A maximum likelihood based NOMA signal spectrum sensing method

Abstract

The application discloses a maximum likelihood NOMA signal spectrum sensing method, a plurality of primary users utilize NOMA technology to jointly use a channel, when the number of primary users in the wireless channel is small, secondary users can also access the primary user channel, therefore, the secondary users need to detect not only the idle or busy state of the primary user channel, but also identify the number of primary users on the busy channel, when the primary user transmitting signal is a training sequence used for channel estimation, the transmitting signal constellation point is known, first, a joint probability density function is obtained according to the received signal of the secondary user as a likelihood function, an optimization problem is obtained, then, a global search algorithm is used to solve the optimization problem, an optimal value is obtained, compared with a decision threshold, the number of primary users on the channel is obtained; the method has the advantages that spectrum sensing can be performed under the condition of multiple primary users in the same channel, the number of primary users on the busy channel can be identified, and the optimal result can be obtained by sensing the signal received in only a single sensing duration.

Description

Technical Field

[0001] This invention relates to a cognitive radio technology, and more particularly to a maximum likelihood-based NOMA (Non-Orthogonal Multiple Access) signal spectrum sensing method, which estimates channel coefficients by maximizing the likelihood function to achieve NOMA signal spectrum sensing. Background Technology

[0002] With the rapid development of 5G wireless communication technology, the number of wireless devices needing to access networks via radio is increasing dramatically, and people's demand for wireless communication is growing. Although research on millimeter-wave communication is gradually increasing, its higher frequency bands result in drawbacks such as poor penetration and limited coverage, making large-scale coverage via millimeter-wave communication very costly. Therefore, improving the spectrum utilization of mid- and low-frequency bands is equally important for the development of 5G and the Internet of Things (IoT). To improve the spectrum utilization of mid- and low-frequency bands, many technologies have been designed, including Non-Orthogonal Multiple Access (NOMA) and Cognitive Radio. NOMA can allocate orthogonal resources in the same frequency, time, or code domain to multiple users, increasing system throughput without any bandwidth expansion. The basic idea of ​​Cognitive Radio is spectrum reuse; secondary users can use spectrum sensing technology to perceive the surrounding radio environment, detect spectrum holes, and access idle frequency bands without affecting the normal communication of primary users. Given that NOMA and Cognitive Radio can improve spectrum utilization in different ways, combining the two technologies is more helpful for improving spectrum efficiency.

[0003] Based on the theory that NOMA can allocate a time-frequency resource to multiple users, there are multiple primary users and one secondary user in the same frequency band. In traditional spectrum sensing, whether a secondary user senses the presence or absence of a primary user is transformed into a binary hypothesis problem. However, in NOMA-based spectrum sensing, since there are multiple primary users, the secondary user needs to sense the number of primary users present, so this can actually be transformed into a multiple hypothesis testing problem. Because different primary users in the same channel have different powers, the secondary user can adjust its power to match the power of any inactive primary user based on the sensing results, and then access the channel for communication.

[0004] When the primary user's transmitted signal is a training sequence used for channel estimation, the constellation points of the transmitted signal are known, and there is currently no relevant work. When the constellation points of the transmitted signal are unknown, existing methods utilize machine learning algorithms such as K-means and K-nearest-neighbor (KNN) to solve the spectrum sensing problem based on NOMA signals. Researchers classify the energy vectors received by multiple secondary users from multiple sensing durations (or time slots). These methods are based on clustering algorithms, dividing signals from different sensing durations into multiple categories. Therefore, they require multiple sensing durations of the received signal to achieve spectrum sensing, limiting their application. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a maximum likelihood-based NOMA signal spectrum sensing method. In this method, multiple primary users share the same channel using NOMA technology. Given that the constellation points of the primary users' transmitted signals are known, a maximum likelihood function is constructed to estimate the channel, the number of primary users working in the channel is sensed, and the channel is accessed based on the power of the non-working primary users. This method can effectively improve spectrum utilization and does not require multiple sensing durations; only the signal received during a single sensing duration is needed for spectrum sensing.

[0006] The technical solution adopted by this invention to solve the above-mentioned technical problems is: a NOMA signal spectrum sensing method based on maximum likelihood, characterized by including the following steps:

[0007] Step 1: In the cognitive radio system, assume there is only one secondary user and set the maximum number of primary users sharing the same channel using NOMA technology to be K, and assign serial numbers to these primary users starting from 1; in any sensing time slot, the secondary user receives a continuous signal as the NOMA signal, and then samples the NOMA signal received by the secondary user in any sensing time slot N times to obtain N discrete signals; where K≥2, N represents the number of samplings, and N≥200;

[0008] Step 2: For the N discrete signals of the secondary user, traverse each discrete signal in the order of sampling, and define the currently traversed discrete signal as the current discrete signal.

[0009] Step 3: Set the current discrete signal as the nth discrete signal, and denote it as r(n). Describe r(n) as: Determine H0 Determine H c Below

[0010] Where n = 1, 2, ..., N, H0 represents the assumption that there is no primary user in the channel, H1 represents the assumption that there is one primary user in the channel, and H...c This indicates that there are c primary users in the channel, H K Let K be the number of primary users in the channel, 1 ≤ c ≤ K, and w(n) represent the noise in r(n), which is independent and identically distributed additive white Gaussian noise with a mean of 0 and a variance of 0. i represents the sequence number of the i-th master user, which is a positive integer in the interval [1, K]. This represents the set of sequence numbers of a primary user when there is one primary user in the channel. This represents the set of sequence numbers of c primary users when there are c primary users in the channel. Ω represents the set of sequence numbers of K primary users when there are K primary users in the channel. i Let represent the power coefficient of the transmitted signal of the i-th primary user, which satisfies And Ω i ∈[0,1], This represents the total power of the primary user's transmitted signal, with a signal-to-noise ratio of . h i h represents the block Rayleigh fading channel coefficient from the i-th primary user to the secondary user. i The phase is uniformly distributed in [-π, π), h i The amplitude follows a Nakagami-m distribution, s i (n) represents the transmitted signal of the i-th primary user. Let represent a complex Gaussian distribution with mean a and variance b. This indicates that when the continuous signal received by the secondary user is sampled for the nth time, the j-th signal transmitted by the i-th primary user... i There are 1 symbols, 1 ≤ j i ≤J, where J represents the order of digital modulation during the process of obtaining multiple symbols through digital modulation of the transmitted signal of the i-th primary user after transmission;

[0011] Step 4: Let Let represent the null hypothesis of the i-th primary user. Let η represent the alternative hypothesis for the i-th primary user. i η represents the case where the i-th main user exists or does not exist. i =0 corresponds to η i =1 corresponds Then Transform into And thus obtain in,

[0012] Step 5: According to Obtain the probability density function of r(n), denoted as f(r(n)). Here, exp() represents an exponential function with base e, and the symbol "||" is the modulo operator;

[0013] Step 6: Traverse the next discrete signal in the order of sampling, take the next traversed discrete signal as the current discrete signal, and then return to step 3 to continue execution until all N discrete signals of the secondary user have been processed, and obtain the probability density function of each discrete signal of the secondary user.

[0014] Step 7: Calculate the joint probability density function of the N discrete signals of the secondary user, denoted as f(r). Then f(r) is used as the likelihood function, based on the maximum likelihood function. To achieve the goal, obtain Furthermore, when the primary user's transmitted signal is a training sequence used for channel estimation, the transmitted signal constellation points are... It is known that the only unknown parameters in the likelihood function can be determined by... This leads to the optimization problem, described as: Where r = [r(1), r(2), ..., r(N)], r(1) represents the first discrete signal, r(2) represents the second discrete signal, and r(N) represents the Nth discrete signal. The symbol for "any" is used.

[0015] Step 8: Solve using a global search algorithm First, in determining Given the search range and search precision, set the search scope for any g. i If there are L candidate values, then L coexist K 1 possible candidate vector; secondly, L K Substitute each of the possible candidate vectors into the likelihood function f(r) to calculate L. K The likelihood values ​​are calculated; then, the candidate vector corresponding to the largest likelihood value is taken as [g1, g2, ..., g K The estimated value of ]; where L≥30;

[0016] Step 9: Perform spectrum sensing on the NOMA signal received by the secondary user in any sensing time slot, targeting g. * =[g1 * g2 * ,…,g K * any element g in ] i * If g i * If the value is greater than or equal to the decision threshold λ, then a decision is made. If g is true,i * If it is less than the decision threshold λ, then a decision is made. Established; simultaneously, statistics on g * =[g1 * g2 * ,…,g K * The number of elements in the [] that are greater than or equal to the decision threshold λ, which is the number of working master users; where the value of λ is determined based on the false alarm probability.

[0017] In step 8, a global search algorithm is used to solve the problem. The process is as follows:

[0018] Step 8_1: According to The parameter represented Sure The required search range and search precision, g i The required search range is based on the power coefficient Ω of the transmitted signal of the i-th primary user. i Signal-to-noise ratio and the block Rayleigh fading channel coefficient h from the i-th primary user to the secondary user i To determine; then according to The required search range and search precision are set for any g. i If there are L candidate values, then L coexist K There are q possible candidate vectors, and the q-th possible candidate vector is denoted as g. (q) , Where 1≤q≤L K , G represents (q) Candidate values ​​for g1, G represents (q) Candidate values ​​for g2, G represents (q) China K Candidate values;

[0019] Step 8_2: Place L K Substitute each of the possible candidate vectors into the likelihood function f(r) to calculate L. K One likelihood value;

[0020] Step 8_3: Find L K The possible candidate vector corresponding to the largest likelihood value among the , denoted as g. * g * =[g1 * g2 * ,…,g K * ]; where g1* g2 * ,…,g K * Corresponding to g * The first element, the second element, ..., the Kth element in the array;

[0021] Step 8_4: Place g * =[g1 * g2 * ,…,g K * As [g1,g2,…,g] K The estimated value of ], i.e., g1 * g2 * ,...,g K * Corresponding to g1, g2, ..., g K The estimated value.

[0022] Compared with the prior art, the advantages of the present invention are as follows:

[0023] 1) In the present invention, when multiple primary users share the same channel using NOMA technology, secondary users perform spectrum sensing and can identify the number of primary users on the busy channel.

[0024] 2) This invention proposes a novel spectrum sensing method that requires only a single sensing duration signal to effectively perform spectrum sensing, unlike traditional clustering algorithms which require multiple signals. This method can significantly improve the speed and efficiency of spectrum sensing.

[0025] 3) The method of the present invention can maximize the likelihood function and obtain the optimal result when the primary user transmits a training sequence for channel estimation. Attached Figure Description

[0026] Figure 1 This is a block diagram illustrating the overall implementation of the method of the present invention;

[0027] Figure 2 This is a probability map of different states under different signal-to-noise ratios obtained using the method of the present invention. Detailed Implementation

[0028] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0029] The present invention proposes a maximum likelihood-based NOMA signal spectrum sensing method, the overall implementation block diagram of which is shown below. Figure 1 As shown, it includes the following steps:

[0030] Step 1: In a cognitive radio system, assume there is only one secondary user and set the maximum number of primary users sharing the same channel using NOMA technology to be K. Number these primary users starting from 1. In any sensing time slot, the secondary user receives a continuous signal as the NOMA signal. Then, the NOMA signal received by the secondary user in any sensing time slot is sampled N times to obtain N discrete signals. K ≥ 2. Theoretically, more primary users are better, but a larger number increases the demodulation complexity at the receiver. Researchers typically consider two primary users in the power domain NOMA, i.e., K = 2. Therefore, in this embodiment, K = 2. Without NOMA technology, K = 1. N represents the number of samples, N ≥ 200. In this embodiment, N = 250.

[0031] Step 2: For the N discrete signals of the secondary user, traverse each discrete signal in the order of sampling, and define the currently traversed discrete signal as the current discrete signal.

[0032] Step 3: Set the current discrete signal as the nth discrete signal, and denote it as r(n). Describe r(n) as: For example, when K=2 It is easy to see that, under different assumptions, r(n) follows a complex Gaussian distribution with different means and the same variance. Determining H0... Determine H c Below

[0033] Where n = 1, 2, ..., N, H0 represents the assumption that there is no primary user in the channel, H1 represents the assumption that there is one primary user in the channel, and H... c This indicates that there are c primary users in the channel, H K Let K be the number of primary users in the channel, 1 ≤ c ≤ K, and w(n) represent the noise in r(n), which is independent and identically distributed additive white Gaussian noise with a mean of 0 and a variance of 0. i represents the sequence number of the i-th master user, which is a positive integer in the interval [1, K]. This represents the set of sequence numbers of a primary user when there is one primary user in the channel. This represents the set of sequence numbers of c primary users when there are c primary users in the channel. This represents the set of sequence numbers of K primary users when there are K primary users in the channel. All are subsets of the set {1,2,...,K}, for example, when K=2 or Ω i Let represent the power coefficient of the transmitted signal of the i-th primary user, which satisfies And Ωi ∈[0,1], This represents the total power of the primary user's transmitted signal, with a signal-to-noise ratio of . h i h represents the block Rayleigh fading channel coefficient from the i-th primary user to the secondary user. i The phase is uniformly distributed in [-π, π), h i The amplitude follows a Nakagami-m distribution, and the distribution function is: The symbol "||" is the modulo operator, u is the independent variable of the distribution function, Ω represents the average power gain of the channel (in this embodiment, Ω = 1), m represents the degree of channel fading (a larger value of m indicates more severe channel fading; in this embodiment, m = 3), Γ(m) represents the gamma function, and e is the natural cardinality (e = 2.71…). i s1(n) represents the transmitted signal of the i-th primary user, H2 indicates that there are 2 primary users in the channel, Ω1 represents the power coefficient of the transmitted signal of the first primary user, Ω2 represents the power coefficient of the transmitted signal of the second primary user, h1 represents the coefficient of the block Rayleigh fading channel from the first primary user to the secondary user, h2 represents the coefficient of the block Rayleigh fading channel from the second primary user to the secondary user, s1(n) represents the transmitted signal of the first primary user, and s2(n) represents the transmitted signal of the second primary user. Let represent a complex Gaussian distribution with mean a and variance b. This indicates that when the continuous signal received by the secondary user is sampled for the nth time, the j-th signal transmitted by the i-th primary user... i There are 1 symbols, 1 ≤ j i ≤J, where J represents the order of digital modulation during the process of obtaining multiple symbols through digital modulation of the transmitted signal of the i-th primary user after transmission. Digital modulation methods include BPSK, QPSK, 16QAM, etc. When the digital modulation method is BPSK, J = 2. When the digital modulation method is QPSK, J=4. When the digital modulation scheme is 16QAM, J = 16.

[0034] Step 4: The purpose of spectrum sensing is to determine which hypothesis r(n) belongs to, i.e., H0, H1, ..., H2. c ... H K In which case, since hypothesis testing methods only apply to one alternative hypothesis corresponding to the null hypothesis, they cannot be directly applied to this spectrum sensing problem; therefore, a multiple hypothesis testing problem must be considered. Let... Let the null hypothesis (that the i-th primary user does not exist) be denoted by . Let η represent the alternative hypothesis (exists) for the i-th primary user. i η represents the case where the i-th main user exists or does not exist. i=0 corresponds to η i =1 corresponds Then Transform into And thus obtain in,

[0035] Step 5: The key to the spectrum sensing problem is to use estimate according to Obtain the probability density function of r(n), denoted as f(r(n)). Here, exp() represents an exponential function with the natural base e as the base, and the symbol "||" is the modulo operator.

[0036] Step 6: Traverse the next discrete signal in the order of sampling, take the next traversed discrete signal as the current discrete signal, and then return to step 3 to continue execution until all N discrete signals of the secondary user have been processed, and obtain the probability density function of each discrete signal of the secondary user.

[0037] Step 7: Calculate the joint probability density function of the N discrete signals of the secondary user, denoted as f(r). Then f(r) is used as the likelihood function, based on the maximum likelihood function. To achieve the goal, obtain Furthermore, when the primary user's transmitted signal is a training sequence used for channel estimation, the transmitted signal constellation points are... It is known that the only unknown parameters in the likelihood function can be determined by... This leads to the optimization problem, described as: Where r = [r(1), r(2), ..., r(N)], r(1) represents the first discrete signal, r(2) represents the second discrete signal, and r(N) represents the Nth discrete signal. The symbol is "any".

[0038] Step 8: Solve using a global search algorithm First, in determining Given the search range and search precision, set the search scope for any g. i If there are L candidate values, then L coexist K 1 possible candidate vector; secondly, L K Substitute each of the possible candidate vectors into the likelihood function f(r) to calculate L. K The likelihood values ​​are calculated; then, the candidate vector corresponding to the largest likelihood value is taken as [g1, g2, ..., g K The estimated value of ]; where L≥30, and in this embodiment L=31.

[0039] In this specific embodiment, step 8 uses a global search algorithm to solve the problem. The process is as follows:

[0040] Step 8_1: According to The parameter represented Sure The required search range and search precision, g i The required search range is based on the power coefficient Ω of the transmitted signal of the i-th primary user. i Signal-to-noise ratio and the block Rayleigh fading channel coefficient h from the i-th primary user to the secondary user i To determine; then according to The required search range and search precision are set for any g. i If there are L candidate values ​​(i.e., L values ​​are selected from the search range), then L coexist K There are q possible candidate vectors, and the q-th possible candidate vector is denoted as g. (q) , In this embodiment, the search range is [0,3] and the search precision is 0.1. In this embodiment, L = 31. K =31 2 =961, 1≤q≤L K , G represents (q) Candidate values ​​for g1, G represents (q) Candidate values ​​for g2, G represents (q) China K The candidate values, due to Ω i It is a given value. It is based on the given The obtained, h i It is a variable that follows a distribution and has strong randomness, therefore it is used To estimate The product value, such as Ω i Equals 0.3333 equals 1, h i The probability of a value greater than 3 is extremely small, so the search range is [0, 1.7320].

[0041] Step 8_2: Place L K Substitute each of the possible candidate vectors into the likelihood function f(r) to calculate L. K One likelihood value.

[0042] Step 8_3: Find L KThe possible candidate vector corresponding to the largest likelihood value among the , denoted as g. * g * =[g1 * g2 * ,…,g K * ]; where g1 * g2 * ,…,g K * Corresponding to g * The first element, the second element, ..., the Kth element in the array.

[0043] Step 8_4: Place g * =[g1 * g2 * ,…,g K * As [g1,g2,…,g] K The estimated value of ], i.e., g1 * g2 * ,…,g K * Corresponding as g1, g2, ..., g K The estimated value.

[0044] Step 9: Perform spectrum sensing on the NOMA signal received by the secondary user in any sensing time slot, targeting g. * =[g1 * g2 * ,…,g K * any element g in ] i * If g i * If the value is greater than or equal to the decision threshold λ, then a decision is made. If g is true, i * If it is less than the decision threshold λ, then a decision is made. Established; simultaneously, statistics on g * =[g1 * g2 * ,…,g K * The number of elements in the [] that are greater than or equal to the decision threshold λ, which is the number of working master users; wherein, the value of λ is determined according to the false alarm probability, and in this embodiment the value of the false alarm probability is 0.1.

[0045] The following simulations further illustrate the feasibility and effectiveness of the method of the present invention.

[0046] In the simulation, the maximum number of primary users sharing the same channel using NOMA technology is K=2, the power coefficient ratio of the transmitted signals of the two primary users is Ω1:Ω2=1:2, the number of samplings per time slot is N=250, and the variance of the noise is... The signal-to-noise ratio is -5 to 7, the average power gain of the channel is Ω = 1, the channel fading degree is m = 3, and the false alarm probability is 0.1.

[0047] Figure 2 The probability maps of different state detection under different signal-to-noise ratios obtained using the method of the present invention are given. Figure 2 This paper introduces three correct curves for judging H0 in case H0, H1 in case H1, and H2 in case H2, as well as nine incorrect curves for judging H0, H1, and H2 as other states respectively. Figure 2 As can be seen, with a fixed false alarm probability, the curve for correctly identifying H0 in case H0 remains essentially unchanged as the signal-to-noise ratio (SNR) increases. The curves for correctly identifying H1 in case H1 and H2 in case H2 gradually rise, while the curves for the other six incorrect identifications decrease as the SNR increases. The fact that the probability of correct identification increases with increasing SNR indicates that a higher SNR results in higher accuracy, demonstrating the effectiveness of the method described in this invention.

Claims

1. A method for sensing the spectrum of NOMA signals based on maximum likelihood, characterized in that... Includes the following steps: Step 1: In the cognitive radio system, assume there is only one secondary user and set the maximum number of primary users sharing the same channel using NOMA technology to be K, and assign serial numbers to these primary users starting from 1; in any sensing time slot, the secondary user receives a continuous signal as the NOMA signal, and then samples the NOMA signal received by the secondary user in any sensing time slot N times to obtain N discrete signals; where K≥2, N represents the number of samplings, and N≥200; Step 2: For the N discrete signals of the secondary user, traverse each discrete signal in the order of sampling, and define the currently traversed discrete signal as the current discrete signal. Step 3: Set the current discrete signal as the nth discrete signal, and denote it as r(n). Describe r(n) as: Determine H0 Determine H c Below Where n = 1, 2, ..., N, H0 represents the assumption that there is no primary user in the channel, H1 represents the assumption that there is one primary user in the channel, and H... c This indicates that there are c primary users in the channel, H K Let K be the number of primary users in the channel, 1 ≤ c ≤ K, and w(n) represent the noise in r(n), which is independent and identically distributed additive white Gaussian noise with a mean of 0 and a variance of 0. i represents the sequence number of the i-th master user, which is a positive integer in the interval [1, K]. This represents the set of sequence numbers of a primary user when there is one primary user in the channel. This represents the set of sequence numbers of c primary users when there are c primary users in the channel. Ω represents the set of sequence numbers of K primary users when there are K primary users in the channel. i Let represent the power coefficient of the transmitted signal of the i-th primary user, which satisfies And Ω i ∈[0,1], This represents the total power of the primary user's transmitted signal, with a signal-to-noise ratio of . h i h represents the block Rayleigh fading channel coefficient from the i-th primary user to the secondary user. i The phase is uniformly distributed in [-π, π), h i The amplitude follows a Nakagami-m distribution, s i (n) represents the transmitted signal of the i-th primary user. Let represent a complex Gaussian distribution with mean a and variance b. This indicates that when the continuous signal received by the secondary user is sampled for the nth time, the j-th signal transmitted by the i-th primary user... i There are 1 symbols, 1 ≤ j i ≤J, where J represents the order of digital modulation during the process of obtaining multiple symbols through digital modulation of the transmitted signal of the i-th primary user after transmission; Step 4: Let Let represent the null hypothesis of the i-th primary user. Let η represent the alternative hypothesis for the i-th primary user. i η represents the case where the i-th main user exists or does not exist. i =0 corresponds to η i =1 corresponds Then Transform into And thus obtain in, Step 5: According to Obtain the probability density function of r(n), denoted as f(r(n)). Here, exp() represents an exponential function with base e, and the symbol "||" is the modulo operator; Step 6: Traverse the next discrete signal in the order of sampling, take the next traversed discrete signal as the current discrete signal, and then return to step 3 to continue execution until all N discrete signals of the secondary user have been processed, and obtain the probability density function of each discrete signal of the secondary user. Step 7: Calculate the joint probability density function of the N discrete signals of the secondary user, denoted as f(r). Then f(r) is used as the likelihood function, based on the maximum likelihood function. To achieve the goal, obtain Furthermore, when the primary user's transmitted signal is a training sequence used for channel estimation, the transmitted signal constellation points are... It is known that the only unknown parameters in the likelihood function can be determined by... This leads to the optimization problem, described as: Where r = [r(1), r(2), ..., r(N)], r(1) represents the first discrete signal, r(2) represents the second discrete signal, and r(N) represents the Nth discrete signal. The symbol is "any". Step 8: Solve using a global search algorithm First, in determining Given the search range and search precision, set the search scope for any g. i If there are L candidate values, then L coexist K 1 possible candidate vector; secondly, L K Substitute each of the possible candidate vectors into the likelihood function f(r) to calculate L. K The likelihood values ​​are calculated; then, the candidate vector corresponding to the largest likelihood value is taken as [g1, g2, ..., g K The estimated value of ]; where L≥30; Step 9: Perform spectrum sensing on the NOMA signal received by the secondary user in any sensing time slot, targeting g. * =[g1 * g2 * ,…,g K * any element g in ] i * If g i * If the value is greater than or equal to the decision threshold λ, then a decision is made. If g is true, i * If it is less than the decision threshold λ, then a decision is made. Established; simultaneously, statistics on g * =[g1 * g2 * ,…,g K * The number of elements in the [] that are greater than or equal to the decision threshold λ, which is the number of working master users; where the value of λ is determined based on the false alarm probability.

2. The NOMA signal spectrum sensing method based on maximum likelihood according to claim 1, characterized in that... In step 8, a global search algorithm is used to solve the problem. The process is as follows: Step 8_1: According to The parameter represented Sure The required search range and search precision, g i The required search range is based on the power coefficient Ω of the transmitted signal of the i-th primary user. i Signal-to-noise ratio and the block Rayleigh fading channel coefficient h from the i-th primary user to the secondary user i To determine; then according to The required search range and search precision are set for any g. i If there are L candidate values, then L coexist K There are q possible candidate vectors, and the q-th possible candidate vector is denoted as g. (q) , Where 1≤q≤L K , G represents (q) Candidate values ​​for g1, G represents (q) Candidate values ​​for g2, G represents (q) China K Candidate values; Step 8_2: Place L K Substitute each of the possible candidate vectors into the likelihood function f(r) to calculate L. K One likelihood value; Step 8_3: Find L K The possible candidate vector corresponding to the largest likelihood value among the , denoted as g. * g * =[g1 * g2 * ,…,g K * ]; where g1 * g2 * ,…,g K * Corresponding to g * The first element, the second element, ..., the Kth element in the array; Step 8_4: Place g * =[g1 * g2 * ,…,g K * As [g1,g2,…,g] K The estimated value of ], i.e., g1 * g2 * ,…,g K * Corresponding as g1, g2, ..., g K The estimated value.

Images