CN103746756B - Cognitive radio networks interference estimation method based on simulated main customer attack - Google Patents

Abstract

The present invention relates to cognitive radio frequency spectrum management and security technology area, particularly to cognitive radio networks interference estimation method based on simulated main customer attack, including: trust time user and receive channel power, utilize any one channel in channel power detection cognitive radio networks the most occupied, channel occupation status and channel power are sent to a bunch head；Bunch head judges that whether occupied channel is for being hacked；Estimate to be hacked the interfering energy distribution of channel；Cognitive user of the present invention is by the radio environment around perception, detect and whether channel has malicious user and estimates the interfering energy distribution of malicious user, set up interference situation map further, by primary user's energy and malicious user energy separation, both can draw and only embody the distribution of malicious user interfering energy, and can combine with frequency spectrum cavity-pocket detection means again and draw frequency spectrum situation map accurately.

Description

Interference estimation method of cognitive radio network based on imitation of master user attack

Technical Field

The invention relates to the technical field of cognitive radio spectrum management and safety, in particular to an interference estimation method for a cognitive radio network based on imitation of master user attack, which constructs the interference situation of a malicious user by identifying an information source identity and estimating interference energy.

Background

With the development of scientific technology, spectrum resources are increasingly tense, but in the current spectrum management mode, even if an authorized User (also called a Primary User, PU for short) does not use an authorized frequency band, an unauthorized User (also called a trusted Secondary User, SU for short) cannot use the frequency band, and the utilization rate of the spectrum is greatly reduced under the condition. In order to solve the contradiction, a Cognitive Radio (CR) technology is proposed, in which an unauthorized user senses an idle frequency band through a spectrum sensing algorithm, and makes full use of spectrum resources on the premise of not affecting communication of the authorized user. All the trusted users mentioned in the method are trusted users, namely all data sent to the cluster head by the trusted users are real data or calculated according to the real data, and are not subjected to any malicious modification.

However, some new security issues are also introduced during spectrum access. Such as: the method comprises the following steps of imitating a Primary User Emulation Attack (PUEA for short), namely sending a signal imitating a Primary User signal by a Malicious User (MU for short), misleading a secondary User to believe that the Primary User is sending the signal, and judging that a channel is occupied by the Primary User and cannot use the channel.

At present, research aiming at PUEA mainly focuses on how a single trust secondary user detects attacks, and the influence of the attacks is not analyzed from a large-scale angle, and a corresponding coping scheme is made. For example, for the counterfeit main user attack, the anti-attack research proposed in the existing literature is to add some matching algorithms, such as HASH function, main user energy fingerprint, etc., at the transmitting and receiving ends to avoid the counterfeit attack from affecting the trust of the secondary user to correctly judge whether the channel is idle. The methods basically guarantee the dominance of all idle channels, but for the channel with malicious users, if the channel distribution center distributes the interference situation to the trust secondary users indiscriminately, the trust secondary users in the interference range still cannot use the channel safely due to excessive interference even if the trust secondary users obtain the channel, and therefore the safety of the cognitive network communication is reduced.

On one hand, when the condition of detecting whether the attack exists is researched, the research on the distribution situation of the attack energy is also significant to spectrum sensing safety. On the other hand, how to systematically and reliably excavate available spectrum holes in a network is one of the keys for realizing the cognitive radio technology, a spectrum situation graph refers to a 'spectrum map' which vividly and intuitively presents the spectrum holes in the cognitive radio, and the existing documents research the spectrum situation graph without attacks, so that when the attacks of a simulated main user exist, the situation graph can not distinguish the energy of the main user and the energy of a malicious user, the system loses the domination on the channel imitating the attacks, and the system capacity is reduced.

Disclosure of Invention

In order to solve the technical problems, the invention provides an interference estimation method of a cognitive radio network based on imitation of master user attack.

The invention discloses an interference estimation method of a cognitive radio network based on imitation of master user attack, which comprises the following steps: trusting the secondary user to receive the channel power, utilize the channel power to detect whether any channel in the cognitive radio network is occupied, will channel occupation state and channel power send to the cluster head; the cluster head judges whether the occupied channel is attacked or not; the interference energy distribution of the attacked channel is estimated.

Preferably, the detecting whether any channel in the cognitive radio network is occupied by using the channel power includes: if P exists on the channel_rk-T_m<Sigma, determining the channel as a suspected main user occupation, wherein T_mAs interference threshold, P_rkThe channel power of channel r received for the kth trusted secondary user.

Preferably, the determining, by the cluster head, whether the occupied channel is attacked includes:

the cluster head judges whether the channel is occupied according to the channel occupation condition judged by each user and sent by each user;

and the cluster head acquires the number of the information sources of the occupied channel and judges whether the channel is attacked or not according to the number of the information sources.

Preferably, the determining, by the cluster head, whether the channel is occupied according to the channel occupation status determined by each user sent by each user includes: and if any user considers that the channel is occupied by a suspected main user, the channel is considered to be occupied, and if all users consider that the channel is not occupied, the channel is in an idle state.

Preferably, the number of the information sources of the occupied channel acquired by the cluster head adopts a Gauss circle radius method.

Preferably, the determining whether the channel is attacked according to the number of the information sources includes: if the number n of the information sources is more than or equal to 2, the channel is in the attack of the counterfeit master user.

Preferably, the determining whether the channel is attacked according to the number of the information sources includes: if the number n =1 of the sources:

(1) firstly, determining whether the channel is in an occupied state, if Δ P_rIf' is less than or equal to the preset threshold, the channel is in an occupied state;

in the formula, the compound is shown in the specification, ${{\mathrm{\&Delta;P}}_r}^{\&prime;}=P_r^{T1}-P_r^{T2}=26.16\left[\log (f_1/f_2)\right]-[a_1\left(h_{\mathrm{re}}\right)-a_2\left(h_{\mathrm{re}}\right)],$ =26.16[log(f₁/f₂)]-[a₁(h_re)-a₂(h_re)]+x_nwherein:is a period T₁At a certain point pair with center frequency f₁The received power of the channel of (a),is a period T₂At the same position point pair with center frequency f₂Channel received power of a₁(h_re)、a₂(h_re) For shifting the antenna correction factor, x_nIs the average system noise;

(2) judging whether the channel has malicious interference users, if so, judging that the channel has the malicious interference users;

in the formula, the compound is shown in the specification,

$\mathrm{\&Delta;P}=\left(\right|P_{1,T_1}^{f_{D1}}-P_{1,T_1-1}^{f_i}|+|P_{2,T_1}^{f_{D1}}-P_{2,T_1-1}^{f_i}|+...+|P_{k,T_1}^{f_{D1}}-P_{k,T_1-1}^{f_i}|+...+|P_{M,T_1}^{f_{D1}}-P_{M,T_1-1}^{f_i}\left|\right)/M,$ f_i∈D-f_D1d is the set of all channels in the working frequency band,for the kth trusted secondary user at the Tth₁For channel f in one detection period_D1M refers to the number of trusted secondary users.

Preferably, the estimating of the interference energy distribution of the attacked channel includes estimating coordinates and transmission power of a malicious user of the channel, and specifically includes:

303-1A, estimating the angle of arrival of each source at the array antenna 1And the angle of arrival at the array antenna 2

303-1B, angle of arrival theta at array antennas 1 and 2 according to source i_iAnd θ'_iAnd the coordinates (a) of the array antennas 1,2₁,b₁)、(a₂,b₂) Solving the equation y of the line where the malicious information source is located_i=tanθ_i(x_i-a₁)+b₁And y_i=tanθ′_i(x_i-a₁)+b₁；

303-1C, solving s intersection points containing the coordinates of the malicious information source according to the equation of the straight line where the malicious information source is located, determining the coordinates of the malicious user from the s intersection points, and determining the transmitting power of the malicious user.

Preferably, the modifying the interference energy distribution comprises: uniformly distributing a plurality of sensors in an investigation area, calculating a receiving power error value at the position of the sensor, calculating an error correction value between adjacent sensor positions by adopting an interpolation method for the error value, and correcting the transmitting power at the coordinate of the channel malicious user.

According to the method, the cognitive user senses the surrounding radio environment to detect whether the malicious user exists in the channel or not and estimate the interference energy distribution of the malicious user, an interference situation graph is further established, the energy of the main user is separated from the energy of the malicious user, the interference energy distribution of the malicious user can be reflected, and an accurate spectrum situation graph can be obtained by combining with a spectrum cavity detection means.

Drawings

FIG. 1 is a diagram of a prior art network model;

FIG. 2 is a schematic flow chart of a first preferred embodiment of the interference estimation method based on imitating the attack of a master user in the cognitive radio network;

FIG. 3 is a flowchart illustrating a detailed implementation of step 301 of the interference estimation method for a cognitive radio network based on a master user attack simulation according to the present invention;

FIG. 4 is a flowchart illustrating a detailed implementation of step 302 of the interference estimation method for a cognitive radio network based on a simulated master user attack according to the present invention;

FIG. 5 is a flowchart illustrating a detailed implementation of step 303 of the interference estimation method based on imitating the attack of the master user in the cognitive radio network according to the preferred embodiment of the present invention;

FIG. 6 is a schematic flow chart of a second preferred embodiment of the interference estimation method based on imitating the attack of a master user in the cognitive radio network of the present invention;

FIG. 7 is an interference situation diagram obtained by the cognitive radio network of the present invention based on an interference estimation method that mimics the attack of a primary user;

FIG. 8 is a graph of interference situation obtained using the prior art;

FIG. 9 is a simulation diagram comparing the system capacity of the cognitive radio network based on the interference estimation method imitating the attack of the master user and the prior art.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Assume that the investigation region is a cognitive radio network composed of L primary users (each primary user occupies 1 channel), M trusted secondary users, N malicious users, and a pair of array antennas, as shown in fig. 1. The cognitive network adopts a cooperative spectrum sensing model, namely trust secondary users in a certain area are automatically combined into a cluster, and a cluster head is elected through an election algorithm. And each SU sends the detected energy value to the cluster head, and the cluster head performs unified processing. Generally, a primary user in the network is mobile, and a secondary user can freely join or leave the cognitive network under trust, but cannot move in the detection process.

The interference estimation method of the cognitive radio network based on imitating the attack of the master user, as shown in figure 2, comprises the following steps:

step 301, trusting a secondary user to receive channel power, detecting whether any channel in a cognitive radio network is occupied by using the channel power, and sending a channel occupation state and the channel power to a cluster head;

step 302, judging whether the occupied channel is attacked or not by the cluster head;

step 303, estimating the interference energy distribution in the attacked channel.

Embodiments of the various steps of the present invention are described below.

The step 301 of trusting the secondary user to receive the channel power, detecting whether any one channel in the cognitive radio network is occupied by using the channel power, and sending the channel occupation state and the channel power to the cluster head, as shown in fig. 3, includes:

301-1 for the ith channel (denoted f) in the network_i) Each trusted secondary user SU_kReceiving the channel power P_rkJudging whether the channel is occupied by a suspected master user;

suppose a kth trusted secondary user SU_kIs denoted as P_rkIf channel f_iIn presence of P_rk≥T_mThen determine the messageThe lane is occupied by a suspected main user and marked as f_DiOtherwise, the channel is determined to be unoccupied. Wherein, T_mFor the interference threshold, the value of which is different according to the modulation systems accepted by different cognitive users, the detailed setting method thereof is preferably as follows: li yanlin, corning, cunning, senior vitae, erysipelas, extended discussion of frequency point thresholds in cognitive radio systems, proceedings of electrical wave science, 2010, 25 (9): 169-171

301-2, will receive power P_rkAnd the channel occupation condition judged by the user is sent to the cluster head.

Whether SU_kIf the suspected master user occupation is detected, the receiving power P is required_rkAnd sending the data to the cluster head.

The step 302 of judging whether the occupied channel is attacked by the cluster head specifically includes the following steps:

the cluster is a set formed by freely combining users in a certain region range, and the cluster head is a user arbitrarily selected from the cluster; the clusters and cluster heads are common concepts in the art and will not be described further. As shown in fig. 4, the method specifically includes:

302-1, judging whether the channel is occupied or not by the cluster head according to the channel occupation condition judged by each user sent by each user;

according to the principle that a master user in cognitive radio preferentially uses an authorized channel, judging whether the channel is occupied or not after a cluster head obtains the channel occupation condition judged by each user sent by each user in the cluster;

that is, if some users consider that the channel is occupied by a suspected primary user, the channel is considered to be occupied, and if all the users consider that the channel is not occupied, the channel is in an idle state. If the channel is idle, the process returns to step 301 to detect additional channels.

302-2, the cluster head acquires the number of the information sources of the occupied channel, and judges whether the channel is attacked or not according to the number of the information sources;

302-2A, the number of the information sources of the occupied channel acquired by the cluster head

As an implementation manner, the cluster head determines a channel f occupied by a suspected primary user_DiAnd the number n of the medium information sources is used for judging the number of the information sources of a certain channel according to the reporting condition of each trust secondary user.

As another preferred embodiment, the cluster head determines a channel f occupied by a suspected primary user_DiEstimating the channel f occupied by the suspected main user by adopting a Geiger circle radius method according to the number n of the medium information sources_DiNumber n of intermediate sources

The Geiger circle radius method is referred to Chengmuio, left Jingkun, Lidong, etc. Power System harmonic detection based on Geiger circle and TAM [ J ]. Nature science, edition, 2012,38(005): 612-.

302-2B, judging whether the channel is attacked or not according to the number of the information sources

If the number n of the information sources is larger than or equal to 2, the channel is under the attack of counterfeit master users, and because two master users cannot exist in one channel at the same time, the process directly enters the step 303.

If the number n =1 of the information sources, only one user is in the channel, and whether the user is a master user or a malicious user cannot be determined, and the process enters 302-2C.

For judging whether the channel is occupied by the malicious user, the embodiment of the invention provides two ways, which are respectively as described in the following 302-2C and 302-2D, and both can judge whether the channel is occupied by the malicious user.

As an implementation manner, the determining whether the channel is occupied by a malicious user includes:

302-2C, judging whether the channel is occupied by a malicious user or not

(1) Firstly, whether the channel is in an occupied state or not is judged

Due to malicious usersThe position and the transmission power are generally fixed, so that in any two periods T₁And T₂The difference in received power of the signals transmitted on two different channels at the same location being frequency dependent, i.e. the difference is only

${{\mathrm{\&Delta;P}}_r}^{\&prime;}=P_r^{T1}-P_r^{T2}=26.16\left[\log (f_1/f_2)\right]-[a_1\left(h_{\mathrm{re}}\right)-a_2\left(h_{\mathrm{re}}\right)]---\left(1\right)$

Wherein, a₁(h_re)、a₂(h_re) For shifting the antenna correction factor, f₁、f₂Is the center frequency of the two channels.

Then, if each SU is detected at T, it can be obtained from equation (1)₁Period to channel f_D1Received power ofAnd at T₁-1 period pair channel f_D2Received power ofDifference value Δ P of_r' < 2 >, the channel f is indicated_D1Is in an occupied state. Wherein, =26.16[ log (f)₁/f₂)]-[a₁(h_re)-a₂(h_re)]+x_n，x_nIs the average system noise.

(2) If the channel is in the occupied state, judging whether the channel has malicious interference users.

The specific detection algorithm is as follows:

calculating the current detection period T₁In f_D1Upper received power and last period T₁-1 mean value of the difference of the received power of the detection point at all other channels:

$\mathrm{\&Delta;P}=\left(\right|P_{1,T_1}^{f_{D1}}-P_{1,T_1-1}^{f_i}|+|P_{2,T_1}^{f_{D1}}-P_{2,T_1-1}^{f_i}|+...+|P_{k,T_1}^{f_{D1}}-P_{k,T_1-1}^{f_i}|+...+|P_{M,T_1}^{f_{D1}}-P_{M,T_1-1}^{f_i}\left|\right)/M---\left(2\right)$

in the formula (2), f_i∈D-f_D1D is the set of all channels in the working frequency band,for the kth trusted secondary user at the Tth₁For channel f in one detection period_D1M refers to the number of trusted secondary users.

If Δ P is not less than f_D1The main user is different from other channels, and the channel has no malicious counterfeit user and continues to detect the next channel; if Δ P<Then explain f_D1Coincides with the energy fingerprint of a certain channel, indicating that f is now_D1There is malicious user occupancy in the channel, where =26.16[ log (f)₁/f₂)]-[a₁(h_re)-a₂(h_re)]+x_n，x_nIs the average system noise.

As another implementation, the determining whether the channel is occupied by a malicious user includes:

and the receiver of the 302-2D cognitive user receives a signal similar to that sent by an authorized main user, and starts a PUE attack detection system. In an observed time window, a power spectral density calculation is firstly carried out on a time domain sample of a physical layer to obtain a frequency domain sample, and then wavelet decomposition is carried out. Since the variation of fading channel characteristics is generally concentrated in a low frequency band, a low frequency part (an approximation part) is selected when the original signal is reconstructed at a specific scale, and the correlation is enhanced. Corresponding features of the isolated signal are then extracted and detected using a trained support vector machine for training vectors. If the detection result is f_D1Malicious user occupation exists in a channel, which is specifically referred to as: detection of simulated spoofing attacks by authorized users in cognitive radioMethod [ J]China military convertant, 2012,6: 015.

And when the malicious user is detected to occupy, determining the channel as attacked.

As an implementation manner, the step 303 of estimating the interference energy distribution in the attacked channel, specifically as shown in fig. 5, includes the following steps:

303-1, estimating the coordinates and transmitting power of the channel malicious user

303-1A, estimating the angle of arrival of each source at the array antenna 1And the angle of arrival at the array antenna 2

The estimation of the angle of arrival may employ the MUSIC algorithm, which is described in the following references: wangyongliang, chenhui, penning, ten thousand groups, space spectrum estimation theory and algorithm [ M ]. beijing, qinghua university press, 2013. The angle of arrival estimation may also be performed in another way, as described in the following references: yan stiffness, Jinming, John, applied to transform domain two-dimensional angle of arrival fast estimation algorithm [ J ] for arbitrary arrays, academic newspaper, 2013.

303-1B, angle of arrival theta at array antennas 1 and 2 according to source i_iAnd θ'_iAnd the coordinates (a) of the array antennas 1,2₁,b₁)、(a₂,b₂) Solving the equation y of the line where the malicious information source is located_i=tanθ_i(x_i-a₁)+b₁And y_i=tanθ′_i(x_i-a₁)+b₁。

303-1C, solving s intersection points containing the coordinates of the malicious information source according to the equation of the straight line where the malicious information source is located, determining the coordinates of the malicious user from the s intersection points, and determining the transmitting power of the malicious user;

and (3) s-n coordinates which are not the source are selected from the s intersection points containing the malicious source coordinates, and the coordinates of the malicious user are selected according to the Dirichlet principle and the hypothesis testing algorithm.

Dirichlet principle: putting m articles into n (n) in any mode<m) drawers, at least one drawer is provided with two or more articles. The application of the invention is as follows: the correct intersection combination contains n-n or less₁+1 points of intersection on the line of array antenna l, or combinations thereof of no more than n-n₂The +1 intersections are on the direction finding line of the array antenna 2.

Step 302 has found that the number of malicious users in the channel is n, and we divide the s intersections into n groups of combinations, which have the same structureAnd (4) combination. The 1 st group of intersection combinations are detected as follows: from step (21), it is known that the array antenna 1 has n₁A line of direction finding, n₁N is less than or equal to n; if the number of collinear intersections in the combination is more than n-n₁+1, the dirichlet principle is not met, so the combination is excluded. If group 1 is combined with n of array antenna 1₁The strip line of direction-finding conforms to the Dirichlet principle, again with n for the array antenna 2₂In the case of a lateral line (n)₂N) or less), whether the number of collinear intersections in the combination is greater than n-n₂And +1, if the number is larger than the preset value, the Dirichlet principle is not satisfied, and the combination is excluded. If so, the combination is retained and a hypothesis test is waited.

For the restAnd performing the same detection on the group combinations, assuming that C combinations are left after the detection, and assuming that each intersection point in the left combinations is the position of the MU, performing hypothesis test on the C combinations:

(1) for the 1 st combination, selecting the received power P of n SUs received by the cluster head_r1,P_r2,...,P_rnAccording toCalculating emission power PT 'of n malicious users'₁,PT′₂,...,PT′_nAnd according to PT'₁,PT′₂,...,PT′_nCalculating received power P 'of additional L-n SU'_r1,P′_r2,...,P′_r(L-n)And is prepared from P'_rkWith the power P received by the kth SU_rkSubtract (i =1, 2.., L-n) to yield the 1 st Δ P'_r， ${{\mathrm{\&Delta;P}}^{\&prime;}}_r=\underset{i=M+1}{\overset{L}{\mathrm{\&Sigma;}}}{P^{\&prime;}}_{\mathrm{rk}}-P_{\mathrm{rk}}.$

(2) The same calculation was carried out for the remaining C-1 combinations to give a plurality of different combinations of Δ P'_rAnd preferably 10.

(3) Is delta P'_rThe smallest of those combined coordinates is determined as the malicious user coordinates.

(4) Is delta P'_rMinimum group PT'₁,PT′₂,...,PT′_nAnd determining the transmission power of the n malicious users.

Preferably, in order to reduce the calculation error, the n malicious users SU calculating the transmission power PT' of the malicious user have to select SUs closer to the malicious user MU node that needs to be calculated. Since the cluster head knows the location of each SU, and also the location of each group of MU nodes, the n SUs are chosen by the cluster head based on the previous calculations.

The transmit power of the malicious user obtained from the previous step is PT', and preferably, the transmit power of the malicious user may also be modified, so as to obtain a more accurate transmit power of the malicious user, that is, the method further includes:

step 303-2: uniformly arranging a plurality of sensors, calculating the error value of the receiving power at the position of the sensor, calculating the error correction value between the positions of the adjacent sensors by adopting an interpolation method for the error value, and correcting the transmitting power at the position of the channel malicious user coordinate, namely adding the transmitting powers at the corresponding positions.

The theoretical receiving power of the sensor is P after the free space fading and the shadow fading_r', will P_r' with the true value P received by the preset sensor_rAre subtracted by the difference Δ P_rI.e. the error, Δ P, there_r=P_r′-P_r. And then, uniformly inserting X values between the error values of two adjacent sensors to obtain error correction values of each position coordinate between the two sensors, and then correcting the transmission power at the channel malicious user coordinate.

For example, the error at the sensor 1 is Δ P_r1Error at sensor 2 is Δ P_r2And the linear distance between the sensors 1 and 2 is d, the sensor 1 is separated from the sensor 2 by a distance d on a straight line formed by the position coordinates of the sensors 1 and 2₁(d₁<d) An error correction value ofIn this way, error correction values can be calculated for the entire investigation region.

As another implementation manner, in step 303, the interference energy distribution in the attacked channel is estimated, and an interference temperature estimation algorithm based on the Kriging method is adopted.

A large number of sensors are installed in an investigation area, and the interference temperature obtained by each sensor is changed according to the change of the spatial position due to the different positions of the sensors, but due to the correlation of signals sent by interference sources, the interference temperature of the position of each sensor has certain spatial correlation. The interference temperature value obtained by each sensor is the observed value of the space variable model, and the interference temperature space distribution in the area under examination can be estimated by using a Kriging method according to the observed value. See in particular the following documents: interference temperature spatial distribution estimation [ J ] Chongqing university academic newspaper ISTIC EI,2011 and 34(2) based on Kriging method in Von Wenjiang, Lijun cognitive radio.

Preferably, the interference estimation method based on imitating the attack of the master user in the cognitive radio network of the present invention, as shown in fig. 6, further includes: step 304, constructing an interference situation map of the channel, where the step is only for facilitating understanding of an interference display mode of the channel, and is not a necessary step for implementing the present invention, and specifically, the step is:

1. and calculating the distribution value of the interference energy by the following method:

P_r′=PT′-PL （3）

in the formula (3), the reaction mixture is,P_r' interference energy, P, that can be received at each coordinate point_tIs the transmission power of the interference source, PL_sIn order to be a loss caused by the shadow effect,for free space path loss, preferably the free space path loss adopts the HATA model, i.e.

$\stackrel{\&OverBar;}{\mathrm{PL}(f,d)}\left(\mathrm{dB}\right)=69.55+26.16\log f_c-13.82\log h_{\mathrm{te}}-a\left(h_{\mathrm{re}}\right)+(44.9-6.55\log h_{\mathrm{te}})\log d$

From the above three equations, the received power P_r' and P_t、f_c、h_te、h_reAnd d are related.

Under the condition that the same transmitting power is the same and the working frequency is different at the same transmitting point and the same receiving point, the difference value of the receiving power is as follows:

${{\mathrm{\&Delta;P}}_r}^{\&prime;}=P_r^{f1}-P_r^{f2}=P_t^{f1}-{\mathrm{PL}}^{f1}-(P_t^{f2}-{\mathrm{PL}}^{f2})$

because of the fact thatTherefore, it is not only easy to use

${{\mathrm{\&Delta;P}}_r}^{\&prime;}=-{\mathrm{PL}}^{f1}-(-{\mathrm{PL}}^{f2})={\mathrm{PL}}^{f2}-{\mathrm{PL}}^{f1}=\stackrel{\&OverBar;}{{\mathrm{PL}}^{f2}(f,d)}+{\mathrm{PL}}_s^{f2}-(\stackrel{\&OverBar;}{{\mathrm{PL}}^{f1}(f,d)}+{\mathrm{PL}}_s^{f1})$

Since for the same MU and the same SU,h_te、h_red are all equal, so

ΔP_r′=26.16[log(f₁/f₂)]-[a₁(h_re)-a₂(h_re)]

In the HATA model, a (h) is used for urban environments of different scales_re) Different. For large cities, the mobile antenna correction factors are:

a(h_re)=8.29(log1.54h_re)²-1.1dB f_c≤300MHz

a(h_re)=3.2(log11.75h_re)²-4.97dB f_c≥300MHz

for medium and small cities, it is:

a(h_re)=(1.1logf_c-0.7)h_re-(1.56logf_c-0.8)dB

from the above analysis, it can be seen that, for the same transmission point, the power of the signals transmitted at two times is the same but the working frequency f_cDifference value delta P of receiving power of same SU under different conditions_rOnly with the centre frequency f of the two channels_cIt is related. Therefore, set as

=26.16[log(f₁/f₂)]-[a₁(h_re)-a₂(h_re)]+x_n

2. Drawing interference situation map

And adding the error correction value to the calculated value of the interference energy distribution, so that a relatively real interference situation map showing the interference energy distribution in the investigation region can be drawn.

According to the method, the cognitive user senses the surrounding radio environment to detect whether the malicious user exists in the channel or not and estimate the interference energy distribution of the malicious user, an interference situation graph is further established, the energy of the main user is separated from the energy of the malicious user, the interference energy distribution of the malicious user can be reflected, and an accurate spectrum situation graph can be obtained by combining with a spectrum cavity detection means.

The interference situation diagrams are shown in fig. 7 and 8, wherein the colors of the coils in the diagrams are from dark to light, which indicates that the interference power received by the secondary user is decreased from large. Black indicates that the SU receives the largest amount of interference energy, e.g., A, B, C (which are also the locations of malicious users); as the color becomes lighter, the interference energy that the SU can receive becomes smaller, as indicated by the blank area shown in the lower left corner of the figure, where it is almost undisturbed. The interference power in the graph can reach 20dB at most and is only-15 dB at least, if the interference power at a certain position is smaller than the signal-to-noise ratio threshold of a receiver, the channel can still be used under the condition that the channel is known to be attacked by counterfeiting; however, if the interference power is greater than the snr threshold of the receiver, it is not suitable to use the channel.

Fig. 7 is an interference energy distribution situation of an actual malicious user in a T period and a channel i, and fig. 8 is an interference energy distribution situation of a malicious user in a T period and a channel i estimated by using the method. Comparing fig. 7 and 8, it can be seen that the coordinates of the areas with the same color shades in the lines in fig. 7 and 8 are also approximately the same, which indicates that the interference power with the same size in the two figures is approximately the same in the distribution position, and the method can estimate the interference energy distribution more accurately. The cluster head distributes channels to the SU according to the interference energy distribution map of the malicious user obtained by the scheme, so that the successful communication probability is greatly improved, namely, the communication safety is improved.

The channel capacities taking into account the interference situation and not taking into account the interference situation are analyzed as follows:

definition 1: the system capacity is the sum of the average capacities of the channels in the area under consideration. The average capacity of each channel is the average value of the channel capacity which can be obtained by each trusted secondary user if accessing the primary user channel. According to the shannon formula, the maximum channel capacity which can be obtained by each trust secondary user if accessing the primary user channel is as follows:

$C\left(N_i\right)=B\log (1+\frac{S}{N_0+N_i})---\left(4\right)$

wherein B is channel bandwidth, S is trust secondary user transmitting power, N_iReceived power, i.e. interference power, for trusted secondary users to malicious users, N₀As is other noise interference on the channel.

From equation (4), the capacity of the i-th cognitive user based on the interference situation map is

$C\left(N_i\right)=\left\{\begin{array}{c}B\log (1+S/N_0),\\ B\log [1+S/(N_i+N_0)],\end{array}\right.---\left(5\right)$

The capacity of the ith cognitive user without considering the interference situation is

$C\left(N_i\right)=\left\{\begin{array}{c}B\log (1+S/N_0),\\ 0,\end{array}\right.---\left(6\right)$

As can be seen from a comparison of equations (5) and (6), the channel capacities in the two cases differ from each other in the presence of a spoofing attack. Since the latter does not consider the attack situation, when there is a counterfeit attack, the latter may misunderstand as a primary user and give up the dominance for the channel, so that the channel capacity is 0.

Definition 2: capacity C of ith channel_iThe capacity of the channel is averaged for each trusted secondary user in the system. System capacity C_SIs the sum of the capacity of each channel, is

$C_S=\underset{i=1}{\overset{L}{\mathrm{\&Sigma;}}}{C}_i=\underset{i=1}{\overset{L}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}C\left(N_j\right)---\left(7\right)$

For the whole system, the channel capacity under attack can be obtained by considering the spectrum situation diagram of the interference situation, and the capacity of the system is further improved. When there is only one malicious user in the systemThe difference value of the system capacity under the two methods is Delta C_S=Blog[1+S/（N_i+N₀）]When m malicious users exist, the capacity difference value Delta C of the system_S=mBlog[1+S/（N_i+N₀）]So the system capacity difference is larger when there are more malicious users. But this trend is decreasing as the number of malicious users is greater than the number of channels. Because when the number of the malicious users is equal to the number of the channels, each time one MU is added, N is added_iContinued increase in the molecular weight of C (N) obtained from the formula (5)_i) The system capacity is further reduced, that is, the capacity obtained by 1 MU attacking one channel or 2 attacking the same channel formula (5) is different; since the equation (6) does not have such a difference, the capacity decrease tendency becomes significantly slow when the number of MUs is equal to the number of channels.

Fig. 9 shows the system capacity of the multi-channel multi-malicious user environment of the present invention, which is higher than the system capacity of the cognitive network in the ZHANG method.

The purpose, technical solutions and advantages of the present invention are further described in detail by using the embodiments or examples of the present invention, it should be understood that the above embodiments or examples are only preferred embodiments of the present invention, and are not intended to limit the present invention, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (6)

1. The interference estimation method of the cognitive radio network based on imitating the attack of the master user comprises the following steps: trusting the secondary user to receive the channel power, utilize the channel power to detect whether any channel in the cognitive radio network is occupied, will channel occupation state and channel power send to the cluster head; the cluster head judges whether the occupied channel is attacked or not; estimating the interference energy distribution of the attacked channel; the method is characterized in that:

wherein, the cluster head judges whether the occupied channel is attacked or not, including: the cluster head judges whether the channel is occupied according to the channel occupation condition judged by each user and sent by each user; the method comprises the following steps: if a user thinks that the channel is occupied by a suspected master user, the channel is regarded as occupied, the cluster head obtains the number of the information sources of the occupied channel, and whether the channel is attacked or not is judged according to the number of the information sources; if all users consider that the channel is not occupied, the channel is in an idle state;

modifying the interference energy distribution, comprising: uniformly arranging a plurality of sensors, calculating a receiving power error value at the position of the sensor, calculating an error correction value between adjacent sensor positions by adopting an interpolation method for the error value, and correcting the transmitting power of the channel malicious user coordinate; calculating a receiving power error value at the position of the sensor, wherein the receiving power error value is obtained by subtracting a theoretical receiving power of the sensor from a real value received by a preset sensor;

calculating error correction values for the entire investigation region, including: on a straight line formed by the position coordinates of the sensor 1 and the sensor 2, a distance d from the sensor 1₁An error correction value ofWherein Δ P_r1Error at sensor 1, Δ P_r2Is the error at sensor 2 and d is the linear distance between sensors 1 and 2.

2. The interference estimation method based on imitating the attack of the master user in the cognitive radio network according to claim 1, characterized in that: the detecting whether any channel in the cognitive radio network is occupied by using the channel power includes: if P exists on the channel_rk≥T_mDetermining that the channel is occupied by a suspected main user, wherein T_mAs interference threshold, P_rkThe channel power of channel r received for the kth trusted secondary user.

3. The interference estimation method based on imitating the attack of the master user in the cognitive radio network according to claim 1, characterized in that: and the number of the cluster heads acquiring the information sources of the occupied channels adopts a Gauss circle radius method.

4. The interference estimation method based on imitating the attack of the master user in the cognitive radio network according to claim 1, characterized in that: the judging whether the channel is attacked or not according to the number of the information sources comprises the following steps: if the number n of the information sources is more than or equal to 2, the channel is in the attack of the counterfeit master user.

5. The interference estimation method based on imitating the attack of the master user in the cognitive radio network according to claim 1, characterized in that: the judging whether the channel is attacked or not according to the number of the information sources comprises the following steps: if the number n of the information sources is 1, then:

(1) firstly, determining whether the channel is in an occupied state, if Δ P_rIf' is less than or equal to the preset threshold, the channel is in an occupied state; wherein,

${{\mathrm{\&Delta;P}}_r}^{\&prime;}=P_r^{T1}-P_r^{T2}=26.16\&lsqb;log(f_1/f_2)\&rsqb;-\&lsqb;a_1\left(h_{re}\right)-a_2\left(h_{re}\right)\&rsqb;,$

＝26.16[log(f₁/f₂)]-[a₁(h_re)-a₂(h_re)]+x_n，is a period T₁At a certain point pair with center frequency f₁The received power of the channel of (a),is a period T₂At the same position point pair with center frequency f₂Channel received power of a₁(h_re)、a₂(h_re) For shifting the antenna correction factor, x_nIs the average system noise;

(2) judging whether the channel has malicious interference users, if so, judging that the channel has the malicious interference users; wherein,

$\mathrm{\&Delta;}P=\left(\right|P_{1,T_1}^{f_{D1}}-P_{1,T_1-1}^{f_i}|+|P_{2,T_1}^{f_{D1}}-P_{2,T_1-1}^{f_i}|+\mathrm{...}+|P_{k,T_1}^{f_{D1}}-P_{k,T_1-1}^{f_i}|+\mathrm{...}+|P_{M,T_1}^{f_{D1}}-P_{M,T_1-1}^{f_i}\left|\right)/M,$ f_i∈D-f_D1d is the set of all channels in the working frequency band,for the kth trusted secondary user at the Tth₁For channel f in one detection period_D1M refers to the number of trusted secondary users.

6. The interference estimation method based on imitating the attack of the master user in the cognitive radio network according to claim 1, characterized in that: the estimating of the interference energy distribution of the attacked channel includes estimating coordinates and transmission power of a malicious user of the channel, and specifically includes:

303-1A, estimating the angle of arrival of each source at the array antenna 1And the angle of arrival at the array antenna 2

303-1B, angle of arrival theta at array antennas 1 and 2 according to source i_iAnd θ'_iAnd the coordinates (a) of the array antennas 1,2₁,b₁)、(a₂,b₂) Solving the equation y of the straight line where the malicious information source is located_i＝tanθ_i(x_i-a₁)+b₁And y_i＝tanθ'_i(x_i-a₁)+b₁；

303-1C, solving s intersection points containing the coordinates of the malicious information source according to the equation of the straight line where the malicious information source is located, determining the coordinates of the malicious user from the s intersection points, and determining the transmitting power of the malicious user.