CN112564778A - Tag selection mechanism for optimizing probability of privacy interruption in backscatter communication system - Google Patents

Abstract

The invention provides a label selection method with optimal security interruption probability in a backscattering communication system, which is characterized in that S, N labels T are generated by a signal source specially used for generating continuous waves_iA destination node D and an eavesdropper E. The invention takes into account the label T_iDue to the circuit consumption of the tag, the optimal dynamic reflection coefficient meeting the lowest energy consumption of the tag is designed, and the power of a signal reflected by each tag can be maximized. The energy collected by the circuit is ensured to meet the requirement of the label T_iIn each transmission, the invention selects a tag T that maximizes the system's secret capacity_iTo transmit its own information to the destination node. In order to evaluate the privacy interruption probability of the label selection scheme in the system, the invention derives the privacy interruption probability and the diversity order of the system in an independent non-uniformly distributed Rayleigh fading channel. Compared with other schemes, the optimal label selection method provided by the invention has the best performance.

Description

Tag selection mechanism for optimizing probability of privacy interruption in backscatter communication system

Technical Field

The invention relates to the technical field of wireless communication, in particular to a label selection method for minimizing the security interruption probability of a system in a multi-label environment.

Technical Field

Backscatter communication can modulate and reflect radio frequency signals in the surrounding environment to complete the communication. Compared with the traditional active communication technology which generates radio frequency signals by itself, the backscattering equipment does not need to generate the radio frequency signals, so high-power-consumption elements such as an analog/digital converter, a digital/analog converter, an oscillator and the like are omitted, and the passive communication technology belongs to. For this reason, backscatter devices have lower power consumption, on the order of milliwatts. Based on the framework of backscatter communication systems, it can be divided into three categories: single station, dual station and ambient backscatter communication systems. In a two-station backscatter communication system, there is a dedicated radio frequency signal source that generates a continuous sine wave.

Under the coexistence of conventional link and Backscatter communication systems, the communication probabilities in the documents (Y.Ye, L.Shi, X.Chu and G.Lu, "On the output Performance of the electromagnetic backscattering communication," IEEE Internet of Things J., vol.7, No.8, pp.7265-7278, Aug.2020; H.Ding, D.B.da Costa, and J.Ge, "output Analysis for cooperative electromagnetic backscattering system," IEEE Wireless Commun.Lett., vol.9, No.5, NO. 601-605, May 2020; D.Darsena, G.Gelli and F.Verde, "Modeling and Analysis of Wireless Networks, Audio, App.7, App.174, and IEEE transaction Analysis of communication systems, respectively) are analyzed. In the above-described operation, the performance of the conventional link will deteriorate due to co-channel interference introduced by the backscatter devices. However, in the literature (D.Li. "Two bits With One Stone: expanding decoding-and-Forward zooming for operating non-coherent amplitude Backscattering," IEEE Trans.Commun., vol.68, No.3, pp.1405-1416, Mar.2020; Y.Liu, Y.Ye, G.Yan and Y.Zo. "outer Performance Analysis for an operating non-coherent Source selected Two-way coherent amplitude Backscattering Communication Systems," IEEE Commun.Lett., doi:10.1109/LCOMM.2020.3029246), a Cooperative Backscattering Communication system is proposed in which a relay node tag is embedded as a conventional Backscattering device, which uses the power generated by the user to achieve Backscattering and does not affect any link Performance.

Unlike the above-mentioned documents, which consider only Backscatter Communication Systems With only one Tag, authors in the document (d.li, w.pen and f.hu, "Capacity of Backscatter Communication Systems With Tag Selection," IEEE trans.veh.technol., vol.68, No.10, pp.10311-10314, oct.2019;) propose Backscatter Communication Systems With multiple tags, which can implement Tag Selection to maximize traversal Capacity. In the above document, all tags use a fixed reflection coefficient, and the energy collected by each tag satisfies the power consumption of the tag itself, so that all tags have the ability to reflect signals. In practice, a fixed reflection coefficient may not ensure that the energy that has been harvested can meet the power requirements of each tag, even if the power consumed is very small (a few milliwatts). Similar to conventional wireless systems, in backscatter communication systems, due to the broadcast nature, the signal reflected by the tag may be received by a malicious eavesdropper, thus requiring privacy and security concerns. In the literature (y.zhang, f.gao, l.fan, x.lei and g.k.karaginanidis, "Secure Communications for Multi-Tag Backscatter Systems," IEEE Wireless communications.let., vol.8, No.4, pp.1146-1149, aug.2019), the optimal Tag selected in a Backscatter communication system with multiple tags can minimize the probability of a privacy outage and the corresponding probability of a system outage is deduced. We note that the above document does not take into account the power consumption problem of the tag itself. Both of these documents are based on independent identically distributed rayleigh channels. Considering that multiple tags are randomly distributed in a practical backscatter communication system, it is appropriate to describe the channel with independent non-uniform distributions. In consideration of the current research situation, the invention provides an optimal label selection method in a backscattering communication system on the basis of considering the self power consumption of a label so as to minimize the confidentiality interruption probability of the system.

Disclosure of Invention

The invention provides a label selection mechanism with optimal security interruption probability in a backscattering communication system, which consists of S, N labels of a source special for generating continuous waves, a destination node D and an eavesdropper E. The invention designs the optimal dynamic reflection coefficient to maximize the power of the reflected signal on the premise of meeting the self energy consumption of the tag. Further, in each transmission, all canSelecting a label T capable of maximizing the system secret capacity from labels with enough energy collected to meet self energy consumption_iTo transmit its own information to the destination node D. To evaluate the privacy outage probability of a signature selection scheme in the system, we derive a closed-form solution to the privacy outage probability based on independent, unsynchronized rayleigh channels. The tag selection scheme we propose has the best performance compared to the existing methods. (hereinafter "tag", in the Internet of things, means a radio device with backscatter communications capability, English tag)

In order to achieve the above object, the present invention provides the following technical solutions: a tag selection method with optimal privacy interruption probability in a backscatter communication system, comprising the steps of:

1. all channels are assumed to obey independent non-uniformly distributed rayleigh fading channels. The direct link interference from the source S to the destination node D and the source S to the eavesdropper E can be removed by serial interference cancellation techniques. Label T_iThe received signal can be written as

Wherein P is_sIs the power transmitted through S, and st is the signal with unit energy transmitted through S. It is noted that the noise power is introduced through the tag antenna, and is negligible small compared to the rf signal power, and is not listed in the above equation.

2. The received signal may be divided into two parts:

the circuitry used to drive the tag is such that,

used as a reflected signal for loading the tag's own information, wherein_iPresentation label T_iNext, we will discuss in detail the dynamic reflection coefficient: in the label T_iTo the collected energy

Can be expressed as follows:

wherein

Is shown in label T_iThe input power for energy harvesting, namely:

P_maxindicating the saturation power when the input power is very large. v. of₀Representing the sensitivity of the energy harvesting circuit; v. of₁And v₂Is a fixed parameter determined by the resistance, capacitance and diode switching voltage of the energy harvesting circuit. Let P_cPresentation label T_iThe power consumed by the circuit. If it is not

Label T_iThe ability to modulate and reflect radio frequency signals. In order to maximize the power of the reflected signal, the

And take into account beta_iThe fact of ≦ 1, the optimal dynamic reflection coefficient can be expressed as follows:

wherein the content of the first and second substances,

remarks 1: by observing

Can be found as long as

Label T_iCan reflect the signal; and is

Meaning the label T_iInsufficient energy is collected to drive the circuit and, correspondingly, there is no ability to reflect the signal.

3. Based on the above analysis, tag T_iThe effective snr of the signal reflected to the destination node D can be expressed as

Where η represents the reflection efficiency, meaning the label T_iThe residual energy of the signal which can be effectively reflected is consumed by the self circuit. Γ represents the performance gap from shannon capacity caused by the modulation method employed for backscatter communications. Likewise, by tag T_iThe effective snr of the signal reflected to an eavesdropper can be written as

Secret capacity of system C_sDefined as the instantaneous achievable rate C at the destination node D_dAnd instantaneous reachable rate C at eavesdropper E_eCan be expressed as

If the instantaneous secret capacity is lower than the preset target speed R_sI.e. C_s＜R_sThe system will be interrupted.

Compared with the prior art, the invention has the following advantages:

1. the invention considers the self power consumption of the label as the backscattering equipment, and adopts the optimal dynamic reflection coefficient to maximize the power of the reflected signal on the premise of meeting the self power consumption of the label.

2. In each transmission, the label which can maximize the security capacity of the system is selected from all labels which can collect enough energy to drive the circuit of the label to send the information of the label to the destination node, so that the security interrupt capacity of the system is minimized. Compared with the existing method, the method proposed by the inventor has the best performance of secrecy interruption.

Drawings

FIG. 1, System model

FIG. 2 comparison of proposed scheme and random scheme for probability of privacy disruption

Figure 3, the proposed scheme and the scheme using independent unsynchronized rayleigh fading channel to compare the probability of privacy interruption

FIG. 4 comparison of the probability of a privacy disruption for different schemes at different privacy data rates

Detailed Description

In this embodiment, the specific parameters are set as follows:

the backscattering system of the present invention consists of S, N tags T from a single dedicated signal source emitting Continuous Wave (CW)_i(i ═ 1,2, …, N), a destination node D, and an eavesdropper EE. T is_iThe energy of the radio-frequency signal of the S is collected to drive a circuit to work, the incident wave is modulated and reflected to send self information to the D, and meanwhile the T can be eavesdropped by the E_iThe reflected electromagnetic wave. h is_i、g_i、g_0iRespectively representing links S → T_i、T_i→D、T_iChannel coefficient of → E, | h_i|²、|g_i|²、|g_0i|²Respectively representing links S → T_i、T_i→D、T_iChannel gain of → E. S, D, E are (0,0) m, (8,0) m, and (9,2) m, respectively. The coordinates of the label are (d)_1i0) m, and d_1iIn the interval [3,5 ]]m are uniformly distributed. Assuming that all channels are subject to independent non-identically distributed Rayleigh fading, the corresponding channel gain | h_i|²、|g_i|²、|g_0i|²Subject to independent non-uniform distribution of the index distributions, the mean values being respectively

Wherein: f denotes the frequency of the rf signal transmitted by S and let f 915 MHz; g_s、G_D、G_T、G_ES, D, tag and eavesdropper antenna gains and G_s＝G_D＝G_T＝G_E＝6dBi；d_2i＝8-d_1i，

α ═ 2 represents the path attenuation factor. Power consumption P of tag_c8.9 μ W. The parameters of the nonlinear energy harvesting circuit are set as follows: p_max＝240μW、v₀＝5μW、v₁＝5000、v₂0.0002. The reflection efficiency is set to 0.6 and the white gaussian noise power at the destination node is set to σ²The performance difference is set to Γ 5dB at-60 dBm.

As shown in FIG. 1, the proposed backscatter communications system consists of a dedicated source S S generating a continuous sine wave, a destination node D, an eavesdropper E and N tags T_iAnd (4) forming. All nodes are equipped with a single antenna. From S → T_i、T_i→D、T_iThe channel coefficients → E are respectively denoted by h_i、g_i、g_0i. In FIG. 1, i^*Indicating the index of the selected tab. All channels are assumed to obey independent non-uniformly distributed Rayleigh fading channels, and the corresponding channel gains obey independent non-uniformly distributed exponential distributions with the mean value of lambda_1i、λ_2iAnd λ_0i. In the prior art, it is assumed that the direct link interference of S → D and S → E can be completely eliminated with the help of the successive interference cancellation technique. In the label T_iThe received signal can be expressed as:

wherein P is_sIs the power sent over S and S t is the signal with unit energy transmitted over S. It can be pointed out that the noise power is introduced through the tag antenna, and compared with the radio frequency signal power, the power is very small and negligible, so that the expression (1) is not shown. Received signalCan be divided into two parts:

the circuit used to drive the tag itself,

for use as a reflected signal loaded with tag information, wherein_iPresentation label T_iNext, we will discuss in detail the dynamic reflection coefficient: in the label T_iTo the collected energy

Can be expressed as follows:

wherein

Is shown in label T_iThe input power for energy harvesting, namely:

。P_maxindicating the saturation power when the input power is very large. v. of₀Representing the sensitivity of the energy harvesting circuit; v. of₁And v₂Is a fixed parameter determined by the resistance, capacitance and diode switching voltage of the energy harvesting circuit. Let P_cPresentation label T_iThe power consumed by the circuit. If it is not

Label T_iCapable of reflecting the radio frequency signal to the destination node D. In order to maximize the power of the reflected signal, the

And take into account beta_iThe fact of ≦ 1, the optimal dynamic reflection coefficient can be expressed as follows:

wherein the content of the first and second substances,

(the derivation of equation (3) can be found in Ye, Y., Shi, L., Chu, X., et al. 'On the outer performance of the active backscatter communications', IEEE Internet Things J.,2020,7, (8), pp.7265-7278)

Remarks 1: by observing

Can be found as long as

Label T_iCan reflect the signal; and is

Meaning the label T_iInsufficient energy is collected to drive the circuit and, correspondingly, there is no ability to reflect the signal.

Based on the above analysis, the destination node D receives the label T_iThe effective snr of the reflected signal can be expressed as

Where η is the backscattering efficiency, meaning the label T_iThe effective signal can be reflected. Γ represents the performance gap from shannon capacity caused by the modulation method employed by backscatter communications. Likewise, by tag T_iThe effective snr of the signal reflected to an eavesdropper can be written as

Secret capacity of system C_sDefined as the instantaneous achievable rate C at the destination node D D_dAnd instantaneous reachable rate C at eavesdropper E E_eCan be expressed as

If the instantaneous secret capacity is lower than the preset target speed R_sI.e. C_s＜R_sThe system will be interrupted.

Tag selection scheme and privacy disruption probability analysis

In the proposed system, it is assumed that tags that are able to satisfy their own power consumption and to maximize the privacy capacity are selected to backscatter the signal, i.e. to satisfy:

based on (7), we can select the optimal tag T by the following two steps_i. First, by judging

Finding all the labels capable of collecting enough energy to drive the circuit of the label forming set C, namely T_iE.g. C. Second, in set C, the label that is found to maximize the privacy capacity is selected as the optimal label. With the optimal tag, the corresponding minimized privacy disruption probability can be deduced. In order to derive a closed-form solution of the secret interruption probability of a system in an independent non-uniformly distributed Rayleigh fading channel, a two-step optimal label selection scheme is converted into an equivalent scheme as follows: first, calculate the arbitrary link S → T_i→D_,E probability of privacy interruption when selected

Then all channels are mutually independentThe standing assumption is that the privacy disruption probability of the system is written as:

remarks 2: the application of our proposed tag selection scheme requires obtaining global channel state information, i.e. in each transmission slot, the channel gain | h_i|²、|g_i|²、|g_0i|². The information may be obtained by a probe signal before the formal start of each transmission slot.

Arbitrary link S → T_iThis interruption of link privacy capacity when → D, E is selected will occur in two scenarios: case 1: in the label T_iWhen the collected energy can not drive the circuit, namely:

case 2: label T_iThe harvested energy meets its power consumption requirements, but the privacy capacity at the destination node is below a threshold R_sNamely:

in the above-described analysis, the analysis was carried out,

can be calculated as:

further, P₁The derivation can be as follows:

P₂the derivation can be as follows:

the threshold signal-to-noise ratio can be expressed as in (12)

In order to easily derive the analytical expression, the variable substitution X ═ h is adopted in formula (13)_i|²，Y＝|g_i|²，Z＝|g_0i|²The corresponding probability density function is

And

equation (14) replacement with variables

The second double integral in equation (15) can be obtained. Use of integration formula for integration in formula (16)

(wherein K is₁Bessel function of the second class of first order modifications), we can obtain the final closed analytic expression. In the above-described derivation,

representing the transmitted signal-to-noise ratio with a performance gap.

In conjunction with equations (10) and (17), the result of the closed resolution of the arbitrary link privacy outage probability can be written as follows:

using equations (8) and (18), the probability of a privacy disruption for the proposed tag selection method can be found as follows:

at high S/N ratio, utilize

And

the high frequency signal-to-noise ratio approximation of (c) can be written as follows:

in the formula (20), in the following formula,

representing the transmission signal-to-noise ratio of the source S. Thus, P_outThe high frequency signal-to-noise ratio is approximately expressed as

Remarks 3: looking at equation 20, at high snr, the probability of a privacy break for any link is determined only by the mean of two channels, i.e., link T_iLambda of → D_2iLink T_iLambda of → E_0iAnd threshold signal-to-noise ratio τ, but with link S → T_iIt has no relation. This can be explained as follows: at high signal-to-noise ratio, each tag T_iThe probability of harvesting enough energy to drive the circuit is so high that system performance is also determined by the other links. In addition, we can see that at high snr, the probability of any link and system privacy disruption will remain constant and not decrease as the input snr increases. In other words, IThe proposed diversity order of the system is zero, i.e.

This is because in the high signal-to-noise ratio region, the increase of the system privacy interruption probability is limited due to the existence of the eavesdropper. From equation (21), we can find that as the number of tags N increases, the performance of the system security interruption probability will be improved because

Remarks 4: we propose that the superiority of the tag selection scheme can be understood from two aspects. First, the solution designs an optimal dynamic reflection coefficient. On the premise of meeting the consumption of the tag circuit, the design can ensure the tag T_iReflects as much signal power as possible and can maximize the received signal-to-noise ratio of the destination node D. Second, the tag that maximizes the security capacity of the system is selected to transmit its own information per transmission, thereby minimizing the security disruption probability of the system.

As shown in FIG. 2, the analytic result can be completely fitted with the simulation result in the whole SNR region, when P is_sThe privacy outage probability will not change any more > 40dBm, consistent with the analysis of the diversity performance of the system in note 3. In addition, as the number of tags N increases, the system privacy interruption performance does improve. In addition, the performance of the random selection scheme is also shown in fig. 2. We can see that the performance of the random selection scheme is always inferior to the scheme we propose, and in the random selection scheme, as the number of tags increases, the system performance is difficult to improve. (random selection scheme: in each transmission, a tag is randomly selected to send information, and the secrecy interruption probability of the system at the moment is considered)

Fig. 3 shows the performance of the system in independent non-co-distributed and independent co-distributed rayleigh channels in comparison. In the independent co-distribution scheme, d₁Representing the source S to all tags T_iA distance of, i.e. d_1i＝d₁And the joint channel mean can be written as

And

wherein d is₂＝8-d₁And

the secret interruption probability analysis result for the independent same distribution scheme can also be obtained by using the formulas (8) and (18), as long as the lambda is obtained_0i、λ_1iAnd λ_2iBy substitution of λ₀、λ₁And λ₂. As shown in fig. 3, in an actual communication environment, if the channels are subject to independent non-co-distributed channels, errors will be introduced by using independent co-distributed channels.

Given P_sFig. 4 studies the comparison of performance at different privacy rates at 30dBm and N5. In FIG. 4, to ensure contrast fairness, we set d for all schemes_1i4. In the existing solution simulation, we first select the optimal tag according to the method in the prior art and then study whether the energy collected by the selected tag can satisfy the self energy consumption P_c. If the harvested energy is not sufficient for its own power consumption, the system will be interrupted.

In fig. 4, the analytical expressions and the full fit of the simulation curves in our proposed label selection scheme again confirm the correctness of the theoretical analysis. Further, fig. 4 shows that the performance of the other schemes is worse. With the prior art solutions, the design of the fixed reflection coefficient does not ensure that the tag can maximize the power of the reflected signal in each transmission. Furthermore, the selected optimal tag may not be able to collect enough energy to drive the tag circuit, which would degrade privacy interruption performance. For the two reasons mentioned above, the privacy interruption probability of the scheme existing in the prior art is inferior to that of the scheme. By observing the formula equation (6) in

Middle to first order of betaAfter derivation can be obtained

Wherein

And

from a statistical point of view, we can find

This is due to λ_0i＜λ_2iWherein

Represents the desired operator. Thus, the prior art scheme, for β, the privacy capacity is a decreasing function. The smaller β is, the larger the security capacity is, and the smaller the security interruption capacity is. However, the smaller β, the greater the probability that the energy collected by the tag will not meet its own power consumption. The two reasons mentioned above lead to the performance curves of the prior art scheme shown in fig. 4.

Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, improvement and the like made within the content and principle of the present invention shall be included in the protection scope of the present invention.

Claims (5)

1. A tag selection mechanism for optimizing the probability of a privacy disruption in a backscatter communication system,

the method comprises the following steps:

(1) constructing a system model: the system is composed of S, N tags T which are specially used for generating continuous waves_iA destination node D and an eavesdropper E;

(2) deducing the tag T_iThe optimal dynamic reflection coefficient is used for maximizing the power of the reflected signal;

(3) secret capacity C of the proposed system_sDefined as the instantaneous achievable rate C at the destination node_dInstantaneous reachable rate C with eavesdropper_gThe difference between the two; the instantaneous secret capacity is compared with a preset target speed rate R_sAnd comparing to finally determine the system security interruption condition.

2. The tag selection mechanism for optimizing probability of outage in a backscatter communication system according to claim 1, wherein step (1) is implemented in that all nodes are equipped with a single antenna; all channels are subject to independent non-uniformly distributed rayleigh fading channels.

3. The tag selection mechanism for optimizing the probability of outage in a backscatter communication system according to claim 1, wherein the specific method of the following step (1) is:

all channels are assumed to obey independent non-uniformly distributed rayleigh fading channels; direct link interference from the information source S to the destination node D and from the information source S to the eavesdropper E can be eliminated with the help of a serial interference deletion technology; label T_iThe received signal can be written as

Wherein P is_sIs the power transmitted by S, st is the signal with unit energy transmitted by S; it is to be noted here that: the power of the noise introduced by the tag antenna is small and negligible compared to the power of the received radio frequency signal.

4. The tag selection mechanism for optimizing the probability of outage in a backscatter communication system according to claim 1, wherein the specific method of the following step (2) is:

the received signal may be divided into two parts:

the circuitry used to drive the tag is such that,

used as a reflected signal for loading the tag's own information, wherein_iPresentation label T_iDynamic reflection coefficient of, next, at the tag T_iTo the collected energy

Can be expressed as follows:

wherein

Presentation label T_iThe input power of the energy harvester of (1), namely:

P_maxrepresents the saturation power when the input power is very large; v. of₀Indicating a sensitivity threshold; v. of₁And v₂Is a fixed parameter determined by the resistor, capacitor and diode switching voltage; let P_cPresentation label T_iThe power consumed by the circuit; if it is not

Label T_iThe signal of the information source S can be reflected to a destination node D; in order to maximize the power of the reflected signal, the

And take into account beta_iThe fact of ≦ 1, the optimal dynamic reflection coefficient can be expressed as follows:

wherein the content of the first and second substances,

5. the tag selection mechanism for optimizing the probability of outage in a backscatter communication system according to claim 1, wherein the specific method of the following step (3) is:

by the label T at the destination node D_iThe effective snr of the reflected signal can be expressed as

Where η is the backscattering efficiency, meaning the label T_iThe effective signal can be reflected; Γ represents the performance gap from shannon capacity caused by the modulation method employed by the backscatter communication; likewise, the effective SNR of a signal reflected by a tag to an eavesdropper can be written as

Secret capacity of system C_sDefined as the instantaneous achievable rate C at the destination node D_dWith instantaneous reachable rate C at eavesdropper E_gThe difference, which can be expressed as

If the instantaneous secret capacity is lower than the preset target speed R_sI.e. C_s＜R_sThe system will be interrupted.

Images