the probability that the secondary user can correctly perceive the primary user's transmission, denoted by P d =Pr(T k >τ th |H1). At this time, under the condition of H1, the test power is expressed as Due to s 0,i and n 0,i All obey complex Gaussian distribution, T k Similarly, the degree of freedom is 2N s The chi-square distribution approximation of the detection accuracy can be re-expressed as

[0063]

[0064] Because h n,m ,g k,m are independent and identically distributed random variables, so When M is large, it can be approximated by Gaussian distribution, using the STAR-RIS optimal phase, First, derive the expression of its mean:

[0065]

[0066] where μ k,i ,μ k,r μ k The real and imaginary parts of . Using the statistically optimized phase, Z k can be rewritten as

[0067]

[0068] Since T1 is a constant after beam design, Var(T1)=0, For independent and identically distributed random variables with a mean of 0, we can get Var(Z k )=p s ×(Var(T2)+Var(T3)+Var(T4)).

[0069] First derive the expression of Var(Re{T2})

[0070]

[0071] because is a complex Gaussian random variable with unit variance, so Therefore, the above formula can be re-expressed as

[0072]

[0073] The rest of the derivation is done in the same way, and we can get

[0074]

[0075] So Z k The expression for the variance of the real and imaginary parts is

[0076]

[0077] Because Z k Approximately obeys the complex Gaussian distribution, then It obeys the non-central chi-square distribution with degrees of freedom ρ = 2, and its probability density function is

[0078]

[0079] Among them L k , is the noncentral chi-square distribution parameter, I α (·) is a first-order Bessel function.

[0080] Therefore, the probability of correct detection is expressed as

[0081]

[0082] Substitute the above formula and use Gauss-Laguerre numerical integration to approximate it. The approximate expression is:

[0083]

[0084] where t 1,i is the Laguerre polynomial L nThe i-th root of (x), weight N p Represents the polynomial order.

[0085] The spectrum sensing performance of the STAR-RIS assisted cognitive radio system is evaluated by computer simulation. The distance between the base station PT and STAR-RIS is 50m, and the distances between STAR-RIS and user T and user R are 30m and 15m respectively. The Ricean factor of different channels has the same value, that is, κ = κ h =κ g,t =κ g,r , the departure angle and arrival angle in the LoS component of the Ricean channel are randomly selected in [0,2π). The detection threshold τ th It is determined by a given false alarm probability threshold. The specific simulation parameters are shown in the table below:

[0086]

[0087] Figure 3 Figure 2 shows the simulated and theoretical curves of the correct detection probability as a function of transmission power for different numbers of STAR-RIS elements. As can be seen from the figure, the theoretical and simulated values ​​are highly consistent, validating the correctness of the theoretical formula. As transmission power increases, the correct detection probability gradually approaches 1. Furthermore, at the same transmission power, a larger number of elements increases the correct detection probability. This is because increasing the number of elements increases the number of communication channels, resulting in greater channel gain and, consequently, improved spectrum perception for users.

[0088] Figure 4 The figure shows how the correct detection probability changes with the number of sampling times, for transmit antennas of 2 and 4, respectively. As can be seen, as the number of sampling times increases, the correct detection probability increases for the same number of transmit antennas. This is because the detection threshold is determined based on a given false alarm probability. When the ambient noise remains essentially unchanged, the false alarm probability is primarily determined by the number of sampling times. Increasing the number of sampling times improves detection accuracy, resulting in more precise perception results. Furthermore, increasing the transmit power also helps improve the correct detection probability.

[0089] In summary, the performance analysis method proposed in the present invention can effectively analyze the detection correctness probability of STAR-RIS assisted cognitive radio networks, and the simulation results fully demonstrate its effectiveness.

[0090] The above description is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements can be made without departing from the principles of the present invention. These improvements should also be regarded as the scope of protection of the present invention.

Claims

1. A spectrum sensing analysis method in a STAR-RIS assisted cognitive radio system, characterized by: Here are the steps: Step S1: Establish a system model for spectrum sensing in a STAR-RIS-assisted cognitive radio system. Consider a cognitive network consisting of a base station PT equipped with N antennas, a single-antenna primary user P, and two single-antenna secondary users T and R. The network uses a STAR-RIS-assisted communication consisting of M elements. PT sends a primary user signal to the primary user P. The secondary users perform spectrum sensing with the assistance of STAR-RIS to detect whether the primary user is communicating. Step S2: Channel from base station PT to STAR-RIS STAR-RIS channels to secondary users T and R k∈{t,r} is modeled as a Rice channel, with the LoS component Written as h LoS =(u M (θ R,a )) H v N (θ PT,d ), where θ PT,d ,θ R,a ,θ Sk,d are the departure angle from PT to STAR-RIS, the arrival angle from PT to STAR-RIS, the departure angle from STAR-RIS to the secondary user, and v N (·),u M (·),q k,M (·) represents the corresponding array response vector, and the signal received by the secondary user is expressed as in is the beamforming vector, n0 is the noise at the receiving end, which has a mean of 0 and a variance of Gaussian distribution, p s is the transmission signal power of the base station PT, s0 is the transmission signal sent to the primary user, m∈{1,...,M} is the transmission and reflection factor matrix of STAR-RIS, and the perceived signal-to-noise ratio at the secondary user is in Step S3: Optimize the phase of STAR-RIS based on the statistical channel information, and use the triangle inequality to obtain the optimal phase: Step S4: Based on the probability distribution function and the optimal phase, the Gauss-Laguerre formula is used to obtain an approximate expression for the probability of correct detection: where t 1,i is the Laguerre polynomial L n The i-th root of (x), weight N p represents the polynomial order, τ th is the detection threshold, L k , is the noncentral chi-square distribution parameter, N s is the sampling number, I α (·) is a first-order Bessel function.

2. The spectrum sensing analysis method in a STAR-RIS assisted cognitive radio system according to claim 1, wherein: The design method of the optimal phase in step S3 is as follows: The perceptual signal-to-interference-noise ratio is expressed as where ξ h ,η h ,ξ g,k ,η g,k is the parameter of the Rice channel, since only is phase-dependent, so maximizing the signal-to-noise ratio through STAR-RIS phase design is equivalent to Using the triangle inequality, the conditions for it to hold are This gives the expression for the optimal phase.

3. The spectrum sensing analysis method in a STAR-RIS assisted cognitive radio system according to claim 2, wherein: For the derivation of the probability density function in step S4, let According to the central limit theorem, Z k Approximately follows a Gaussian distribution, It obeys the non-central chi-square distribution with 2 degrees of freedom, and the probability density function is Therefore, the probability of correct detection is expressed as By substituting the variables in the above formula and using Gauss-Laguerre numerical integration for approximation, we can obtain an approximate formula for the probability of correct detection.

Images