A backscatter communication system

Abstract

The application discloses a backscattering communication system, channel estimation error is modeled as a Gaussian random variable independent of channel coefficient estimation value, on the basis of which, a signal source selects a tag capable of maximizing the signal-to-noise ratio of a destination node to transmit each time. In order to truly describe the performance of an actual system, the energy consumption of the tag itself is also taken into account, that is, only the tag capable of driving the circuit of the tag itself by collected energy can send information. The application analyzes the outage probability of the system when the channel estimation error exists, and simulation results show that the channel estimation error can significantly affect the performance of the tag selection scheme; the energy consumption of the tag itself has a significant effect on the outage performance of the system; if the energy consumption of the tag itself is not considered, the obtained result has a great deviation compared with the actual result.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, specifically: in backscatter communication, when considering the impact of channel estimation error, the channel estimation error is modeled as a Gaussian random variable independent of the channel estimation value. Based on this, the interruption probability of the tag selection strategy that maximizes the signal-to-noise ratio is derived, and the impact of channel estimation error and tag energy consumption on the performance of the tag selection mechanism is analyzed. Background Technology

[0002] Backscatter communication enables low-speed communication by modulating and reflecting radio frequency signals from the surrounding space, boasting extremely low power consumption down to the microwatt level, and holds broad application prospects in the future Internet of Things (IoT). Backscatter communication has received significant attention from both industry and academia in recent years.

[0003] In backscatter communication, the backscatter communication device does not have a signal amplification function, and the reflected signal will undergo dual channel fading. In order to improve the signal-to-noise ratio of the destination node, the relay selection method in cooperative communication can be used as a reference. In the multi-tag backscatter communication system, the appropriate tag is selected to send the signal in the current time slot, thereby improving the system performance. For example, references [1] and [2].

[0004] [1] D.Li, W.Peng, andF.Hu, "Capacity of backscatter communication systems with tag selection," IEEETrans.Veh.Technol., vol.68, no.10, pp.10311–10314, Oct.2019.

[0005] [2] Y. Liu, Y. Ye, and R. Q. Hu, "Secrecyoutage probability in backscatter communication systems with tag selection," IEEE Wireless Commun. Lett., vol. 10, no. 10, pp. 2190–2194, Oct. 2021.

[0006] In backscatter communication, most existing studies have adopted the assumption of perfect channel state information (CSI). In fact, due to the limitations of the channel estimation methods themselves, there will always be channel estimation error (CEE). In existing studies that consider imperfect CSI (references [3], [4], [5], [6]), the impact of tag energy consumption on system performance has not been taken into account.

[0007] [3]

[0008] [4] D.Deng,

[0009] [5]TSMuratkar, A.Bhurane, PKSharma, andA.Kothari, "Analysis of multi-tagambientbackscattercommunicationundertime-selectivefading," IEEECommun.Lett., vol.26, no.1, pp.40–43, Jan.2022.

[0010] [6] TS Muratkar, A.Bhurane, PK Sharma et al., "Physical layer security analysis in ambient backscatter communication with node mobility and imperfect channel estimation," IEEE Commun. Lett., vol. 26, no. 1, pp. 27–30, Jan. 2022.

[0011] Although the tag itself consumes only microwatts, it has a significant impact on system performance. This invention considers both channel estimation error and tag power consumption on the outage performance of the tag selection scheme, and derives the corresponding outage probability and diversity gain expressions. Summary of the Invention

[0012] To address the aforementioned technical problems, this invention provides a method for evaluating the impact of channel estimation error on the performance of tag selection schemes. The channel estimation error is modeled as a Gaussian random variable independent of the estimated channel coefficients. Based on this, the signal source selects the tag that maximizes the signal-to-noise ratio of the destination node for transmission each time.

[0013] Only tags that have collected enough energy to power their own circuitry can send information.

[0014] A backscatter communication system consists of one source node (S) and N tags T with backscattering capabilities. i (i = 1, 2, ..., N), consisting of a destination node (D, DestinationNode);

[0015] S is a dedicated radio frequency signal source; S to T i T i The small-scale channel fading coefficients up to D are represented by h. 1i h 2i It means that, among them, h x ~CN(0,λ) x ), x = 1i or 2i; CN(·) represents a complex Gaussian distribution; i * Indicates the subscript of the selected label; S to T i T i The distance to D is represented by d. 1i d 2i express;

[0016] Using the Least Mean Square Estimator (LMSE), the true value h of the channel's small-scale fading is obtained. x and estimated value The relationship between them is expressed by the following formula:

[0017]

[0018] In the above formula, e x Is with h x Independent Gaussian random variables, and have Therefore, there is in,

[0019] T i The signal received at the location is

[0020]

[0021] In the formula, P sLet be the transmission power of the source node, s(t) be the unit energy signal transmitted by the source node, and α be the path loss factor.

[0022] Using a fixed reflection coefficient β, by T i The reflected signal and the signal used to drive T i The signals of the circuit itself are written as follows: and

[0023] The tag itself consumes P c It means that when At that time, T i It has the ability to reflect signals; at this time, the received signal at point D is:

[0024]

[0025] (3) In the formula, c(t) is T i The transmitted information signal has unit energy; n(t) is a signal with variance σ. 2 Zero-mean additive white Gaussian noise; η represents the signal that can be effectively reflected by the tag.

[0026] Expanding equation (3) according to the signal term and the noise term, we get:

[0027]

[0028] (4) In the formula, the first term is the signal term; the noise term consists of three parts, the first two parts are the noise introduced by CEE, and the last term is Gaussian white noise;

[0029] The received signal-to-noise ratio at point D is:

[0030]

[0031] In the formula, and This indicates the input signal-to-noise ratio.

[0032] In each transmission, the tag that maximizes the signal-to-noise ratio at point D is selected for transmission, i.e.:

[0033]

[0034] The interruption probability of the tag selection scheme in equation (6) is as follows:

[0035]

[0036] (7) In this equation, step (a) is based on the channel independence assumption; τ = 2 R -1 represents the threshold signal-to-noise ratio, and R represents the target rate.

[0037] (7) In the formula, any link S→T i →The probability of interruption of D is calculated according to the following formula:

[0038]

[0039] If T i The harvested energy is insufficient to drive its own circuit, so the link S→T i →D interruption; if the harvested energy can drive its own circuit, but the corresponding signal-to-noise ratio is below the threshold, link S→T i →D interruption; The interruption probabilities in the above two cases are represented by P1 and P2 in equation (8), respectively;

[0040] P1 is:

[0041]

[0042] P2 is:

[0043]

[0044] Replace X and Y respectively The corresponding probability density function is Considering that both X and Y are greater than 0, and Y may be less than 0. Obtain step (c) in equation (11); apply variable substitution to equation (12). Equation (13) is obtained; a special integration function is used for the integral in equation (13). Finally, we obtain equation (14); K1(·) is the second-order modified Bessel function of the first kind;

[0045] Combining equations (9) and (14), for any link S→T i →The probability of interruption for D is:

[0046]

[0047] In the above formula, From equation (15), when selecting the tag that maximizes the signal-to-noise ratio among all tags, the system's outage probability is:

[0048]

[0049] In the high signal-to-noise ratio region, there is γ in →∞ and Δ→0; at this time, the corresponding approximate system interruption probability is:

[0050]

[0051] In equation (17), In step (d), limits were used. When considering CEE, the diversity order of the tag selection scheme that maximizes the signal-to-noise ratio is:

[0052]

[0053] If CEE does not exist, that is: The system's interrupt probability under perfect CSI conditions is:

[0054]

[0055] In the above formula,

[0056] In the high signal-to-noise ratio region, Θ→0, the usage limit is reached. and e -x ≈1-x (x→0), the diversity order under perfect CSI condition is:

[0057]

[0058] This invention proposes a method for evaluating the impact of channel estimation error on tag selection scheme performance. The channel estimation error is modeled as a Gaussian random variable independent of the estimated channel coefficients. Based on this, the signal source selects the tag that maximizes the signal-to-noise ratio (SNR) of the destination node for transmission each time. To accurately describe the performance of the actual system, the tag's own energy consumption is also considered; that is, only tags whose collected energy can power their own circuitry can transmit information. This invention analyzes the system's outage probability when channel estimation error exists. Simulation results show that channel estimation error significantly affects the performance of the tag selection scheme; the tag's own energy consumption has a significant impact on the system's outage performance; if the tag's own energy consumption is not considered, the obtained results deviate greatly from the actual results. Attached Figure Description

[0059] The following detailed description, in conjunction with the accompanying drawings and specific embodiments, further illustrates the following:

[0060] Figure 1 This is a schematic diagram of the label selection model.

[0061] Figure 2 The relationship between interruption probability and the number of tags (R=6, β=0.8, )picture.

[0062] Figure 3 The probability of interruption varies with the channel estimation error power (R = 6, β = 0.8). )picture.

[0063] Figure 4 The impact of target velocity and reflection coefficient on system performance (N=5, P) s =30dBm, (α=4) Figure. Detailed Implementation

[0064] The system model in this invention is as follows Figure 1 As shown, Figure 1 As shown, the backscatter communication system proposed in this invention consists of one source node (S, SourceNode) and N tags T with backscattering capabilities. i (i = 1, 2, ..., N), consisting of a destination node (D). S is a dedicated radio frequency signal source. S to T i T i The small-scale channel fading coefficients up to D are represented by h. 1i h 2i It means that, among them, h x ~CN(0,λ) x ), x = 1i or 2i. CN(·) denotes a complex Gaussian distribution. * Indicates the index of the selected label. S to T i T i The distance to D is represented by d. 1i d 2i express.

[0065] If the Least Mean Square Estimator (LMSE) is used, the true value of the small-scale fading of the channel, h x and estimated value The relationship between them is expressed by the following formula:

[0066]

[0067] In the above formula, e x Is with h x Independent Gaussian random variables, and have Therefore, there is in,

[0068] T i The received signal is written as

[0069]

[0070] In the above formula, P s Let be the transmit power of the source node, and s(t) be the unit energy signal transmitted by the source node. α represents the path loss factor. Therefore, according to the above definition, the channel gain of the signal can be written as... Unlike the independent and identically distributed assumption in existing studies, the use of equation (2) implies that the channel gain will follow independent and non-distributed Rayleigh fading, which can more accurately describe the actual characteristics of the channel.

[0071] In this invention, a fixed reflection coefficient β is used, so that the reflection coefficient T i The reflected signal and the signal used to drive T i The signals of the circuit itself are written as follows: and

[0072] The tag itself consumes P c It means that when At that time, T i It has the ability to reflect signals. At this time, the received signal at point D can be written as:

[0073]

[0074] (3) In the formula, c(t) is T i The transmitted information signal has unit energy. n(t) is a signal with variance σ. 2 The zero-mean additive white Gaussian noise. Due to the limitations of the device itself, the tag cannot reflect all the signal; signal loss occurs during reflection. η represents the signal that can be effectively reflected by the tag.

[0075] Expanding equation (3) according to the signal term and the noise term, we can write it as follows:

[0076]

[0077] In equation (4), the first term is the signal term. The noise term consists of three parts: the first two parts are noise introduced by CEE, and the last term is Gaussian white noise.

[0078] The received signal-to-noise ratio at point D is written as:

[0079]

[0080] In the above formula, and This indicates the input signal-to-noise ratio.

[0081] In this invention, in each transmission, the tag that maximizes the signal-to-noise ratio at point D is selected for transmission, that is:

[0082]

[0083] Since the channels are independent, the interruption probability of the tag selection scheme in equation (6) can be written as follows:

[0084]

[0085] In equation (7), step (a) is based on the channel independence assumption. τ=2 R-1 represents the threshold signal-to-noise ratio, and R represents the target rate. In (7), any link S→T i →The probability of interruption of D is calculated according to the following formula:

[0086]

[0087] If T i The harvested energy is insufficient to drive its own circuit, so the link S→T i →D interruption; Additionally, if the harvested energy is sufficient to power its own circuitry, but the corresponding signal-to-noise ratio is below the threshold, link S→T i →D Interruption. The interruption probabilities in the above two cases are represented by P1 and P2 in equation (8), respectively.

[0088] P1 is:

[0089]

[0090] In equation (9), the actual channel value h is used when calculating the corresponding interruption probability. 1i This is because the tag only needs to simply divide the received signal into two parts, and this operation does not require channel estimation.

[0091] P2 is:

[0092]

[0093] In equation (10), step (b) is derived based on the independence between variables. To simplify the expression, X and Y are used to replace... The corresponding probability density function is written as Considering that both X and Y are greater than 0, and Y may be less than 0. We obtain step (c) in equation (11). We then apply variable substitution to equation (12). Equation (13) is obtained. A special integration function is used for the integral in equation (13). Finally, equation (14) is obtained. K1(·) is the first-order modified Bessel function of the second kind.

[0094] Combining equations (9) and (14), for any link S→T i →The final probability of D being interrupted is:

[0095]

[0096] In the above formula, From equation (15), when selecting the tag that maximizes the signal-to-noise ratio among all tags, the system's outage probability is:

[0097]

[0098] In the high signal-to-noise ratio region, there is γ in →∞ and Δ→0. At this point, the corresponding approximate system interruption probability is:

[0099]

[0100] In equation (17), Observing equation (17), in the high signal-to-noise ratio region, P1→0, and the system interruption probability is mainly determined by P2. This is because in the high signal-to-noise ratio region, the energy collected by the tag is almost always sufficient to meet its own energy consumption requirements. In step (d), the limit is used. When considering CEE, the diversity order of the tag selection scheme that maximizes the signal-to-noise ratio is derived as follows:

[0101]

[0102] Observing equations (8) and (14)-(16), if CEE does not exist, that is: The system's interrupt probability under perfect CSI conditions is:

[0103]

[0104] In the above formula,

[0105] In the high signal-to-noise ratio region, Θ→0, the usage limit is reached. and e -x ≈1-x (x→0), the derivation of the diversity order under the perfect CSI condition is as follows:

[0106]

[0107] The existence of channel estimation error makes the system's diversity gain zero. If CEE does not exist, then the system's diversity order equals the number of tags. Furthermore, observing equation (15), because... Therefore, the probability of system outage when CEE is present will still decrease as the number of tags increases.

[0108] Simulation results

[0109] The simulation parameters are set as follows: It is assumed that the estimation error power of all newly arrived channels is equal, i.e.: The coordinates of S and D are (0,0)m and (16,0)m, respectively. All labels are randomly distributed within a circle centered at coordinates (8,0). Any label T... i The coordinates are (8+rcosθ) i rsinθ i )m, where r = 8m is the radius of the circle, θ i The random distribution is in the interval (0, 2π). Other parameters are set as follows: λ1i =λ 2i =1, P c =8.9 microwatts, α=3, η=0.6, σ 2 = -90dBm.

[0110] With parameters set to R = 6 and β = 0.8, hour, Figure 2 The impact of the number of labels on system performance was examined. The analytical solution was obtained using formula (16), and the Monte Carlo results were obtained using formulas (6-8). Figure 2 As can be seen, the analytical solution completely coincides with the Monte Carlo results, verifying the correctness of the theoretical analysis. The system's interrupt performance does indeed improve with the increase of the number of tags, when P... s When the value is greater than 40dBm, the performance curve gradually becomes horizontal, which verifies the analysis of the system diversity gain performance.

[0111] Depend on Figure 3 It can be seen that when When the system's interruption probability is 1, the system will always be in an interrupted state. This shows that channel estimation error has a direct impact on system performance. Furthermore, when... In perfect CSI, in the high signal-to-noise ratio region, for every 10 dBm increase in input power, the interruption probability decreases by 5 orders of magnitude, verifying the analysis of system diversity performance under perfect CSI.

[0112] Figure 4 The effects of target velocity and reflection coefficient on system performance were investigated. When the target velocity R > 10, the system is always in an interrupted state. Figure 4 In this context, "E.-C." is an abbreviation for the English phrase "Energy-Causality." For a tag selection scheme that considers the tag's own energy consumption (the scheme proposed in this invention), the smaller the reflection coefficient, the better the performance. The smaller the reflection coefficient, the smaller the reflected signal power, the smaller the signal-to-noise ratio received by D, and the higher the interruption probability should be. However, considering the tag's own energy consumption, the smaller the reflection coefficient, the more energy can be supplied to the tag, the greater the energy collected by the tag, and the lower the probability of the tag's own interruption, that is, the smaller P1 in equation (8), and the lower the system's interruption probability. In other words, P1 plays a major role in the system's interruption probability.

[0113] If we disregard the tag's own power consumption, meaning that regardless of the reflectivity value, the signal energy allocated to the tag is assumed to be sufficient to drive the circuit, the performance curve under this condition is also plotted on [the graph]. Figure 4As analyzed above, in this case, the larger the reflection coefficient, the greater the reflected signal power, and the lower the probability of system interruption. It is evident that if the tag's own energy consumption is not considered in the performance analysis, the results obtained will deviate significantly from the actual system. Figure 4 In this case, when R=1, the performance can deviate by approximately eight orders of magnitude.

[0114] In multi-tag backscatter communication systems, system performance can be enhanced through appropriate tag selection schemes. This invention proposes a method for evaluating the impact of channel estimation error on tag selection scheme performance. The channel estimation error is modeled as a Gaussian random variable independent of the estimated channel coefficients. Based on this, the signal source selects the tag that maximizes the signal-to-noise ratio (SNR) of the destination node for transmission each time. To accurately describe the performance of the actual system, the tag's own energy consumption is also considered; that is, only tags whose collected energy can power their own circuitry can transmit information. This invention analyzes the system's outage probability when channel estimation error exists. Simulation results show that channel estimation error significantly affects the performance of the tag selection scheme; the tag's own energy consumption has a significant impact on the system's outage performance. Without considering the tag's own energy consumption, the obtained results deviate significantly from the actual results.

[0115] Although the present invention has been described in detail above with general descriptions and specific embodiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. A backscatter communication system, characterized in that: It consists of one source node (S, SourceNode) and N tags T with backscattering capability. i (i = 1, 2, ..., N), consisting of a destination node (D, DestinationNode); S is a dedicated radio frequency signal source; S to T i T i The small-scale channel fading coefficients up to D are represented by h. 1i h 2i It means that, among them, h x ~CN(0,λ) x ), x = 1i or 2i; CN(·) represents a complex Gaussian distribution; i * Indicates the subscript of the selected label; S to T i T i The distance to D is represented by d. 1i d 2i express; Using the Least Mean Squares Estimator (LMSE), the true value h of the channel's small-scale fading is obtained. x and estimated value The relationship between them is expressed by the following formula: In the above formula, e x Is with h x Independent Gaussian random variables, and have Therefore, there is in, T i The signal received at the location is In the formula, P s Let be the transmission power of the source node, s(t) be the unit energy signal transmitted by the source node, and α represent the path loss factor. Using a fixed reflection coefficient β, by T i The reflected signal and the signal used to drive T i The signals of the circuit itself are written as follows: and The tag itself consumes P c It means that when At that time, T i It has the ability to reflect signals; at this time, the received signal at point D is: (3) In the formula, c(t) is T i The transmitted information signal has unit energy; n(t) is a signal with variance σ. 2 Zero-mean additive white Gaussian noise; η represents the signal that can be effectively reflected by the tag; Expanding equation (3) according to the signal term and the noise term, we get: (4) In the formula, the first term is the signal term; the noise term consists of three parts, the first two parts are the noise introduced by CEE, and the last term is Gaussian white noise; The received signal-to-noise ratio at point D is: In the formula, and Indicates the input signal-to-noise ratio; In each transmission, the tag that maximizes the signal-to-noise ratio at point D is selected for transmission, i.e.: The interruption probability of the tag selection scheme in equation (6) is as follows: (7) In this equation, step (a) is based on the channel independence assumption; τ = 2 R -1 represents the threshold signal-to-noise ratio, and R represents the target rate.

2. The backscatter communication system according to claim 1, characterized in that: (7) In the formula, any link S→T i →The probability of interruption of D is calculated according to the following formula: 。 3. The backscatter communication system according to claim 2, characterized in that: If T i The harvested energy is insufficient to drive its own circuit, so the link S→T i →D interruption; if the harvested energy can drive its own circuit, but the corresponding signal-to-noise ratio is below the threshold, link S→T i →D interruption; The interruption probabilities in the above two cases are represented by P1 and P2 in equation (8), respectively; P1 is: P2 is: Replace X and Y respectively The corresponding probability density function is Considering that both X and Y are greater than 0, and Y may be less than 0. Obtain step (c) in equation (11); apply variable substitution to equation (12). Equation (13) is obtained; a special integration function is used for the integral in equation (13). Finally, we obtain equation (14); K1(·) is the second-order modified Bessel function of the first kind; Combining equations (9) and (14), for any link S→T i →The probability of interruption for D is: In the above formula, From equation (15), when selecting the tag that maximizes the signal-to-noise ratio among all tags, the system's outage probability is: In the high signal-to-noise ratio region, there is γ in →∞ and Δ→0; at this time, the corresponding approximate system interruption probability is: In equation (17), In step (d), limits were used.

4. The backscatter communication system according to claim 3, characterized in that: When considering CEE, the diversity order of the tag selection scheme that maximizes the signal-to-noise ratio is: 。 5. The backscatter communication system according to claim 4, characterized in that: If CEE does not exist, that is: The system interrupt probability under perfect CSI conditions is: In the above formula, 6. The backscatter communication system according to claim 5, characterized in that: In the high signal-to-noise ratio region, Θ→0, the usage limit is reached. and e -x ≈1-x (x→0), the diversity order under perfect CSI condition is:

Images