CN105227508A - Electromagnetic leakage signal recovers and Enhancement Method and system - Google Patents

Abstract

The present invention relates to a kind of electromagnetic leakage signal to recover and Enhancement Method and system.The method comprises: environmentally the feature of noise and wireless channel noise sets up the electromagnetic leakage signal model comprising ambient noise and wireless channel noise; High-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part is divided into also to carry out Adaptive inverse filtering process respectively received electromagnetic leakage signal according to electromagnetic leakage signal model; By described high-frequency electromagnetic leakage signal part and described low frequency electromagnetic leakage signal part through respective blind equalizer to obtain respective output signal, utilize convex combination mode to merge two output signals and recover and enhancing model to obtain electromagnetic leakage signal.The recovery that the embodiment of the present invention provides and Enhancement Method, according to the physical characteristic of electromagnetic leakage signal, can be good at recovering to make the vision signal received have stronger intelligibility with increase vision signal, thus improve the attacking ability of TEMPEST.

Description

Electromagnetic leakage signal recovers and Enhancement Method and system

Technical field

The present invention relates to electromagnetic environment technical field of safety protection, particularly relate to a kind of electromagnetic leakage signal and recover and Enhancement Method and system.

Background technology

Electromagnetic leakage comprises radiation leakage and leakage.At present, TEMPEST is day by day perfect, and the leakage electromagnetic wave signal that digital system sends is very faint.When if desired utilizing electromagnetic leakage signal, this electromagnetic leakage signal can be submerged in a large amount of noises.But there is uncertainty due to noise in actual environment, cause extracting and recover electromagnetic leakage signal there is challenge.

Vision signal reprocessing rate is restriction TEMPEST (TransientElectromagneticPulseEmanationSurveillanceTechno logy, electromagnetic environment security protection) one of key technology, pass through after-treatment system, even reducing under the condition that hardware device is required under identical hardware condition, still intercepting and capturing and the reduction of information be can complete, quality and the definition of image and word improved.

Because electromagnetic radiation environment has uncertainty, retrieving algorithm needs to have adaptivity to received electromagnetic leakage signal, and traditional filter and amplification method has then been difficult to.In prior art, Detection of Weak Signals and method of reducing mostly need certain priori conditions, do not utilize solution signals revivification.For this reason, scholars have carried out large quantity research, to improve the signal to noise ratio of Received signal strength.But, only improve the reprocessing that signal to noise ratio carries out vision signal, recovery and the enhancing of vision signal can not be completed well.

Summary of the invention

One of them object of the present invention is to provide a kind of electromagnetic leakage signal to recover and strengthens and system, only improves signal to noise ratio can not recover the technical problem with augmented video signal very well to solve in prior art.

First aspect, embodiments provides a kind of electromagnetic leakage signal and recovers and Enhancement Method, comprising:

Environmentally the feature of noise and wireless channel noise sets up the electromagnetic leakage signal model comprising ambient noise and wireless channel noise;

High-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part is divided into also to carry out Adaptive inverse filtering process respectively received electromagnetic leakage signal according to described electromagnetic leakage signal model;

By described high-frequency electromagnetic leakage signal part and described low frequency electromagnetic leakage signal part through respective blind equalizer to obtain respective output signal, utilize convex combination mode to merge two output signals and recover and enhancing model to obtain electromagnetic leakage signal.

Alternatively, the feature of described environmentally noise and wireless channel noise also comprises after setting up the step of the electromagnetic leakage signal model comprising ambient noise and wireless channel noise:

Form the cost function of blind equalizer according to statistical independence parameter, with make the output signal of estimation model of receptions electromagnetic leakage signal and the electromagnetic leakage signal that receives close as far as possible.

Alternatively, describedly form the cost function of blind equalizer according to statistical independence parameter, with make after the output signal of estimation model of receptions electromagnetic leakage signal and the close as far as possible step of the electromagnetic leakage signal that receives, also comprise:

Probability-distribution function and the scoring function of high-frequency electromagnetic leakage signal part is obtained according to mixed Gauss model; And according to being uniformly distributed probability-distribution function and the scoring function of model acquisition low frequency electromagnetic leakage signal part.

Alternatively, described probability-distribution function and the scoring function obtaining high-frequency electromagnetic leakage signal part according to mixed Gauss model; And according to being uniformly distributed after model obtains the probability-distribution function of low frequency electromagnetic leakage signal part and the step of scoring function, also comprise:

The iteration coefficient of the equalizer corresponding to high-frequency electromagnetic leakage signal part is obtained according to the probability-distribution function of described high-frequency electromagnetic leakage signal part; And the iteration coefficient of the equalizer corresponding to low frequency electromagnetic leakage signal part is obtained according to the probability-distribution function of described low frequency electromagnetic leakage signal part.

Alternatively, described electromagnetic leakage signal recovery with the expression formula strengthening model is:

$\hat{f}(x,y)=\mathrm{\λ}\left(n\right)f_L(x,y)*w_L(x,y)+(1-\mathrm{\λ}(n\left)\right)f_H(x,y)*w_H(x,y),$

In formula, for electromagnetic leakage signal recovers and the output valve strengthening model, f _l(x, y) and f _h(x, y) is respectively high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part, weighting function when λ (n) is merging, and λ (n) span is [0,1], and sgm [x]=[1+e ^-x] ^-1for sigmoid function, for positive maximum, w _l(x, y) and w _h(x, y) is respectively the weight coefficient of high-frequency electromagnetic leakage signal part and the blind equalizer corresponding to low frequency electromagnetic leakage signal part.

Second aspect, the embodiment of the present invention additionally provides a kind of electromagnetic leakage signal and recovers and enhancing system, comprising:

Electromagnetic leakage signal model building module, the feature for environmentally noise and wireless channel noise sets up the electromagnetic leakage signal model comprising ambient noise and wireless channel noise;

Adaptive inverse filtering processing module, for being divided into high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part also to carry out Adaptive inverse filtering process respectively received electromagnetic leakage signal according to electromagnetic leakage signal model;

Recover and strengthen model acquisition module, for by high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part through respective blind equalizer to obtain respective output signal, utilize convex combination mode to merge two output signals and recover and enhancing model to obtain electromagnetic leakage signal.

Alternatively, also comprise cost function generation module, for forming the cost function of blind equalizer according to statistical independence parameter, with make the output signal of estimation model of receptions electromagnetic leakage signal and the electromagnetic leakage signal that receives close as far as possible.

Alternatively, also comprise probability-distribution function acquisition module, for obtaining probability-distribution function and the scoring function of high-frequency electromagnetic leakage signal part according to mixed Gauss model; And according to being uniformly distributed probability-distribution function and the scoring function of model acquisition low frequency electromagnetic leakage signal part.

Alternatively, also comprise iteration coefficient acquisition module, for obtaining the iteration coefficient of the equalizer corresponding to high-frequency electromagnetic leakage signal part according to the probability-distribution function of high-frequency electromagnetic leakage signal part; And the iteration coefficient of the equalizer corresponding to low frequency electromagnetic leakage signal part is obtained according to the probability-distribution function of low frequency electromagnetic leakage signal part.

Compared with prior art, the recovery that the embodiment of the present invention provides and Enhancement Method are according to the physical characteristic of electromagnetic leakage signal, can be good at recovering to make the vision signal received have stronger intelligibility with increase vision signal, thus improve the attacking ability of TEMPEST.

Accompanying drawing explanation

Can understanding the features and advantages of the present invention clearly by reference to accompanying drawing, accompanying drawing is schematic and should not be construed as and carry out any restriction to the present invention, in the accompanying drawings:

Fig. 1 is that a kind of electromagnetic leakage signal that the embodiment of the present invention provides recovers and Enhancement Method flow chart;

Fig. 2 is that a kind of electromagnetic leakage signal that the embodiment of the present invention provides recovers and strengthens system principle diagram;

Fig. 3-a ~ Fig. 3-d is electromagnetic leakage signal reprocessing comparative result schematic diagram;

Fig. 4 is the electromagnetic leakage signal mean square error curve comparison result schematic diagram received by many algorithms process.

Embodiment

Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples for illustration of the present invention, but are not used for limiting the scope of the invention.

Embodiments provide a kind of electromagnetic leakage signal to recover and Enhancement Method, as shown in Figure 1, comprising:

The feature of S100, environmentally noise and wireless channel noise sets up the electromagnetic leakage signal model comprising ambient noise and wireless channel noise;

S200, according to electromagnetic leakage signal model received electromagnetic leakage signal to be divided into high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part and to carry out Adaptive inverse filtering process respectively;

S300, by described high-frequency electromagnetic leakage signal part and described low frequency electromagnetic leakage signal part through respective blind equalizer to obtain respective output signal, utilize convex combination mode to merge two output signals and recover and enhancing model to obtain electromagnetic leakage signal.

The electromagnetic leakage signal provided the embodiment of the present invention below recovers and each step of Enhancement Method is launched to describe in detail.

First the feature introducing S100, environmentally noise and wireless channel noise sets up the step of the electromagnetic leakage signal model comprising ambient noise and wireless channel noise.

A large amount of ambient noises and wireless channel noise is there is in received electromagnetic leakage signal.Wherein ambient noise comprises the noise that the electromagnetic radiation source outside earth atmosphere causes, and this noise presents irregular change, and is also uneven in spatial distribution; Wireless channel noise can regard the noise of device interior as, mainly shot noise, be by discrete charge move form the random noise that electric current causes, belong to thermal noise.Therefore received electromagnetic leakage signal can regard the signal mixed by Convolution Noise and white Gaussian noise as, and theoretical model is such as formula shown in (1):

f(x,y)＝s(x,y)*c(x,y)+n(x,y)(1)

Wherein, s (x, y) is received electromagnetic leakage signal, and c (x, y) is the convolution function of transmission channel, system, and * represents and carries out convolution algorithm, and n (x, y) is white Gaussian noise.

Next is introduced S200, is divided into high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part also to carry out the step of Adaptive inverse filtering process respectively received electromagnetic leakage signal according to electromagnetic leakage signal model.

As shown in Figure 2, according to the feature of received electromagnetic leakage signal, adopt many sub-band processings mode, by this electromagnetic leakage signal high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part.Then respectively Adaptive inverse filtering process is carried out, shown in (2) to high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part:

f(x,y)＝s _L(x,y)*c _L(x,y)+s _H(x,y)*c _H(x,y)+n(x,y)(2)

Wherein, f _l(x, y) and f _h(x, y) is that the high-frequency electromagnetic of received electromagnetic leakage signal leaks part and low frequency electromagnetic leakage part, c respectively _h(x, y) and c _l(x, y) is respectively the transmission channel that high-frequency electromagnetic leaks part and low frequency electromagnetic leakage part, and n (x, y) is white Gaussian noise.

Finally introduce S300, by described high-frequency electromagnetic leakage signal part and described low frequency electromagnetic leakage signal part through respective blind equalizer to obtain respective output signal, utilize convex combination mode to merge two output signals and recover and enhancing model to obtain electromagnetic leakage signal.

High-frequency electromagnetic leaks part and low frequency electromagnetic leaks part respectively through respective blind equalizer w _l(x, y) and w _houtput signal after (x, y) carries out Adaptive inverse filtering process, and carries out superposing recovery and the enhancing model that reconstruct obtains received electromagnetic leakage signal, shown in (3):

$\begin{array}{c}\hat{f}(x,y)=\mathrm{\λ}\left(n\right)z_L+(1-\mathrm{\λ}(n\left)\right)z_H\\ =\mathrm{\λ}\left(n\right)f_L(x,y)*w_L(x,y)+(1-\mathrm{\λ}(n\left)\right)f_H(x,y)*w_H(x,y)\end{array}---\left(3\right)$

Wherein, z _land z _hbe respectively the output signal that high-frequency electromagnetic leaks part and the blind equalizer corresponding to low frequency electromagnetic leakage part, for the output signal recovered with strengthen model, be also the estimated value to electromagnetic leakage signal, λ (n) is the weighting function of superposition reconstruct:

Wherein, sgm [x]=[1+e ^-x] ^-1for sigmoid function, for for recovering the positive maximum with the output signal strengthening model, λ (n) span is [0,1].

For making recovery close as far as possible with the electromagnetic leakage signal received with the output signal strengthening model; Wherein said close as far as possible, refer to that output signal is less than preset value with the difference of the amplitude of corresponding electromagnetic leakage signal.Such as, optimal scene is, equal with the amplitude of electromagnetic leakage signal for its output signal of same signal.For realizing above-mentioned effect, alternatively, adopt KL divergence (Kullback-Leiblerdivergence) as cost in the embodiment of the present invention.Wherein, KL divergence is the parameter of statistical independence, represents the tolerance of the asymmetry of the difference of two probability distribution, shown in (5):

$w(x,y)=\arg \underset{w}{\min }DL\left(p_z\left(z,w\right)\left|p_s\right(s\left)\right)\right)---\left(5\right)$

Wherein, the weight coefficient that w (x, y) is blind equalizer, p _z(z, w) and p _ss () is respectively the probability density function of blind equalizer output signal z, source signal s.P _z(z, w) and p _ss () can adopt formula (6) to calculate.

$\begin{array}{c}DL\left(p_s\right(s\left)\right|p_{\hat{s}}\left(\hat{s}\right))\∫p_s\left(s\right)\log \frac{p_s\left(s\right)}{p_{\hat{s}}\left(\hat{s}\right)}ds\\ =\∫p_s\left(s\right){\mathrm{logp}}_s\left(s\right)ds-\∫p_{\hat{s}}\left(\hat{s}\right){\mathrm{logp}}_{\hat{s}}\left(\hat{s}\right)ds\\ =H\left(s\right)-\∫p_{\hat{s}}\left(\hat{s}\right){\mathrm{logp}}_{\hat{s}}\left(\hat{s}\right)ds\\ \≈H\left(s\right)-E\[{\mathrm{logp}}_{\hat{s}}\left(\hat{s}\right)\]\\ =H\left(s\right)-H\left(\hat{s}\right)\\ =(-H(\hat{s}\left)\right)-(-H(s\left)\right)\\ =(-H(z\left)\right)-(-H(\hat{s}\left)\right)\end{array}---\left(6\right)$

Can find out, cost function can represent probability density function p _z(z, w) and p _sthe Euclidean minimum range of (s) negentropy.The embodiment of the present invention defines new cost function according to formula (6):

E (x, y)=tan (H (z)-g (E [logp _z(x, y)])) in (7) formula, g () is non-linear memoryless function, and form can be expressed as:

g(·)＝atanh(·)

Then new cost function becomes:

J(x,y)＝tan(H(z)-g(E[logp _z(x,y)]))(8)

The embodiment of the present invention constructs the cost function under negentropy territory, and have chosen tan as memoryless nonlinear function, is solved the constant amount in tan by iterative manner.A and b is respectively constant, can be upgraded try to achieve by formula (9) ~ (10) iteration:

$\mathrm{\Δ}a=-\mathrm{\α}\frac{\∂J(x,y)}{\∂a}=-\mathrm{\α}E\[(g-z)\frac{g}{a}\]---\left(9\right)$

$\mathrm{\Δ}b=-\mathrm{\β}\frac{\∂J(x,y)}{\∂b}=-\mathrm{\β}E\[(g-z)(a^2-g^2)\frac{z}{a}\]---\left(10\right)$

Wherein, J (x, y) is cost function, and expectation is asked in E [] expression.

Therefore, the cost function of embodiment of the present invention neotectonics has good symmetry, can improve the performance of Adaptive inverse filtering.

In the embodiment of the present invention, the electromagnetic leakage signal received is vision signal.The high-frequency video signal part of vision signal and low-frequency video signal part represent hopping part in vision signal and integral energy respectively, therefore the distribution of hopping part and integral energy also there are differences, high-frequency video signal part by a small amount of large coefficient and a large amount of coefficients close to zero signal form, adopt mixed Gauss model to describe the probability distribution situation of this high-frequency video signal part in the embodiment of the present invention; The energy of low-frequency video signal part is then comparatively level and smooth, adopts the probability distribution situation being uniformly distributed this low-frequency video signal of model description in the embodiment of the present invention.

Alternatively, probability-distribution function and the scoring function of high-frequency electromagnetic leakage signal part is obtained according to mixed Gauss model; And according to being uniformly distributed probability-distribution function and the scoring function of model acquisition low frequency electromagnetic leakage signal part.Therefore, the probability density function of high-frequency video signal is such as formula shown in (11):

$P_H(z,{\mathrm{\θ}}_H)=\frac{(1-a)}{{\mathrm{\σ}}_1\sqrt{2\mathrm{\π}}}\exp \[\frac{-{(z-{\mathrm{\μ}}_1)}^2}{2{\mathrm{\σ}}_1^2}\]+\frac{a}{{\mathrm{\σ}}_2\sqrt{2\mathrm{\π}}}\exp \[\frac{-{(z+{\mathrm{\μ}}_2)}^2}{2{\mathrm{\σ}}_2^2}\]---\left(11\right)$

Wherein, θ _hrepresent the set of the probability statistics function distributed constant of high-frequency video signal part, z is the output signal of high-frequency video signal equalizer.

The scoring function of high-frequency video signal shown in (12):

Wherein, z _hfor the output signal of high-frequency video signal equalizer, p (z _h) be high-frequency video signal part z _hprobability-distribution function, p'(z _h) be p (z _h) derivative, tanh (z _h) be z _hhyperbolic tangent function value.

The probability density function of low-frequency video signal is such as formula shown in (13):

$p_z({\mathrm{\θ}}_L,w_L)=\left\{\begin{array}{cc}\frac{z_L}{2{\mathrm{\τ}}_L}& z_L\∈\[-{\mathrm{\τ}}_L,{\mathrm{\τ}}_L\]\\ 0,& otherwise\end{array}\right.---\left(13\right)$

The scoring function of low-frequency video signal shown in (14):

Obtaining cost function according to formula (8) is:

J(y,w)＝H(z)-g(E[logp _z(x,y)])

＝H(y)+log|detw|-g(E[logp _z(z)])(15)

The weight coefficient utilizing Natural Gradient Algorithm to obtain blind equalizer is:

$\begin{array}{c}\mathrm{\Δ}w=-\mathrm{\η}\frac{\∂J(y,w)}{\∂w}\\ =-\mathrm{\η}\[w^{-T}\left(n\right)-\frac{{\mathrm{dlogp}}_z\left(z\right)}{dz}\]g{(\·)}^{\′}\end{array}---\left(16\right)$

In formula, η is iteration step length, and g () ' is non-linear memoryless function derivative.

The iteration coefficient of the equalizer corresponding to high-frequency electromagnetic leakage signal part is obtained according to the probability-distribution function of high-frequency electromagnetic leakage signal part; And obtain the iteration coefficient of the equalizer corresponding to low frequency electromagnetic leakage signal part according to the probability-distribution function of low frequency electromagnetic leakage signal part, shown in (17) ~ (18):

$w_H(n+1)=w_H\left(n\right)+{\mathrm{\η}}_H\[I-\[z_H-\tanh \left(z_H\right)\]z_H^T\left(n\right)\]w_H\left(n\right)\·{\sec }^2\left(H\right(z_H\left)\right)---\left(17\right)$

$w_L(n+1)=w_L\left(n\right)+{\mathrm{\η}}_L\[I-\mathrm{\δ}(n\±{\mathrm{\τ}}_L)\·z_L^T\left(n\right)\]w_L\left(n\right)\·{\sec }^2\left(H\right(z_L\left)\right)---\left(18\right)$

By formula, (17 ~ (18) substitute into formula (3) can obtain the recovery of electromagnetic leakage signal and the output signal strengthening model.

The recovery provided for the embodiment embodiment of the present invention and the superiority of Enhancement Method, the recovery provided below in conjunction with Fig. 3, Fig. 4 and the table 1 pair embodiment of the present invention and the effect of Enhancement Method are described.

In the embodiment of the present invention, adopt MISO-NEF method that classical image processing method NAS-RIF algorithm in prior art (NAS-RIF algorithm is the recursion filter of nonnegativity and support region constraint) and SISO-NEF algorithm (single-input single-output image equalization algorithms of the prior art) and the embodiment of the present invention provide (in the embodiment of the present invention, due to received electromagnetic leakage signal high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part respectively, therefore can regard as multichannel difference processing method).Fig. 3-a is received electromagnetic leakage signal; Fig. 3-b adopts NAS-RIF algorithm to process rear result schematic diagram to received electromagnetic leakage signal; Fig. 3-c adopts SISO-NEF algorithm to process rear result schematic diagram to received electromagnetic leakage signal; Fig. 3-d is result schematic diagram after the MISO-NEF method process adopting the embodiment of the present invention to provide.

As can be seen from Table 1, the Y-PSNR of the electromagnetic leakage signal adopted received by NAS-RIF algorithm process is 15.23; The Y-PSNR adopting the electromagnetic leakage signal received by SISO-NEF algorithm process is 15.74; The Y-PSNR of the electromagnetic leakage signal received by the process of MISO-NEF method adopting the embodiment of the present invention to provide is 16.28.

Table 1 Y-PSNR

Method	NAS-RIF	SISO-NEF	MISO-NET
				PNSR	15.23	15.74	16.28

Can find out, the recovery that the embodiment of the present invention provides and the process picture quality that Enhancement Method obtains obviously are better than other two kinds of methods of the prior art, thus illustrate that this method has good effect.

Fig. 4 is the electromagnetic leakage signal mean square error curve comparison result schematic diagram received by many algorithms process.The convergence rate of the recovery that provides of the embodiment of the present invention and Enhancement Method is faster than other two algorithms as seen from Figure 4, and steady-state error is little, superior performance.

The recovery that the embodiment of the present invention provides and Enhancement Method comprise noise situations according to received electromagnetic leakage signal, and ambient noise and wireless channel noise are set up electromagnetic leakage signal model; And be divided into high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part according to the feature of received electromagnetic leakage signal, and by both by carrying out superposition reconstruct after respective blind equalizer, obtain electromagnetic leakage signal and recover and strengthen model.Compared with prior art, the recovery that the embodiment of the present invention provides and Enhancement Method are according to the physical characteristic of electromagnetic leakage signal, can be good at recovering to make the vision signal received have stronger intelligibility with increase vision signal, thus improve the attacking ability of TEMPEST.

Embodiment two

Second aspect, a kind of electromagnetic leakage signal provided for embodying the embodiment of the present invention recovers the superiority with Enhancement Method, and the embodiment of the present invention additionally provides a kind of electromagnetic leakage signal and recovers and enhancing system, comprising:

Electromagnetic leakage signal model building module, for according to comprising ambient noise in received electromagnetic leakage signal and wireless channel noise sets up electromagnetic leakage signal model;

Adaptive inverse filtering processing module, also carries out Adaptive inverse filtering process respectively for received electromagnetic leakage signal being divided into high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part;

Recover and strengthen model acquisition module, recover for the electromagnetic leakage signal that high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part are carried out merging to obtain received electromagnetic leakage signal through the output signal of respective blind equalizer and strengthen model.

Alternatively, also comprise cost function generation module, for forming the cost function of blind equalizer according to statistical independence parameter, with make the output signal of estimation model of receptions electromagnetic leakage signal and the electromagnetic leakage signal that receives close as far as possible.

Alternatively, also comprise probability-distribution function acquisition module, for obtaining probability-distribution function and the scoring function of high-frequency electromagnetic leakage signal part according to mixed Gauss model; And according to being uniformly distributed probability-distribution function and the scoring function of model acquisition low frequency electromagnetic leakage signal part.

Alternatively, also comprise iteration coefficient acquisition module, for obtaining the iteration coefficient of the equalizer corresponding to high-frequency electromagnetic leakage signal part according to the probability-distribution function of high-frequency electromagnetic leakage signal part; And the iteration coefficient of the equalizer corresponding to low frequency electromagnetic leakage signal part is obtained according to the probability-distribution function of low frequency electromagnetic leakage signal part.

The up near field communication system that the embodiment of the present invention provides, with up near field communication method mentioned above based on same goal of the invention, solve same technical problem, reach same technique effect, the embodiment of the present invention repeats no longer one by one.

In sum, the electromagnetic leakage signal that the embodiment of the present invention provides recovers to comprise noise situations with Enhancement Method and system according to received electromagnetic leakage signal, and ambient noise and wireless channel noise are set up electromagnetic leakage signal model; And be divided into high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part according to the feature of received electromagnetic leakage signal, and by both by carrying out superposition reconstruct after respective blind equalizer, obtain electromagnetic leakage signal and recover and strengthen model.Compared with prior art, the recovery that the embodiment of the present invention provides and Enhancement Method are according to the physical characteristic of electromagnetic leakage signal, can be good at recovering to make the vision signal received have stronger intelligibility with increase vision signal, thus improve the attacking ability of TEMPEST.

Although describe embodiments of the present invention by reference to the accompanying drawings, but those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, such amendment and modification all fall into by within claims limited range.

Claims (9)

1. electromagnetic leakage signal recovers and an Enhancement Method, it is characterized in that, comprising:

Environmentally the feature of noise and wireless channel noise sets up the electromagnetic leakage signal model comprising ambient noise and wireless channel noise;

High-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part is divided into also to carry out Adaptive inverse filtering process respectively received electromagnetic leakage signal according to described electromagnetic leakage signal model;

By described high-frequency electromagnetic leakage signal part and described low frequency electromagnetic leakage signal part through respective blind equalizer to obtain respective output signal, utilize convex combination mode to merge two output signals and recover and enhancing model to obtain electromagnetic leakage signal.

2. electromagnetic leakage signal according to claim 1 recovers and Enhancement Method, it is characterized in that, the feature of described environmentally noise and wireless channel noise also comprises after setting up the step of the electromagnetic leakage signal model comprising ambient noise and wireless channel noise:

Form the cost function of blind equalizer according to statistical independence parameter, with make the output signal of estimation model of receptions electromagnetic leakage signal and the electromagnetic leakage signal that receives close as far as possible.

3. electromagnetic leakage signal according to claim 2 recovers and Enhancement Method, it is characterized in that, the described cost function forming blind equalizer according to statistical independence parameter, with make after the output signal of estimation model of the reception electromagnetic leakage signal step close as far as possible with the electromagnetic leakage signal that receives, also comprise:

Probability-distribution function and the scoring function of high-frequency electromagnetic leakage signal part is obtained according to mixed Gauss model; And according to being uniformly distributed probability-distribution function and the scoring function of model acquisition low frequency electromagnetic leakage signal part.

4. electromagnetic leakage signal according to claim 3 recovers and Enhancement Method, it is characterized in that, described probability-distribution function and the scoring function obtaining high-frequency electromagnetic leakage signal part according to mixed Gauss model; And according to being uniformly distributed after model obtains the probability-distribution function of low frequency electromagnetic leakage signal part and the step of scoring function, also comprise:

The iteration coefficient of the equalizer corresponding to high-frequency electromagnetic leakage signal part is obtained according to the probability-distribution function of described high-frequency electromagnetic leakage signal part; And the iteration coefficient of the equalizer corresponding to low frequency electromagnetic leakage signal part is obtained according to the probability-distribution function of described low frequency electromagnetic leakage signal part.

5. electromagnetic leakage signal according to claim 1 recovers and Enhancement Method, it is characterized in that, described electromagnetic leakage signal recovers and the expression formula of enhancing model is:

$\hat{f}(x,y)=\mathrm{\λ}\left(n\right)f_L(x,y)*w_L(x,y)+(1-\mathrm{\λ}(n\left)\right)f_H(x,y)*w_H(x,y),$

In formula, for electromagnetic leakage signal recovers and the output valve strengthening model, f _l(x, y) and f _h(x, y) is respectively high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part, weighting function when λ (n) is merging, and λ (n) span is [0,1], and sgm [x]=[1+e ^-x] ^-1for sigmoid function, for positive maximum, w _l(x, y) and w _h(x, y) is respectively the weight coefficient of high-frequency electromagnetic leakage signal part and the blind equalizer corresponding to low frequency electromagnetic leakage signal part.

6. electromagnetic leakage signal recovers and strengthens a system, it is characterized in that, comprising:

Electromagnetic leakage signal model building module, the feature for environmentally noise and wireless channel noise sets up the electromagnetic leakage signal model comprising ambient noise and wireless channel noise;

Adaptive inverse filtering processing module, for being divided into high-frequency electromagnetic leakage signal part and low frequency electromagnetic leakage signal part also to carry out Adaptive inverse filtering process respectively received electromagnetic leakage signal according to described electromagnetic leakage signal model;

Recover and strengthen model acquisition module, for by described high-frequency electromagnetic leakage signal part and described low frequency electromagnetic leakage signal part through respective blind equalizer to obtain respective output signal, utilize convex combination mode to merge two output signals and recover and enhancing model to obtain electromagnetic leakage signal.

7. electromagnetic leakage signal according to claim 6 recovers and strengthens system, it is characterized in that, also comprise cost function generation module, for forming the cost function of blind equalizer according to statistical independence parameter, with make the output signal of estimation model of receptions electromagnetic leakage signal and the electromagnetic leakage signal that receives close as far as possible.

8. electromagnetic leakage signal according to claim 7 recovers and strengthens system, it is characterized in that, also comprises probability-distribution function acquisition module, for obtaining probability-distribution function and the scoring function of high-frequency electromagnetic leakage signal part according to mixed Gauss model; And according to being uniformly distributed probability-distribution function and the scoring function of model acquisition low frequency electromagnetic leakage signal part.

9. electromagnetic leakage signal according to claim 8 recovers and strengthens system, it is characterized in that, also comprise iteration coefficient acquisition module, for obtaining the iteration coefficient of the equalizer corresponding to high-frequency electromagnetic leakage signal part according to the probability-distribution function of described high-frequency electromagnetic leakage signal part; And the iteration coefficient of the equalizer corresponding to low frequency electromagnetic leakage signal part is obtained according to the probability-distribution function of described low frequency electromagnetic leakage signal part.