Design method of unmanned system direct current power line carrier communication impedance matching coupler

Abstract

The invention discloses a design method of an unmanned system direct-current power line carrier communication impedance matching coupler, and relates to the field of power line carrier communication. The invention aims to solve the problems of low communication rate and poor communication reliability caused by low efficiency and impedance mismatch of filtering and impedance matching in a broadband in the existing communication impedance matching method. The method comprises the following steps: acquiring a transfer function and an actual frequency of the elliptic low-pass filter so as to obtain a topological structure of the elliptic low-pass filter; obtaining a transfer function and an actual frequency of the Chebyshev II-type high-pass filter so as to obtain an actual topological structure of the Chebyshev II-type high-pass filter; cascading the Chebyshev II-type high-pass filter topological structure with the elliptic low-pass filter topological structure to obtain a composite band-pass filter; constructing a double-L-shaped impedance matching network; and the composite band-pass filter and the double-L-shaped impedance matching network jointly form a direct-current power line carrier communication impedance matching coupler. The method is used for designing the communication impedance matching coupler.

Description

Technical Field

[0001] This invention relates to the field of power line carrier communication, and in particular to a design method for an impedance matching coupler for DC power line carrier communication in unmanned systems. Background Technology

[0002] With the widespread application of intelligent unmanned systems such as autonomous vehicles, drones, and robots in industries, agriculture, and the military, the demand for integrated internal energy supply and information interaction is becoming increasingly prominent. Traditional independent power supply and communication architectures suffer from problems such as complex cabling, weight redundancy, and low reliability. DC power line carrier communication technology, by reusing power lines, simultaneously achieves the dual functions of "power transmission" and "data communication," significantly reducing system complexity and cost, thus becoming an important research direction for embedded communication in unmanned systems. Especially in the aerospace context, DC PLC technology has significant advantages: Lightweight: Reusing the power distribution network reduces dedicated communication cables, directly reducing spacecraft mass and optimizing carrying capacity; High reliability: In complex electromagnetic environments, the DC bus balanced transmission scheme has better suppression capabilities for switching ripple noise than traditional signal cables; Plug and play: Adapting to the trend of distributed power supply, it provides convenient communication links for dispersed functional units, avoiding the wiring complexity of traditional buses (such as CAN).

[0003] However, applying DC PLC technology to unmanned autonomous systems (especially spacecraft and drones) faces severe technical challenges. These systems typically feature multi-branch distributed topologies, dynamic switching of high-frequency switching loads, and harsh electromagnetic environments, leading to the following complex characteristics in DC PLC channels: Deep frequency-selective fading: Impedance mismatch at distributed nodes in multi-branch power line channels causes severe signal reflection. Wideband nonlinear time-varying impedance: High-frequency switching power supplies and pulsed loads cause nonlinear changes in channel input impedance over a wide bandwidth, which traditional lumped parameter models cannot accurately characterize. Non-stationary signal attenuation: In twisted-pair cables used in spacecraft and other scenarios, the skin effect and proximity effect exacerbate signal attenuation, with attenuation fluctuations exceeding 20 dB in the 2-80 MHz frequency band. These characteristics result in high communication error rates and limited transmission bandwidth for DC PLC systems, severely restricting their engineering applications in unmanned systems.

[0004] To address impedance mismatch, researchers have proposed various impedance matching schemes, including hardware-implemented matching networks, traditional matching models, and broadband impedance matching. However, these methods still have the following key limitations: Hardware-implemented matching networks primarily focus on narrowband communication scenarios, and their insufficient bandwidth makes them unsuitable for the optimization needs of broadband PLCs in the wide frequency range, resulting in low filtering and impedance matching efficiency and consequently low communication speeds. Traditional matching models lack a systematic consideration of the capacitive-inductive alternation of power line impedance in broadband communication bands, leading to low accuracy in describing the dynamic characteristics of broadband impedance and resulting in impedance mismatch, ultimately causing low communication speeds due to severe signal reflection. Existing broadband impedance matching schemes mainly employ discrete-point amplitude matching strategies, failing to establish a quantitative mapping relationship between impedance matching parameters and key performance indicators such as bit error rate and channel capacity of the communication system. This results in the inability to achieve high-speed and effective impedance matching, leading to poor communication reliability and low speeds. Summary of the Invention

[0005] This invention addresses the problems of low efficiency and impedance mismatch in existing communication impedance matching methods over wide bandwidths, which lead to low communication rates and poor communication reliability. It proposes a design method for an impedance matching coupler for DC power line carrier communication in unmanned systems.

[0006] A design method for an impedance matching coupler for DC power line carrier communication in an unmanned system, specifically as follows:

[0007] Step 1: Set the design goal of the elliptic low-pass filter. Based on the design goal, obtain the transfer function and actual frequency of the elliptic low-pass filter, thereby obtaining the topology of the elliptic low-pass filter.

[0008] Step 2: Set the design goals for the Chebyshev Type II high-pass filter. Based on the design goals, obtain the transfer function and actual frequency of the Chebyshev Type II high-pass filter, thereby obtaining the actual topology of the Chebyshev Type II high-pass filter.

[0009] Step 3: Cascade the Chebyshev Type II high-pass filter topology obtained in Step 2 with the elliptic low-pass filter topology obtained in Step 1. Then, cascade two high-frequency suppression inductors at the output of the elliptic low-pass filter topology to obtain a composite bandpass filter.

[0010] Step 4: Obtain the required component types and values ​​for the double L-type impedance matching network, thereby constructing the double L-type impedance matching network;

[0011] Step 5: Combine the composite bandpass filter obtained in Step 3 with the double L-type impedance matching network obtained in Step 4 to form a DC power line carrier communication impedance matching coupler.

[0012] The beneficial effects of this invention are as follows:

[0013] This invention relates to a DC power line carrier communication impedance matching coupler comprising a composite bandpass filter and a dual L-type impedance matching network. The composite bandpass filter is composed of a Scheff II type high-pass filter, a fifth-order elliptic low-pass filter, and an inductor cascaded together. The elliptic low-pass filter features a steep transition band, effectively suppressing high-frequency interference outside the passband. A high-frequency suppression inductor is cascaded at the low-pass filter output, improving high-frequency filtering performance. The Scheff II type high-pass filter exhibits strict monotonicity and ripple-free operation within its passband, avoiding ripple superposition after cascading with the elliptic low-pass filter. The dual L-type impedance matching network effectively compresses impedance mismatch over a wide bandwidth, reducing signal reflection and thus improving communication speed. The composite bandpass filter features a flat passband and a steep transition band, strongly suppressing out-of-band noise and switching power supply harmonics, thereby enhancing communication reliability. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of a PLC transceiver circuit.

[0015] Figure 2 The topology diagram of a 5th-order elliptic low-pass filter;

[0016] Figure 3 The diagram shows the topology of the low-pass filter after adding a high-frequency suppression inductor.

[0017] Figure 4 For a fifth-order elliptic low-pass filter S 2,1 Parameter simulation diagram;

[0018] Figure 5 The low-pass filter S after adding a high-frequency suppression inductor 2,1 Parameter simulation diagram;

[0019] Figure 6 This is a Chebyshev Type II high-pass filter topology;

[0020] Figure 7 For Chebyshev Type II high-pass filter S 2,1 Simulation curves;

[0021] Figure 8 This is a schematic diagram of the cascaded bandpass filter topology;

[0022] Figure 9 For the cascaded bandpass filter S 2,1 Simulation result diagram;

[0023] Figure 10 This is a schematic diagram of a bidirectional L-type impedance matching network;

[0024] Figure 11 This is an L-shaped matching network diagram;

[0025] Figure 12 Smith chart for normalized impedance and reflection coefficient;

[0026] Figure 13 A Smith chart showing eight regions mapped based on impedance values;

[0027] Figure 14 An example diagram of the movement strategy for impedance mapping points in a Smith chart;

[0028] Figure 15 A flowchart for determining the parameters of L-type impedance matching network components using a Smith chart;

[0029] Figure 16 Here is the overall flowchart of the impedance matching algorithm;

[0030] Figure 17 This is a schematic diagram of the structure of an unmanned system DC PLC experimental platform.

[0031] Figure 18 For unmanned systems experimental platforms;

[0032] Figure 19 This is a diagram of the platform's cable routing.

[0033] Figure 20 The communication rate curves are shown for four different cases in branch A.

[0034] Figure 21 Optimization amplitude curves for matching under various conditions in branch A;

[0035] Figure 22 The communication rate curves are shown for four different cases in branch C.

[0036] Figure 23 The matching optimization amplitude curves for branch C under various conditions are presented. Detailed Implementation

[0037] Specific Implementation Method 1: The specific process of the design method for an impedance matching coupler for DC power line carrier communication in an unmanned system described in this implementation method is as follows:

[0038] Step 1: Define the design goals of the elliptic low-pass filter. Based on these goals, obtain the transfer function and actual frequency of the elliptic low-pass filter, thereby obtaining its topology. Specifically:

[0039] Step 11: Set the design goals for the elliptic low-pass filter based on the chip bandwidth:

[0040] The design objectives of the elliptic low-pass filter include: passband cutoff frequency. Stopband start frequency Minimum stopband attenuation Maximum passband attenuation and port impedance ;

[0041] Steps 1 and 2: Obtain the normalized passband cutoff frequency and normalized stopband start frequency according to the design objectives, specifically as follows:

[0042]

[0043] in, It is the normalized passband cutoff frequency. It is the normalized stopband start frequency.

[0044] In this step, the present invention uses normalization processing, so that the subsequent pole and zero distribution and transfer function derivation can be performed under dimensionless conditions, which facilitates parameter optimization.

[0045] Step 13: Based on the maximum passband attenuation A p Obtain the passband ripple coefficient Based on the minimum stopband attenuation A s Obtain the stopband ripple coefficient and utilize and Obtain the selectivity factor of the filter :

[0046]

[0047]

[0048]

[0049] in, It is the passband ripple factor. It is the maximum attenuation in the passband. It is a logarithmic calculation with base 10. It is the minimum stopband attenuation. It is the stopband ripple factor;

[0050] This step, and These respectively reflect the maximum permissible fluctuation in the passband and the minimum suppression requirement in the stopband; The selectivity factor of the filter is defined, which directly affects the choice of the order of the elliptic low-pass filter.

[0051] Step 1.4: Filter-based selectivity factor To obtain the minimum order of an elliptic low-pass filter, specifically:

[0052] First, the design of elliptic filters relies on elliptic function theory, and its core parameter is the modulus. and mold repair Modulus Defined as the reciprocal of the normalized stopband start frequency:

[0053]

[0054] Then, using the modulus Determine the construction of the first kind of complete elliptic integral :

[0055]

[0056] in, It is the curvature of an ellipse;

[0057] For ease of calculation, based on the properties of the first kind of complete elliptic integral, hour, , hour, ;

[0058] Finally, through the modulus , and selection factor The minimum order of a fifth-order elliptic low-pass filter can be derived. The calculation formula is as follows:

[0059]

[0060] in, It is a positive integer; greater than or equal to The smallest positive integer;

[0061] Step 15: Using the Jacobian elliptic function To obtain the pole and zero angular frequencies of an elliptic low-pass filter, specifically:

[0062] First, construct the first type of elliptic integral function:

[0063]

[0064] in, It's an angle;

[0065] Then, construct the Jacobian elliptic function based on the elliptic integral function of the first kind:

[0066]

[0067]

[0068]

[0069]

[0070]

[0071] in, It is a complex variable. It is the cosine of an ellipse. It is the sine of an ellipse. It is an elliptical mode chord;

[0072] Then, using MATLAB, the real part of the pole locations of the elliptic filter is obtained using the Jacobian elliptic function. and the virtual part The real part at the zero point and the virtual part And using the real part of the zero point position and the virtual part Obtaining the angular frequency at zero point ;

[0073] Step 16, according to The transfer function is constructed using the angular frequencies of the transmission zeros and poles of the first-order elliptic low-pass filter, specifically:

[0074]

[0075] in, It is a complex frequency. yes The total number of poles of an elliptic low-pass filter. yes Pole labels of an elliptic low-pass filter. yes The angular frequency of the first zero of the elliptic low-pass filter. yes The angular frequency of the second zero of the elliptic low-pass filter. yes The first elliptic low-pass filter The real part of each pole yes The first elliptic low-pass filter The imaginary part of each pole yes The first elliptic low-pass filter The real part of each pole yes The first elliptic low-pass filter The imaginary part of each pole yes The transfer function of a first-order low-pass filter;

[0076] In this step, the sign of the denominator of the transfer function depends on the specific form of the poles. The basic form of the transfer function is: If the pole is Substituting this into the basic form of the transfer function, the sign in the denominator of the expanded transfer function will be a minus sign; if Substituting this into the basic form of the transfer function, the sign in the denominator of the expanded transfer function is a plus sign.

[0077] Step 17: Obtain normalized component parameter values ​​based on the transfer function, specifically as follows:

[0078] For a filter with impedance Z at both ends, the reflection coefficient is obtained using the transfer function, and then the system input impedance is calculated using the reflection coefficient:

[0079]

[0080]

[0081] in, It is the reflection coefficient. It is the normalized port impedance. It is the system input impedance;

[0082] Again Expanding the continuous fraction yields:

[0083]

[0084] in, , , , and It is the capacitance value. and It is the inductance value;

[0085] Using the fraction expansion The normalized component values ​​are calculated as follows:

[0086] To reduce the number of inductors used, a capacitor is connected in parallel first. The first parallel capacitor: as s→∞, the input impedance is mainly due to… Decision. By extracting The behavior at infinity yields the normalized capacitance value of the first capacitor. Then, the normalized capacitance value of the second capacitor is calculated. The normalized capacitance value of the third capacitor The normalized capacitance value of the fourth capacitor The normalized capacitance value of the fifth capacitor The normalized inductance value of the first inductor The normalized inductance value of the second inductor .

[0087] Step 18: Convert the normalized passband cutoff frequency and the normalized stopband start frequency into actual frequencies, specifically as follows:

[0088]

[0089] in, Pick or , It is the actual frequency after conversion;

[0090] Step 19: Denormalize the normalized component parameter values ​​to the passband cutoff frequency. By considering the port impedance Z, the actual values ​​of the component parameters can be obtained, specifically:

[0091]

[0092]

[0093] Step 10: Based on the transfer function of the fifth-order elliptic low-pass filter obtained in Step 16, the actual frequency obtained in Step 18, and the actual values ​​of the component parameters obtained in Step 19, construct the topology of the fifth-order elliptic low-pass filter. To enhance the filter's suppression of common-mode noise, the single-ended structure is converted into a symmetrical balanced structure, such as... Figure 2 As shown.

[0094] Step 2: Set the design goals for the Chebyshev II high-pass filter. Based on these goals, obtain the transfer function and actual frequency of the Chebyshev II high-pass filter, thereby obtaining its actual topology. Specifically:

[0095] Step 21: Set the design goals for the Chebyshev Type II high-pass filter based on the chip bandwidth:

[0096] The design objectives of the Chebyshev Type II high-pass filter include: passband cutoff frequency. Stopband cutoff frequency Maximum ripple in passband Minimum stopband attenuation and port impedance ;

[0097] Step 22: Set the passband cutoff frequency Using the reference frequency, Obtain the normalized stopband start frequency :

[0098]

[0099] in, It is the normalized stopband start frequency. It is the normalized passband cutoff frequency. It is the stopband cutoff frequency;

[0100] Steps two and three: Utilizing the minimum stopband attenuation Obtain the stopband ripple coefficient and utilize Obtaining the passband ripple coefficient Specifically:

[0101]

[0102]

[0103] in, It is the stopband ripple factor. It is the minimum stopband attenuation. It is the passband ripple factor. It has the lowest passband attenuation;

[0104] Step 24: Use the normalized stopband start frequency obtained in Step 22. Obtaining the order of a Chebyshev Type II filter Specifically:

[0105]

[0106] in, is a positive integer, It is greater than The smallest positive integer;

[0107] Step 25: Utilizing the order of the Chebyshev Type II filter Obtain the zeros of the normalized low-pass prototype filter. and normalized low-pass prototype filter poles and utilize and The transfer function for constructing the low-pass prototype filter is as follows:

[0108] First, using the order of the Chebyshev Type II filter Obtain the zeros of the normalized low-pass prototype filter. and normalized low-pass prototype filter poles This allows us to obtain the angular frequencies of the zeros of the normalized low-pass prototype filter and the real and imaginary parts of the poles of the normalized low-pass prototype filter, specifically:

[0109]

[0110]

[0111]

[0112]

[0113]

[0114] in, It is an ellipse shape parameter used to control pole distribution. It is the real contribution item at the pole. It is the contribution term of the imaginary part of the pole. It is the angular frequency at zero point. It is the real part of the pole. It is the imaginary part of the pole;

[0115] Then, the transfer function of the low-pass prototype filter is constructed using the angular frequencies of the zeros of the normalized low-pass prototype filter and the real and imaginary parts of the poles of the normalized low-pass prototype filter, specifically:

[0116]

[0117] in, These are pole labels. It is the first The imaginary part of each pole It is the first The real part of each pole It is a complex frequency. It is the first angular frequency at zero point, It is the total number of poles. It is the transfer function of the low-pass prototype filter;

[0118] Step 26: Convert the low-pass prototype filter into a high-pass prototype filter to obtain the transfer function of the high-pass prototype filter. Specifically:

[0119]

[0120]

[0121] in, It is the complex frequency variable of the normalized high-pass prototype filter. It is the complex frequency variable in the target high-pass prototype filter. It is an intermediate variable (unit is) ), It is the minimum value of the chip carrier wave;

[0122] Step 27: Utilizing the transfer function of the high-pass prototype filter Obtain the normalized component parameter values ​​of the Chebyshev Type II filter, convert the normalized passband cutoff frequency and normalized stopband start frequency to the actual frequencies, and then inversely normalize the normalized component parameter values ​​back to the passband cutoff frequency. and port impedance To obtain the actual values ​​of the component parameters, specifically:

[0123] First, using the transfer function of the Qualcomm prototype. Obtaining the reflection coefficient The system input impedance is obtained using the reflection coefficient. Specifically:

[0124]

[0125]

[0126] in, It is the reflection coefficient. It is the normalized port impedance. It is the system input impedance;

[0127] Again Expanding the continuous fraction yields:

[0128]

[0129] in, , and It is the capacitance value. and It is the inductance value;

[0130] Using the fraction expansion Solve for the normalized component values, specifically:

[0131] Choose to connect the capacitors in parallel first. The first capacitor in parallel: As the input impedance approaches infinity, it is mainly determined by… Decision. By extracting The behavior at infinity yields the normalized capacitance value of the first capacitor. Then, the normalized capacitance value of the second capacitor is calculated. The normalized capacitance value of the third capacitor The normalized inductance value of the first inductor The normalized inductance value of the second inductor ;

[0132] Then, the normalized passband cutoff frequency and the normalized stopband start frequency are converted into actual frequencies, specifically as follows:

[0133]

[0134] in, Pick or , It is the actual frequency after conversion;

[0135] Then, the normalized component parameter values ​​are denormalized to the passband cutoff frequency. and port impedance To obtain the actual values ​​of the component parameters, specifically:

[0136]

[0137]

[0138] Step 28: Based on the transfer function of the high-pass prototype filter obtained in Step 26. The actual frequency and actual values ​​of component parameters obtained in step 27 are used to construct the topology of a fourth-order Chebyshev Type II filter.

[0139] Step 3: Cascade the Chebyshev Type II high-pass filter topology obtained in Step 2 with the fifth-order elliptic low-pass filter topology obtained in Step 1. Then, cascade two 100nH high-frequency suppression inductors at the output of the fifth-order elliptic low-pass filter topology to obtain a composite bandpass filter.

[0140] Step 4: Obtain the required component types and values ​​for the double L-type impedance matching network, and then construct the double L-type impedance matching network. Specifically:

[0141] Step 41: Transform the load impedance into the target matching impedance using a double L-type impedance matching network. According to the target matching impedance With the real part of the load impedance Based on the relationship, the circuit topology of the double L-type impedance matching network is obtained and the component values ​​are calculated, specifically:

[0142] A1. Obtain the connection relationship between two components in an L-type impedance matching network, and the corresponding component values ​​for each connection relationship. Specifically:

[0143] when Greater than If yes, then execute a; otherwise, execute b.

[0144] a. When Greater than When, the matched elements connected in series will be Place it at the load end, and then obtain the component value, specifically:

[0145] Make the real part of the load impedance R A With the imaginary part X A After the component reactance With component reactance The target matching impedance after using an L-type impedance matching network. Equal, component reactance Component reactance Target matching impedance after using an L-type impedance matching network Relationship:

[0146]

[0147] in, It is the target matching impedance after using an L-type impedance matching network. It is the imaginary unit. , It is the reactance of two components. It is the real part of the load impedance. It is the imaginary part of the load impedance;

[0148] Then, the component reactance is obtained. With component reactance Specifically:

[0149]

[0150] b. When Less than Then the matched elements connected in series Place it at the source end, and then obtain the component value, specifically:

[0151] Make the real part of the load impedance With the imaginary part After the component reactance With component reactance The target matching impedance after using an L-type impedance matching network. Equal, component reactance Component reactance Target matching impedance after using an L-type impedance matching network Relationship:

[0152]

[0153] Then, the component reactance is obtained. With component reactance Specifically:

[0154]

[0155] In this invention, the dual L-type impedance matching network and matching strategy are as follows: Figure 10 As shown. Where R A For the load resistance or any impedance that needs to be matched, the real part is... It is its virtual part. and This represents the reactance of two components in an L-type impedance matching network. The target matching impedance after using an L-shaped impedance matching network is the load impedance in this invention's impedance matching method. and go through and After impedance transformation and They are equal. The specific matching process involves two cases: Figure 11 shown Greater than and Less than R A , Figure 11 (a) in the middle is Greater than Corresponding matching network, Figure 11 (b) in the middle is Less than The corresponding matching network places the series-connected matching components at the load and source ends, respectively. Each L-type impedance matching network component value has two sets, and the double L-type matching has two sets at the TX end and two sets at the RX end, resulting in four sets of matching component values.

[0156] A2. Based on each connection relationship between components and the corresponding component value, obtain multiple sets of component matching values;

[0157] Each L-type impedance matching network element value corresponds to two sets, and the double L-type matching has two sets at the TX end and two sets at the RX end, resulting in four sets of matching element values.

[0158] Step 4.2: Optimize the circuit topology and component values ​​of the double L-type impedance matching network using the Smith chart, specifically as follows:

[0159] B1. Utilizing the target matching impedance and load impedance To obtain the reflection coefficient, specifically:

[0160]

[0161]

[0162] in, It is the normalized load impedance. It is the reflection coefficient. It is the load impedance;

[0163] The Smith chart is a graphical impedance analysis tool that visualizes complex impedance / admittance transformations as normalized parameters using polar coordinates. It consists of resistance and reactance circles of varying radii and is widely used in RF circuit and matching network design. Electromagnetic waves travel along the characteristic impedance... The transmission line transmits the signal, reaching an impedance of Z at the end. L The reflection of a signal under load can be described and analyzed by the reflection coefficient Γ.

[0164] B2. After performing a complex-plane mathematical transformation on the normalized load impedance, map the transformed normalized load impedance onto the Smith chart, and then calculate the reflection coefficient. Mapping onto the Smith chart to construct the Smith chart, as follows: Figure 12 As shown, (a) is the normalized impedance. (a) is the Smith chart of the reflectance coefficient Γ; (b) is the Smith chart of the reflectance coefficient Γ.

[0165] When the normalized load impedance after mathematical transformation of the complex plane is mapped onto the Smith chart, the horizontal axis represents a purely resistive line, meaning that the impedance value mapped to points on this line has only the real part; the impedance value mapped to points on the outermost circle has only the imaginary part. When the load impedance is equal to the impedance of the chip PLC module, i.e., impedance matching is achieved, the point mapped by the normalized load impedance is located at the center of the Smith chart.

[0166] When mapping the reflection coefficient onto the Smith chart, the outermost circle represents a reflection coefficient magnitude of 1, and the four intersecting points on the axis represent reflection coefficients with only real or imaginary parts. If the mapped reflection coefficient is 0, the current point is at the center of the circle, indicating impedance matching, complete signal transmission, and no reflection in the channel.

[0167] B3. Divide the Smith chart into 8 regions. Based on the position of the impedance to be matched at the load end on the Smith chart, obtain the optimal impedance matching parameters, specifically:

[0168] B3-1. The impedance to be matched is mapped to different regions, representing different impedance values. Based on the mapping relationship, the Smith chart is divided into 8 regions, such as... Figure 13 As shown, specifically:

[0169] Region 1 is the intersection of the area above the red horizontal diameter axis and the yellow circle, indicating that the point magnitude in the region is less than the reference impedance, and the inductive real part of the admittance is greater than the reference admittance.

[0170] The yellow circle is an equal-conductivity circle, with a conductivity value of 1 (after normalization) or 20 mS (before normalization); the red horizontal diameter axis is a pure resistance line, and the black vertical diameter axis is a pure reactance line;

[0171] Region 2 is the area above the red horizontal diameter axis and the area to the left of the black vertical axis, excluding the area of ​​the yellow circle. It indicates that the point magnitude in this region is less than the reference impedance, and the inductive and real admittance are less than the reference admittance.

[0172] Region 3 is the area above the red horizontal diameter axis and the area to the right of the black vertical axis, excluding the area of ​​the blue circle. It indicates that the point magnitude in this region is greater than the reference impedance, indicating inductance, and the real part of the impedance is less than the reference impedance.

[0173] The blue circle is an equal resistance circle, and the resistance value of the equal resistance circle is 1 (after normalization) or 50Ω (before normalization).

[0174] Region 4 is the intersection of the area above the red horizontal diameter axis and the blue circle, indicating that the point magnitude in the region is greater than the reference impedance, which is inductive, and the real part of the impedance is greater than the reference impedance.

[0175] Region 5 is the intersection of the area below the red horizontal diameter axis and the blue circle, indicating that the point magnitude in the region is greater than the reference impedance, and the capacitive impedance real part is greater than the reference impedance.

[0176] Region 6 is the area below the red horizontal diameter axis and the area to the right of the black vertical axis, excluding the area of ​​the blue circle. It indicates that the point magnitude in this region is greater than the reference impedance, capacitive, and the real part of the impedance is less than the reference impedance.

[0177] Region 7 is the area below the red horizontal diameter axis and the area to the left of the black vertical axis, excluding the area of ​​the yellow circle. It indicates that the point magnitude in this region is less than the reference impedance, and the real part of the capacitive and admittance is less than the reference admittance.

[0178] Region 8 is the intersection of the region below the red horizontal diameter axis and the region of the yellow circle, indicating that the modulus is less than the reference impedance, and the real part of the capacitive admittance is greater than the reference admittance;

[0179] This step uses a standard 50Ω impedance as the system reference impedance, such as... Figure 13As shown, the Smith chart uses the horizontal axis (red horizontal axis) as the boundary. Points mapped above the horizontal axis represent inductive impedance, and points mapped below the red horizontal axis represent capacitive impedance. Points mapped to the left of the vertical axis (black vertical axis) indicate that the current impedance magnitude is less than the reference impedance of 50Ω, and points mapped to the right of the black line indicate that the current impedance magnitude is greater than the reference impedance of 50Ω. When the impedance mapping point is located to the right of the black line and above the red line, if the real part of the impedance is less than the reference impedance of 50Ω, the mapping point is located in region 3; if it is greater than the reference impedance of 50Ω, it is located in region 4.

[0180] B3-2. Obtain the region where the impedance to be matched is mapped to the Smith circle. According to the Smith circle matching strategy, after adding the first matching device, the complex impedance point moves to the standard constant conductance circle or constant resistance circle. After adding the second device, the complex impedance point moves along the standard constant conductance circle or constant resistance circle to the center point, i.e., the reference impedance, thereby obtaining the optimized topology of the double L-type impedance matching network.

[0181] The Smith chart matching strategy is specifically as follows:

[0182] When the impedance to be matched is mapped to the Smith circle, the region is region 1. The transmitter L-type impedance matching network is connected in series with an inductor and then connected in parallel with a capacitor. The receiver L-type impedance matching network is connected in parallel with an inductor and then connected in series with a capacitor.

[0183] When the impedance to be matched is mapped to the Smith circle, the region is region 2. The L-type impedance matching network at the transmitting end is connected to a parallel capacitor after a series capacitor, and the L-type impedance matching network at the receiving end is connected to a series inductor after a parallel inductor.

[0184] When the impedance to be matched is mapped to the Smith circle and the region is region 3, the L-type impedance matching network at the transmitting end is connected in parallel with a capacitor followed by a series capacitor, and the L-type impedance matching network at the receiving end is connected in series with an inductor followed by a parallel capacitor.

[0185] When the impedance to be matched is mapped to the Smith circle and the region is region 4, the L-type impedance matching network at the transmitting end is connected to a series capacitor after a parallel inductor, and the L-type impedance matching network at the receiving end is connected to a parallel capacitor after a series inductor.

[0186] When the impedance to be matched is mapped to the Smith circle and the region is region 5, the L-type impedance matching network at the transmitting end is connected to a series inductor after a parallel capacitor, and the L-type impedance matching network at the receiving end is connected to a parallel inductor after a series capacitor.

[0187] When the impedance to be matched is mapped to the Smith circle and the region is region 6, the L-type impedance matching network at the transmitting end is connected in parallel with an inductor and then in series with an inductor, and the L-type impedance matching network at the receiving end is connected in series with a capacitor and then in parallel with a capacitor.

[0188] When the impedance to be matched is mapped to the Smith circle, the region is region 7. The transmitter L-type impedance matching network is connected in series with an inductor and then in parallel with an inductor. The receiver L-type impedance matching network is connected in parallel with a capacitor and then in series with a capacitor.

[0189] When the impedance to be matched is mapped to the Smith circle and the region is region 8, the L-type impedance matching network at the transmitting end is connected to a series capacitor followed by a parallel inductor, and the L-type impedance matching network at the receiving end is connected to a parallel capacitor followed by a series inductor.

[0190] The matching parameter combinations for the transmitter and receiver corresponding to each region of the Smith chart are shown in Table 1.

[0191] Table 1. Matching parameter combinations for transmitters and receivers for each region of the Smith chart.

[0192]

[0193] The parameter values ​​of each component in the dual-L impedance matching network are selected in the following manner:

[0194] The first step is to place any The value and any The values ​​form impedance pairs;

[0195] The second step is to mark the location of the normalized impedance to be matched on the Smith chart. The location of the normalized impedance to be matched (located in...) The location of the equal resistance circle (inside or outside if the value is less than 1, outside if the value is greater than 1) directly determines the structure of the simplest L-type matching network (series first then parallel, or parallel first then series). By moving along the "equal resistance circle" or "equal conductance circle," a complete path from the load point to the matching center (center) can be visually planned. Each step of the movement in the path corresponds to a specific reactance or susceptance value, which can be directly read from the circle diagram to obtain a set of component values.

[0196] The third step involves comparing the multiple impedance value pairs calculated in the first step with the graphical values ​​read from the Smith chart path in the second step. The set of impedance values ​​closest to the graphical values ​​is selected as the final design. The dual-L impedance matching method proposed in this invention matches the input impedance at the load end to the output impedance at the transmitter end using a dual-L impedance matching network. Therefore, the Smith chart is primarily studied from the perspective of impedance mapping. The load input impedance is used as the impedance to be matched. After calculating the corresponding multiple sets of dual-L impedance matching parameters, this paper determines the optimal impedance matching parameters based on the position of the point mapped to the load input impedance in the Smith chart. The impedance point movement strategy is as follows: Figure 14As shown in (a): a series inductor moves the impedance point clockwise along the constant resistance circle; a series capacitor moves the impedance point counterclockwise along the constant resistance circle; a parallel inductor moves the impedance point counterclockwise along the constant conductance circle; and a parallel capacitor moves the impedance point clockwise along the constant conductance circle. The matching strategies also differ when the impedance to be matched is mapped to different regions. Taking any impedance point as an example, such as... Figure 14 As shown in (b), this point can be moved towards the equiconductivity circle by connecting a series capacitor, and by continuing to connect an inductor in parallel, the impedance point can be moved counterclockwise along the equiconductivity circle to the center impedance point. The two devices form an L-shaped impedance matching network to achieve impedance matching at the reference point.

[0197] This invention employs the Smith chart impedance point mapping method for parameter optimization. The corresponding matching networks for different load impedance conditions are as follows: Figure 15 As shown. Based on the load impedance normalized to the target impedance, the location of this point on the Smith chart is mapped, and the path to the target impedance point is determined. Figure 15 In this context, Z0 represents the target impedance, Y0 represents the target impedance, and Z represents the admittance of the load impedance. r Y is the real part of the load impedance, and Y is the load admittance. r Let be the real part of the load admittance. Combining the L-type impedance matching network parameter calculation method, the impedance matching method designed in this invention is as follows: Figure 16 As shown.

[0198] This invention employs a bidirectional L-shaped passive network, deployed at the transmitter (TX) and receiver (RX) ends of a PLC system, respectively, aiming to match the varying load impedance and source impedance to a set target impedance (e.g., 50 Ω). Based on the real part of the load impedance (R... A The relationship between the magnitude of the impedance and the target impedance (Z0) can be divided into two cases (Z0 > R). A Or Z0 < R A This invention directly calculates the reactance values ​​(X1, X2) in the impedance matching network using analytical formulas, yielding two sets of mathematical solutions for each L-shaped impedance matching network. The Smith chart method is then used to optimize the calculated component parameters. This invention maps the complex impedance to be matched onto a Smith chart, queries a preset matching strategy based on its location, and intelligently selects the topology and component types of the L-shaped network to ensure that the impedance point can move to the target impedance point via the shortest path and in the most stable manner. By combining formula calculation with graphical optimization, this invention solves the matching forbidden zone problem of a single L-shaped network and can select the optimal matching parameters near key frequency points (such as 15 MHz) for dynamic impedance changes over a wide bandwidth. Experiments have shown that this significantly improves signal transmission and increases communication speed.

[0199] Step 5: Combine the composite bandpass filter obtained in Step 3 with the double L-type impedance matching network obtained in Step 4 to form a DC power line carrier communication impedance matching coupler, such as... Figure 1 As shown.

[0200] Example 1: This example designs an impedance matching coupler for the CR600 chip, following the process of Specific Implementation Method 1, specifically as follows:

[0201] This embodiment designs an impedance matching coupler for the CR600 chip, following the procedure of Specific Implementation Method 1:

[0202] Step 1: The design goal of the low-pass filter is: passband cutoff frequency. =28MHz, stopband cutoff frequency =33MHz (stopband start frequency), minimum stopband attenuation =40dB (attenuated to 40 dB at 33 MHz), maximum passband attenuation =2dB, port impedance Z=50Ω;

[0203] Obtain the normalized passband cutoff frequency according to the design objectives. and the normalized stopband start frequency =1.178; then calculate It is 0.765. It is 99.99. It is 0.00765;

[0204] Calculated and mold repair The values ​​are 0.848 and 0.529 respectively; according to the first kind of complete elliptic integral function get Since the order must be an integer, the value greater than 4.93 is selected. The fifth-order elliptic filter will contain two transmission zeros, which are precisely positioned at specific frequency points on the stopband edge using elliptic function theory to form a locally infinitely attenuated notch, thereby enhancing the stopband suppression capability.

[0205] Using Jacobi elliptic functions The pole and zero locations of the elliptic filter are obtained, and the five normalized poles are shown in Table 2.

[0206] Table 2. Calculation of poles for a fifth-order elliptic filter

[0207]

[0208] The real and imaginary parts of the zero point are shown in Table 3.

[0209] Table 3 Zero-point calculation for a fifth-order elliptic filter

[0210]

[0211] In this embodiment, the angular frequencies of the transmission zeros of the fifth-order elliptic low-pass filter are finally calculated as follows: ,

[0212] in, It is the angular frequency of the first transmission zero of the fifth-order elliptic low-pass filter. It is the angular frequency of the second transmission zero of the fifth-order elliptic low-pass filter;

[0213] A transfer function is constructed based on the angular frequency and poles of the transmission zero of a fifth-order elliptic low-pass filter, and normalized component parameter values ​​are obtained based on the transfer function.

[0214] For both ends The filter uses the reflection coefficient to obtain the normalized system input impedance:

[0215]

[0216] Again Expanding the continuous fraction yields:

[0217]

[0218] Using the fraction expansion Solve for the component values:

[0219] To reduce the number of inductors used, a capacitor is connected in parallel first. The first parallel capacitor: as s→∞, the input impedance is mainly due to… Decision. By extracting The behavior at infinity yields the normalized capacitance value of the first capacitor. :

[0220]

[0221] in, It's a Ferrari;

[0222] By expanding the extended fraction, the resonant frequency of the first resonant module is calculated as follows: :

[0223] ,

[0224] Solution Continue expanding the continuous fraction and extract... :

[0225]

[0226] Calculate the resonant frequency of the second resonant module. :

[0227] ,

[0228] in, It was Henry;

[0229] Finally extract :

[0230]

[0231] Convert the normalized passband cutoff frequency and the normalized stopband start frequency into actual frequencies. ;

[0232] The normalized component parameter values ​​are denormalized to 28MHz and 50Ω to obtain the actual component parameter values:

[0233]

[0234]

[0235]

[0236]

[0237]

[0238] The component parameter values ​​are shown in Table 4.

[0239] Table 4

[0240]

[0241] The topology of the fifth-order elliptic low-pass filter is constructed based on its transfer function, actual frequency, and component parameter values. Then, the single-ended structure is converted into a symmetrical balanced structure.

[0242] The filter was simulated using a network analyzer (S21 parameters), and the results are as follows: Figure 4 and Figure 5 As shown. The maximum ripple in the passband (0-28MHz) is 1.94 dB, the attenuation reaches -40 dB at 32.71 MHz, and after 46 MHz, due to the effect of the high-frequency suppression inductor, the attenuation stabilizes at about -50 dB, which fully meets the design specifications.

[0243] Step 2: Set the design target for the Chebyshev Type II high-pass filter: passband cutoff frequency. =2MHz, stopband cutoff frequency =800kHz, maximum passband ripple 1dB, minimum stopband attenuation 40dB, port impedance 50 Ω;

[0244] Let the passband cutoff frequency be set. Using the reference frequency, Obtain the normalized stopband start frequency ;

[0245] Utilizing minimum stopband attenuation Obtain the stopband ripple coefficient and utilize Obtaining the selection factor Using the normalized stopband start frequency The order of the Chebyshev Type II filter is approximately n'≈3.81, therefore ;

[0246] Obtain the transfer function of the Chebyshev Type II filter, and then obtain the zeros of the normalized low-pass prototype filter based on the transfer function of the Chebyshev Type II filter. =±j6.532, ±j2.7065 and the poles of the normalized low-pass prototype filter =−0.00112±j1.082、−0.0158±j2.617、

[0247] Using the zeros of the normalized low-pass prototype filter and normalized low-pass prototype filter poles Obtain the transfer function of the low-pass prototype filter, convert the low-pass prototype filter into a high-pass prototype filter, and obtain the transfer function of the high-pass prototype filter.

[0248] Obtain the final required component values ​​for the high-pass filter and obtain the nominal values ​​according to the preset error, as shown in Table 5.

[0249] Table 5

[0250]

[0251] Based on all components and their corresponding nominal values, and the high-pass filter, the topology of the fourth-order Chebyshev Type II high-pass filter is finally obtained. The topology of the fourth-order Chebyshev Type II high-pass filter is as follows: Figure 6 As shown, component values The value is 2.2nH, and the component value is... 30nH, component value The value is 5.6nH, component value The value is 6.2uH, component value The value is 2.7uH. Simulation results ( Figure 7The results show that at 2MHz, S21≈−1dB, the ripple in the passband (>2MHz) is <0.5dB, and the transition band slope is as high as 36.3dB / MHz, which meets the design requirements. Figure 6 In the circuit, the input terminals are connected together. The two ends of the inductor, with two specifications are One end of the capacitor; two The other ends of the capacitors converge to the first common node; the first common node is simultaneously connected to... One end of the inductor, One end of the capacitor has two specifications. One end of the capacitor, The other end of the inductor and The other ends of the capacitors converge at the second common node. The other end of the capacitor is connected to the output terminal of the circuit, respectively.

[0252] Step 3: Cascade the Chebyshev Type II high-pass filter topology with the fifth-order elliptic low-pass filter topology, and then cascade two 100 nH high-frequency suppression inductors at the output of the fifth-order elliptic low-pass filter topology to obtain a composite bandpass filter; In this invention, two 100 nH high-frequency suppression inductors are cascaded at the output of the fifth-order elliptic low-pass filter as follows... Figure 3 As shown, component values 270pF, component value 120pF, component value 300pF, component value 360pF, component value 200pF, component value 100nH, component value The value is 62nH; the composite bandpass filter topology is as follows: Figure 8 As shown. Simulation results after cascading ( Figure 9 The results show that the maximum ripple value is -2.5 dB in the 2-28 MHz passband range, while maintaining their respective steep transition band characteristics. Figure 3 In the middle, the left input terminal of the circuit is connected to two terminals with specifications of... One end of each inductor is connected in a one-to-one correspondence; the other ends of the two L0 inductors serve as the output terminals of the high-frequency suppression inductor module, connected to the input node of the filter network module. The two output nodes of the high-frequency suppression inductor module are respectively connected to... One end of the capacitor One end of the inductor, One end of the capacitor; The other end of the capacitor, The other end of the inductor, The other ends of the capacitors converge at the first common node, which is then connected to... One end of the capacitor One end of the inductor, One end of the capacitor; The other end of the capacitor, The other end of the inductor, The other ends of the capacitors converge at the second common node, which is then connected to... One end of the capacitor; The other end of the capacitor serves as the output of the filter network module and is connected to the corresponding output of the circuit.

[0253] This embodiment employs a fifth-order elliptic function topology, with a passband cutoff frequency of ( The stopband cutoff frequency is 28 MHz. The passband attenuation reaches 40 dB at 30 MHz. Its core advantage lies in its steep transition band, with an attenuation slope of approximately 19 dB / MHz from 28 MHz to 30 MHz, effectively suppressing high-frequency interference outside the passband. To improve high-frequency filtering performance, a high-frequency suppression inductor (e.g., 100 nH) is cascaded at the output of the low-pass filter. Utilizing its inductive reactance increasing with frequency, the overall suppression capability above 30 MHz is increased to approximately -50 dB without degrading the passband performance. A fourth-order Chebyshev II topology is adopted, with a passband cutoff frequency of 2 MHz and a stopband attenuation of 40 dB. Its core advantage is that it is strictly monotonic and ripple-free within the passband, avoiding the superposition of passband ripple after cascading with an elliptic low-pass filter, while achieving an attenuation slope as high as 36.3 dB / MHz. The cascaded composite bandpass filter has a maximum ripple of -2.5 dB within the 2-28 MHz passband, excellent insertion loss performance, and provides reliable frequency domain cleanup capability for the system.

[0254] Step 4: Obtain the required component types and values ​​for the double L-type impedance matching network, thereby constructing the double L-type impedance matching network;

[0255] Step 5: Combine the composite bandpass filter with the dual L-type impedance matching network to form a DC power line carrier communication impedance matching coupler.

[0256] Example 2: To verify the beneficial effects of the present invention, the following experiments were conducted:

[0257] 1. Match node 5 in the experimental platform (system input impedance is 7.2484−j3.9273Ω at 15 MHz) to the target impedance Z0=50Ω:

[0258] (1) Judgment of the situation: real part of load impedance =7.2484Ω, less than the target impedance =50Ω, therefore applicable (Matching strategy).

[0259] (2) Analytical calculation: using =7.2484, =−3.9273, =50 Calculation yields two sets of reactance values ​​(X1, X2);

[0260] (3) Smith chart selection:

[0261] load impedance =7.2484−j3.9273Ω is normalized and mapped on the Smith chart. Based on its position (falling in the capacitive region and with a magnitude much less than 1), Table 3 is consulted to determine the optimal matching strategy as "series inductor followed by parallel capacitor".

[0262] From the two sets of solutions calculated in step 2, select a set of parameters that conforms to this topology strategy.

[0263] (4) Component values: The calculated ideal values ​​are series inductance Lseries = 0.228 μH and parallel capacitor Cparallel = 515.4 pF. Based on the actual component library, the closest nominal values ​​are selected: series inductance 0.22 μH and parallel capacitor 510 pF.

[0264] II. System Integration and Experimental Verification:

[0265] Construction of an unmanned system experimental platform:

[0266] Topology: according to Figure 17 As shown in Table 6, a DC PLC network with one main branch and three sub-branches (A, B, C) was constructed. Power supply: A 24V DC power supply (battery simulation) was used, with a total system power exceeding 100W. Load: Diverse loads were configured according to Table 4, including a camera (3W), LED lights (5W, 7W), a permanent magnet synchronous fan (7.2W, 4.8W), and a variable resistor to simulate the working conditions of a real unmanned system. Cables: Shielded twisted-pair cable (RVSP-1.0) was used for branches A and B, while unshielded twisted-pair cable (RVS-1.0) was used for branch C for comparative analysis. PLC module: A PLC module based on the Clouder CR600 chip was used, with a maximum physical layer speed of 200 Mbps.

[0267] Table 6. Component Models and Parameters Used in the Experimental Platform

[0268]

[0269] The final unmanned autonomous system DC PLC verification platform, as shown in the load end... Figure 18 As shown, the transmission line wiring ends are as follows Figure 19 As shown. For ease of testing, the platform is laid out on three 1.5*1 m flat plates with a rotatable structure in the middle for connection. The cables used to transmit power and carrier signals are laid on one side of the flat plates, and the electrical loads are laid on the other side of the flat plates.

[0270] Impedance matching effect test:

[0271] Test objects: Select branch A (node ​​0 → node 10) with the longest signal transmission path and the most complex impedance change, and branch C (node ​​0 → node 8) with greater cable attenuation.

[0272] Test method:

[0273] The input impedance of each node at 15 MHz was measured using a vector network analyzer (as shown in Table 7). Following the procedure described in 2.2, the transmitter or receiver matching networks for nodes 0, 10, and 8 were designed and fabricated respectively. PLC modules were connected to the nodes, and the end-to-end communication rate was recorded using Nettest software on a PC. Simultaneously, the time-domain waveform and frequency-domain spectrum of the signal were observed using a high-precision oscilloscope (such as a Tektronix MSO56).

[0274] Table 7. System input impedance and corresponding matching parameters for some nodes.

[0275] Test results:

[0276] Branch A: such as Figure 20 As shown, the data rate is approximately 9 Mbps without matching; after matching only at the transmitter, the data rate increases to approximately 13 Mbps; and after matching at both the transmitter and receiver, the data rate further increases to approximately 15 Mbps. (Spectrum analysis) Figure 21 The results show that the matching network effectively improves channel attenuation.

[0277] Branch C: such as Figure 22 As shown, the data rate is approximately 4 Mbps without a match; after a match is achieved at the receiver, the data rate significantly increases to approximately 8 Mbps. (Spectrum analysis) Figure 23 This confirms that the receiver matching significantly reduces high-frequency noise.

[0278] Conclusion: Experimental data fully demonstrate that the dual L-type impedance matching network proposed in this invention can effectively improve the impedance mismatch problem in the DC PLC network of complex unmanned systems, and significantly improve the communication rate and stability.

Claims

1. A design method for an impedance matching coupler for DC power line carrier communication in an unmanned system, characterized in that... The specific process of the method is as follows: Step 1: Set the design goal of the elliptic low-pass filter. Based on the design goal, obtain the transfer function and actual frequency of the elliptic low-pass filter, thereby obtaining the topology of the elliptic low-pass filter. Step 2: Set the design goals for the Chebyshev Type II high-pass filter. Based on the design goals, obtain the transfer function and actual frequency of the Chebyshev Type II high-pass filter, thereby obtaining the actual topology of the Chebyshev Type II high-pass filter. Step 3: Cascade the Chebyshev Type II high-pass filter topology obtained in Step 2 with the elliptic low-pass filter topology obtained in Step 1. Then, cascade two high-frequency suppression inductors at the output of the elliptic low-pass filter topology to obtain a composite bandpass filter. Step 4: Obtain the required component types and values ​​for the double L-type impedance matching network, thereby constructing the double L-type impedance matching network; Step 5: Combine the composite bandpass filter obtained in Step 3 with the double L-type impedance matching network obtained in Step 4 to form a DC power line carrier communication impedance matching coupler.

2. The design method of an impedance matching coupler for DC power line carrier communication in an unmanned system according to claim 1, characterized in that: In step one, setting the design goal of the elliptic low-pass filter, obtaining the transfer function and actual frequency of the elliptic low-pass filter based on the design goal, and thus obtaining the topology of the elliptic low-pass filter, specifically involves: Step 11: Set the design goals for the elliptic low-pass filter; The design objectives of the elliptic low-pass filter include: passband cutoff frequency. Stopband start frequency Minimum stopband attenuation Maximum passband attenuation and port impedance ; Steps 1 and 2: Obtain the normalized passband cutoff frequency and normalized stopband start frequency according to the design objectives, specifically as follows: in, It is the normalized passband cutoff frequency. It is the normalized stopband start frequency; Step 13: Utilize the maximum passband attenuation Obtain the passband ripple coefficient Utilizing minimum stopband attenuation Obtain the stopband ripple coefficient and utilize and Obtain the selectivity factor of the filter : in, It is the passband ripple factor. It is the maximum attenuation in the passband. It is a logarithmic calculation with base 10. It is the minimum stopband attenuation. It is the stopband ripple factor; Step 1.4: Filter-based Selectivity Factor Obtain the minimum order of an elliptic low-pass filter. ,get A first-order low-pass filter, specifically: First, the reciprocal of the normalized stopband start frequency is used as the modulus k, and the modulus complement is obtained using the modulus k. Specifically: Then, using the fixed modulus k, we construct the first kind of complete elliptic integral K(k): in, It is the curvature of an ellipse; Finally, using the modulus , and selection factor Obtain the minimum order of the elliptic low-pass filter. ,get First-order low-pass filter; Among them, the minimum order n of the elliptic low-pass filter is greater than or equal to The smallest positive integer; Step 15: Obtain using the Jacobian elliptic function The pole and zero angular frequencies of the elliptic low-pass filter are as follows: First, construct the first type of elliptic integral function; Then, construct the Jacobian elliptic function based on the first type of elliptic integral function; Finally, the real part of the pole locations of the elliptic filter is obtained using the Jacobian elliptic function. and the virtual part The real part at the zero point and the virtual part And using the real part of the zero point position and the virtual part Obtaining the angular frequency at zero point ; Step 16, Utilize The transfer function is constructed from the angular frequencies and poles of the zeros of a first-order low-pass filter. Step 17: Obtain normalized component parameter values ​​based on the transfer function; Step 18: Convert the normalized passband cutoff frequency and normalized stopband start frequency obtained in Step 12 into actual frequencies, specifically as follows: in, Pick or , It is the actual frequency after conversion; Step 19: Inverse normalize the normalized component parameter values ​​obtained in Step 17 to the passband cutoff frequency. By determining the port impedance Z, the actual values ​​of the component parameters can be obtained. Step 10: Construct a system based on the transfer function obtained in Step 16, the actual frequency obtained in Step 18, and the actual values ​​of the component parameters obtained in Step 19. The topology of the elliptic low-pass filter is derived, and the single-ended structure is transformed into a symmetric balanced structure.

3. The design method of an impedance matching coupler for DC power line carrier communication in an unmanned system according to claim 2, characterized in that: The utilization in step one six The transfer function is constructed using the angular frequencies of the zeros and poles of the first-order low-pass filter. Specifically: in, It is a complex frequency. yes The total number of poles of an elliptic low-pass filter. yes Pole labels of an elliptic low-pass filter. yes The angular frequency of the first zero of the elliptic low-pass filter. yes The angular frequency of the second zero of the elliptic low-pass filter. yes The first elliptic low-pass filter The real part of each pole yes The first elliptic low-pass filter The imaginary part of each pole yes The first elliptic low-pass filter The real part of each pole yes The first elliptic low-pass filter The imaginary part of each pole yes The transfer function of a first-order low-pass filter.

4. The design method of an impedance matching coupler for DC power line carrier communication in an unmanned system according to claim 3, characterized in that: The step 17, obtaining the normalized component parameter values ​​based on the transfer function, specifically involves: First, the reflection coefficient is obtained using the transfer function, and then the system input impedance is calculated using the reflection coefficient: in, It is the reflection coefficient. It is the normalized port impedance. It is the system input impedance; Then, to Expanding the continuous fractions, specifically: in, , , , and It is the capacitance value. and It is the inductance value; Finally, using the fraction expansion Solve for the normalized component values , , , , , and ; in, It is after normalization , It is after normalization , It is after normalization , It is after normalization , It is after normalization , It is after normalization , It is after normalization .

5. The design method of an impedance matching coupler for DC power line carrier communication in an unmanned system according to claim 4, characterized in that: In step 19, the normalized component parameter values ​​obtained in step 17 are inversely normalized to the passband cutoff frequency. By considering the port impedance Z, the actual values ​​of the component parameters can be obtained, specifically: 。 6. The design method of an impedance matching coupler for DC power line carrier communication in an unmanned system according to claim 5, characterized in that: Step two involves setting the design goals for the Chebyshev II high-pass filter. Based on these goals, the transfer function and actual frequency of the Chebyshev II high-pass filter are obtained, thus leading to the actual topology of the filter. Specifically: Step 2:

1. Set the design objectives for the Chebyshev Type II high-pass filter; The design objectives of the Chebyshev Type II high-pass filter include: passband cutoff frequency. Stopband cutoff frequency Maximum ripple in passband Minimum stopband attenuation and port impedance ; Step 22: Set the passband cutoff frequency As a reference frequency, using Obtain the normalized stopband start frequency : in, It is the normalized stopband start frequency. It is the stopband cutoff frequency. It is the normalized passband cutoff frequency; Steps two and three: Utilizing the minimum stopband attenuation Obtain the stopband ripple coefficient Specifically: in, It is the stopband ripple factor. It is the minimum stopband attenuation; Step 24: Use the normalized stopband start frequency obtained in Step 22. Obtaining the order of a Chebyshev Type II filter ; The order of the Chebyshev Type II filter It is greater than The smallest positive integer; Step 25: Utilizing the order of the Chebyshev Type II filter Obtain the zeros of the normalized low-pass prototype filter. and normalized low-pass prototype filter poles and utilize and The transfer function for constructing the low-pass prototype filter is as follows: First, using the order of the Chebyshev Type II filter Obtain the zeros of the normalized low-pass prototype filter. and normalized low-pass prototype filter poles This allows us to obtain the angular frequencies of the zeros of the normalized low-pass prototype filter and the real and imaginary parts of the poles of the normalized low-pass prototype filter, specifically: in, These are the parameters for the ellipse shape. It is the real contribution item at the pole. It is the contribution term of the imaginary part of the pole. It is the angular frequency at zero point. It is the real part of the pole. It is the imaginary part of the pole; Then, the transfer function of the low-pass prototype filter is constructed using the angular frequencies of the zeros of the normalized low-pass prototype filter and the real and imaginary parts of the poles of the normalized low-pass prototype filter, specifically: in, These are pole labels. It is the first The imaginary part of each pole It is the first The real part of each pole It is a complex frequency. It is the first angular frequency at zero point, It is the total number of poles. It is the transfer function of the low-pass prototype filter; Step 26: Convert the low-pass prototype filter into a high-pass prototype filter to obtain the transfer function of the high-pass prototype filter. ; Step 27: Utilizing the transfer function of the high-pass prototype filter Obtain the normalized component parameter values ​​of the Chebyshev Type II filter, convert the normalized passband cutoff frequency and normalized stopband start frequency to the actual frequencies, and then inversely normalize the normalized component parameter values ​​back to the passband cutoff frequency. and port impedance To obtain the actual values ​​of component parameters; Step 28: Based on the transfer function of the high-pass prototype filter obtained in Step 26. The actual frequency and actual values ​​of component parameters obtained in step 27 are used to construct the topology of the Chebyshev Type II filter.

7. The design method of an impedance matching coupler for DC power line carrier communication in an unmanned system according to claim 6, characterized in that: The transfer function of the high-pass prototype filter in step two-six. Specifically: in, It is the complex frequency variable of the normalized high-pass prototype filter. It is the complex frequency variable in the target high-pass prototype filter. It is an intermediate variable. It is the minimum value of the chip carrier wave.

8. The design method of an impedance matching coupler for DC power line carrier communication in an unmanned system according to claim 7, characterized in that: Step four, which involves obtaining the required component types and values ​​for the dual L-type impedance matching network to construct the network, specifically includes: Step 41: Obtain the target matching impedance According to the target matching impedance With the real part of the load impedance The relationship between the components in the double L-type impedance matching network is obtained, along with the corresponding component values ​​for each connection relationship. when Greater than If yes, then execute a; otherwise, execute b. a. Place the series-connected matching element X2 at the load end, and then obtain the element value, specifically: First, obtain the component reactance. Component reactance Target matching impedance after using an L-type impedance matching network The relationship is as follows: in, It is the target matching impedance after using an L-type impedance matching network. It is the imaginary unit. , It is the reactance of two components. It is the real part of the load impedance. It is the imaginary part of the load impedance; Then, the component reactance is obtained. With component reactance Specifically: b. Connect the matched elements in series Place it at the source end, and then obtain the component value, specifically: First, obtain the component reactance. Component reactance Target matching impedance after using an L-type impedance matching network The relationship is as follows: Then, the component reactance is obtained. With component reactance Specifically: Step 4.2: Optimize the circuit topology of the double L-type impedance matching network using the Smith chart.

9. The design method of an impedance matching coupler for DC power line carrier communication in an unmanned system according to claim 8, characterized in that: The circuit topology optimization of the double L-type impedance matching network using the Smith chart in step four-two is specifically as follows: B1. Utilizing the target matching impedance and load impedance To obtain the reflection coefficient, specifically: in, It is the normalized load impedance. It is the reflection coefficient. It is the load impedance; B2. After performing a complex-plane mathematical transformation on the normalized load impedance, map the transformed normalized load impedance onto the Smith chart, and then calculate the reflection coefficient. Mapping onto the Smith chart to construct the Smith circle chart; When the normalized load impedance after mathematical transformation of the complex plane is mapped onto the Smith chart, the horizontal diameter axis of the Smith chart represents a purely resistive line, and the impedance value mapped to the point on the purely resistive line has only the real part; the impedance value mapped to the point on the outermost circle of the Smith chart has only the imaginary part; when the load impedance is equal to the chip impedance, the point mapped by the normalized load impedance is located at the center of the Smith chart. When the reflection coefficient is mapped onto the Smith chart, the outermost circle of the Smith chart represents a reflection coefficient modulus of 1. The point where the outermost circle of the Smith chart intersects with the horizontal diameter axis and the vertical diameter axis of the Smith chart represents a reflection coefficient with only a real part or an imaginary part. The point where the mapped reflection coefficient is 0 is the center of the Smith chart. B3. Divide the Smith chart into 8 regions. Based on the position of the impedance to be matched at the load end on the Smith chart, obtain the topology of the optimized double L-type impedance matching network.

10. The design method of an impedance matching coupler for DC power line carrier communication in an unmanned system according to claim 9, characterized in that: In section B3, the Smith chart is divided into 8 regions. Based on the position of the impedance to be matched at the load end mapped onto the Smith chart, the optimized topology of the double L-type impedance matching network is obtained, specifically as follows: B3-1. Divide the Smith chart into 8 regions, specifically: Region 1 is the intersection of the region above the horizontal diameter axis and the circle of equal conductance. The point magnitude in Region 1 is less than the reference impedance, and the inductive and real part of the admittance are greater than the reference admittance. Region 2 is the region above the horizontal diameter axis and the region to the left of the vertical axis, excluding the region of the circle with equal conductance. The point magnitude in Region 2 is less than the reference impedance, and the real part of the inductive admittance is less than the reference admittance. Region 3 is the region above the horizontal diameter axis and the region to the right of the vertical axis, excluding the region of the circle of equal resistance. The point magnitude in Region 3 is greater than the reference impedance, it is inductive, and the real part of the impedance is less than the reference impedance. Region 4 is the intersection of the region above the horizontal diameter axis and the circle of equal resistance. The point magnitude in Region 4 is greater than the reference impedance, and it is inductive, with the real part of the impedance being greater than the reference impedance. Region 5 is the intersection of the region below the horizontal diameter axis and the circle of equal resistance. The point magnitude in region 5 is greater than the reference impedance, and the real part of the capacitance is greater than the reference impedance. Region 6 is the region below the horizontal diameter axis and the region intersecting the right side of the vertical axis, excluding the region of the circle of equal resistance. The point magnitude in Region 6 is greater than the reference impedance, capacitive, and the real part of the impedance is less than the reference impedance. Region 7 is the region below the horizontal diameter axis and the region to the left of the vertical axis, excluding the region of the circle with equal conductance. The point magnitude in Region 7 is less than the reference impedance, and the real part of the capacitance and admittance is less than the reference admittance. Region 8 is the intersection of the region below the horizontal diameter axis and the region of the circle with equal conductance. The point magnitude in region 8 is less than the reference impedance, and the real part of the capacitive admittance is greater than the reference admittance. B3-2. Obtain the location of the impedance to be matched in the Smith chart. According to the Smith chart matching strategy, after adding the first matching device, the complex impedance point moves to the standard constant conductance circle or constant resistance circle. After adding the second device, the complex impedance point moves to the center of the Smith chart along the standard equal conductance circle or equal resistance circle, thus obtaining the optimized topology of the double L-type impedance matching network. The Smith chart matching strategy is as follows: When the location of the impedance to be matched in the Smith circle is region 1, the L-type impedance matching network at the transmitting end is connected to a parallel capacitor after a series inductor, and the L-type impedance matching network at the receiving end is connected to a series capacitor after a parallel inductor. When the impedance to be matched is mapped to the Smith circle, the region is region 2. The transmitter L-type impedance matching network is connected in series with a capacitor and then in parallel with a capacitor. The receiver L-type impedance matching network is connected in parallel with an inductor and then in series with an inductor. When the impedance to be matched is mapped to the Smith circle and the region is region 3, the L-type impedance matching network at the transmitting end is connected in parallel with a capacitor followed by a series capacitor, and the L-type impedance matching network at the receiving end is connected in series with an inductor followed by a parallel capacitor. When the impedance to be matched is mapped to the Smith circle, the region is region 4. The L-type impedance matching network at the transmitting end is connected to a series capacitor after a parallel inductor, and the L-type impedance matching network at the receiving end is connected to a parallel capacitor after a series inductor. When the impedance to be matched is mapped to the Smith circle and the region is region 5, the L-type impedance matching network at the transmitting end is connected to a series inductor after a parallel capacitor, and the L-type impedance matching network at the receiving end is connected to a parallel inductor after a series capacitor. When the impedance to be matched is mapped to the Smith circle and the region is region 6, the L-type impedance matching network at the transmitting end is connected in parallel with an inductor and then in series with an inductor, and the L-type impedance matching network at the receiving end is connected in series with a capacitor and then in parallel with a capacitor. When the impedance to be matched is mapped to the Smith circle, the region is region 7. The transmitter L-type impedance matching network is connected in series with an inductor and then in parallel with an inductor. The receiver L-type impedance matching network is connected in parallel with a capacitor and then in series with a capacitor. When the impedance to be matched is mapped to region 8 in the Smith circle, the L-type impedance matching network at the transmitting end is connected to a parallel inductor after a series capacitor, and the L-type impedance matching network at the receiving end is connected to a series inductor after a parallel capacitor.

Images