WO2021147628A1

WO2021147628A1 - Method for designing low group-delay fluctuation filter

Info

Publication number: WO2021147628A1
Application number: PCT/CN2020/140942
Authority: WO
Inventors: 庞慰; 徐利军
Original assignee: 诺思(天津)微系统有限责任公司
Priority date: 2020-01-20
Filing date: 2020-12-29
Publication date: 2021-07-29
Also published as: CN111224634B; CN111224634A

Abstract

The present invention relates to the technical field of filters, and particularly relates to a method for designing a low group-delay fluctuation filter, comprising the following steps: according to a filter index, designing a Gaussian low-pass filter, and obtaining the inductance and capacitance values of the Gaussian low-pass filter; converting the low-pass filter to a band-pass filter; connecting a capacitor in parallel with each LC series resonant circuit to form a BVD circuit model; connecting a capacitor in series to each LC parallel resonant circuit so as to convert the parallel branch into a BVD circuit model; taking the capacitance value and inductance value in the BVD circuit model as the initial values of a target optimization function, by means of an iterative method, causing the objective optimization function to approach zero, or reach a minimum value, and recording the capacitance and inductance values at that time; calculating each resonator parameter according to the obtained capacitance value and inductance value. According to the filter obtained by the present design method, as long as performance is constant and there is no serious deterioration, the group-delay fluctuation is significantly reduced.

Description

A Design Method of Low Group Delay Wave Filter

Technical field

The present invention relates to the technical field of filters, in particular to a design method of low group delay fluctuation filters.

Background technique

With the rapid development of mobile communication technology, many radio frequency devices have been widely used in the communication field. For example, there is a large demand for filters and duplexers on mobile terminals such as mobile phones and notebook computers, which are mainly used to filter out unnecessary The radio frequency signal, improve the communication quality, and enhance the user experience. Due to the inherent characteristics of mobile communication terminals, such as portability, lightness and thinness, in addition to higher performance requirements for filters and duplexers, higher requirements are also imposed on the volume and size. The acoustic wave filter based on the principle of mechanical acoustic wave can just meet its requirements. The resonators that constitute this type of acoustic wave filter mainly include: bulk acoustic wave resonator (Film Bulk Acoustic Resonator, FBAR) and surface acoustic wave resonator (Surface Acoustic Wave, SAW) Among them, the bulk acoustic wave resonator, compared with the surface acoustic wave resonator, has a higher Q value and a higher operating frequency, which has attracted the attention of the industry. The bulk acoustic wave resonator uses the piezoelectric effect of a piezoelectric crystal to generate resonance. Since resonance is generated by mechanical waves instead of electromagnetic waves as the source of resonance, the wavelength of mechanical waves is much shorter than that of electromagnetic waves. Therefore, the volume of the bulk acoustic wave resonator and the filter composed of it is greatly reduced compared with the traditional electromagnetic filter. On the other hand, since the crystal orientation growth of the piezoelectric crystal can be well controlled at present, the loss of the resonator is extremely small, the quality factor is high, and it can cope with complex design requirements such as steep transition band and low insertion loss. Due to the small size, high roll-off, and low insertion loss of the bulk acoustic wave filter, the filter with this as the core has been widely used in communication systems.

Another important trend in the development of communication terminals is higher data transmission volume. Whether it is Wi-Fi communication based on WLAN protocol or LTE communication based on 3GPP protocol, the data transmission volume has taken an exponential leap compared with the 3G era. There are two main ways to increase data transmission: one is to achieve channel expansion through carrier aggregation, but this requires major upgrades and changes to the terminal hardware and software design, and the achieved bandwidth expansion is up to 2-3 times. Second, by improving the modulation and demodulation efficiency, from the early debugging of BPSK and QPSK, to 16-QAM, 64-QAM, and even 256-QAM with large data transmission volume, the data transmission rate can theoretically be increased by hundreds of times. However, the improvement of the modulation mode also puts forward higher requirements on the performance of the communication channel. The channel quality is not high, which will directly affect the EVM characteristics, resulting in data transmission rate degradation and signal distortion. For filters, fluctuations in group delay will affect the data transmission rate and signal fidelity.

Summary of the invention

In view of this, the main purpose of the present invention is to provide a method for designing a low group delay fluctuation filter, which ensures that the filter performance remains unchanged or is not seriously deteriorated, so that it has a smaller group delay fluctuation.

To achieve the above objective, according to one aspect of the present invention, a method for designing a low group delay fluctuation filter is provided, which includes the following steps:

Step 1: According to the filter index, design a Gaussian low-pass filter to obtain the inductance and capacitance values of the Gaussian low-pass filter;

Step 2: Convert the low-pass filter into a band-pass filter, where the inductance in the low-pass filter is transformed into an LC series resonant circuit in the band-pass filter, and the capacitor in the low-pass filter is transformed into a band-pass filter The LC parallel resonant circuit in

Step 3: Connect a capacitor in parallel with each LC series resonance circuit to form a BVD circuit model; connect a capacitor in series with each LC parallel resonance circuit to convert the parallel branch into a BVD circuit model;

Step 4: Take the capacitance value and inductance value in the BVD circuit model as the initial value of the objective optimization function, and through the iterative method, make the objective optimization function approach zero or reach the minimum value, and record the capacitance value and inductance value at this time;

Step 5: Calculate each resonator parameter based on the capacitance value and inductance value obtained in step 4, and replace the BVD circuit model with a bulk acoustic wave resonator with this parameter, thereby constructing a low group delay fluctuation filter.

Optionally, in the step 1: the inductance value and the capacitance value of the Gaussian low-pass filter are obtained by the root of the transfer function, and the transfer function is:

Among them, ω _c is the cutoff frequency.

Optionally, in the step 2: the conversion formula for converting the low-pass filter into the band-pass filter is:

Among them, ω ₀ is the center frequency, and Δω is the filter bandwidth.

Optionally, in the step 3: the conversion formula for converting the parallel branch into the BVD circuit model is:

Li1=Li(1+β _i ) ² i=6,...9

Ci1=Ci/(1+β _i ) i=6,...9

C0i1=C0i/(1+β _i ) i=6,...9

Among them, C0i is the capacitor in parallel on the LC series resonant circuit or the capacitor in series on the LC parallel resonant circuit; Ci is the capacitor in the LC series resonant circuit or the capacitor in the LC parallel resonant circuit; Li is the inductance or in the LC series resonant circuit The inductance in the LC parallel resonant circuit; Li1 is the inductance in the parallel branch after being transformed into the BVD circuit model; Ci1 is the parallel capacitance in the parallel branch after being transformed into the BVD circuit model; C0i1 is the parallel branch after being transformed into the BVD circuit model Series capacitors in.

Optionally, in the step 4: the objective optimization function is:

S(ω)=A[H(ω)-H _d (ω)] ² +B[D(ω)-D _d (ω)] ²

Among them, H(ω) is the amplitude-frequency function of the filter, D(ω) is the group delay function of the filter; H _d (ω) is the amplitude-frequency objective function of the filter, and D _d (ω) is the filter’s Group delay objective function.

Optionally, in the step 4: using a Newton iteration method or a genetic algorithm, the objective optimization function is approached to zero or reaches the minimum value.

Optionally, in the step 5: the resonator parameters include the effective electromechanical coupling coefficient

Resonator area A, series resonance frequency f _s .

The design method in the technical scheme of the present invention is based on a Gaussian prototype filter, and a filter with low group delay fluctuation is obtained through optimization iteration; the group delay fluctuation of the filter is obvious when the performance of the filter remains unchanged or is not severely deteriorated. The reduction.

Description of the drawings

For purposes of illustration and not limitation, the present invention will now be described according to preferred embodiments of the present invention, particularly with reference to the accompanying drawings, in which:

FIG. 1 is a flowchart of a design method of a low group delay fluctuation filter provided by this embodiment;

Figure 2 is a comparison diagram of group delay fluctuations between an elliptic function filter and a Gaussian filter;

Figure 3a is a circuit diagram of the inductance in the low-pass filter transformed into the LC series resonant circuit in the band-pass filter;

Fig. 3b is a circuit diagram of a capacitor in a low-pass filter transformed into an LC parallel resonant circuit in a band-pass filter;

Figure 4 is a schematic diagram of the topology of a bandpass filter;

Figure 5 is a schematic diagram of the topological structure of the band-pass filter after correction;

Figure 6 is a BVD equivalent circuit diagram;

Figure 7 is a topological structure diagram of the band-pass filter after conversion;

Figure 8 is a topological structure diagram of a ladder filter composed of BVD;

Figure 9 shows the parameters of each resonator of the low group delay fluctuation filter designed by this method;

FIG. 10 is a comparison diagram of group delay fluctuations between a traditional filter and the filter designed by the method of this embodiment;

11 is a comparison diagram of the passband insertion loss of the traditional filter and the filter designed by the method of this embodiment;

Fig. 12 is a diagram of out-of-band suppression in the 0-3 GHz range of the filter designed by the method of this embodiment.

Detailed ways

Fig. 1 is a flow chart of a design method of a low group delay fluctuation filter provided by this embodiment. As shown in Fig. 1, the method includes the following steps:

S1: According to the filter index, design a Gaussian low-pass filter, and its transfer function is:

Among them, ω _c is the cut-off frequency; the inductance and capacitance values of the Gaussian low-pass filter are obtained through the root of the transfer function;

Compared with the existing elliptic function filter, the Gaussian function filter has a smaller group delay fluctuation; Figure 2 is a comparison diagram of the group delay of the elliptic function filter and the Gaussian filter, as shown in Figure 2. , The solid line is the fluctuation range of the Gaussian filter, and the dashed line is the fluctuation range of the elliptic function filter. As shown in Figure 2, the group delay fluctuation range of the elliptic function filter (about 14ns) is much larger than that of the Gaussian filter. The fluctuation range (about 4ns).

S2: Convert the low-pass filter into a band-pass filter, the conversion formula is:

Among them, ω ₀ is the center frequency, and Δω is the filter bandwidth. Figures 3a and 3b are schematic diagrams of converting a low-pass filter into a band-pass filter. In Figure 3a, the inductance in the low-pass filter is transformed into an LC series resonant circuit in the band-pass filter. The low-pass filter in Figure 3b The capacitor in the band-pass filter is transformed into an LC parallel resonant circuit in the bandpass filter; the above-mentioned LC series-parallel resonant circuit is composed of a ladder structure as shown in Figure 4.

S3: Connect a capacitor in parallel with each LC series resonant circuit, and connect a capacitor in series with each LC parallel resonant circuit to obtain the corrected bandpass filter as shown in Figure 5. Figure 6 is the BVD equivalent circuit. Figure 6 shows that in the circuit shown in Figure 5, each level of the series branch is already a BVD circuit model, while the parallel branch is not a BVD circuit model, and conversion is required to convert the parallel branch into a BVD circuit model. The formula is:

Li1=Li(1+β _i ) ² i=6,...9

Ci1=Ci/(1+β _i ) i=6,...9

C0i1=C0i/(1+β _i ) i=6,...9

Among them, C0i is the capacitor connected in parallel on the LC series resonant circuit or the capacitor connected in series on the LC parallel resonant circuit; Ci is the capacitor in the LC series resonant circuit or the capacitor in the LC parallel resonant circuit; L1 is the capacitor in the LC series resonant circuit in Figure 5 Li1 is the inductance in the parallel branch in Figure 7; Ci1 is the parallel capacitance in the parallel branch in Figure 7; C0i1 is the series capacitance in the parallel branch in Figure 7.

Finally, the converted band-pass filter structure is obtained, as shown in Figure 7; the resonant circuit of each stage of the series branch and the resonant circuit of the parallel branch have been converted into a BVD circuit model, so the corresponding bulk acoustic wave resonance can be further used Instead of the device, the circuit diagram shown in Figure 8 is constructed.

S4: The design objective function is:

S(ω)=A[H(ω)-H _d (ω)] ² +B[D(ω)-D _d (ω)] ²

Among them, H(ω) is the amplitude-frequency function of the filter, D(ω) is the group delay function of the filter; H _d (ω) is the amplitude-frequency objective function of the filter, and D _d (ω) is the filter’s Group delay objective function; the capacitance value and inductance value in the BVD circuit model are used as the initial value of the objective optimization function, and the objective optimization function approaches zero through an iterative method, preferably, Newton iteration method, genetic algorithm, etc. , Or reach the minimum value, record the capacitance value and inductance value at this time;

S5: Calculate each resonator parameter based on the capacitance value and inductance value obtained in step S4, such as the effective electromechanical coupling coefficient

Resonator area A, series resonance frequency f _s, etc., using a bulk acoustic wave resonator with these parameters to replace the BVD circuit model, thereby constructing a low group delay wave filter.

Figure 9 shows the parameters of each resonator of the low group delay wave filter designed by this method. The filter includes 5 series resonators S11, S12, S13, S14 and S15, and 4 parallel resonators P11, P12, P13 and P14, the resonator parameters include series resonance frequency f _s , resonator area A and effective electromechanical coupling coefficient

Figure 10 is a comparison diagram of fluctuations between a traditional filter and the filter designed by the method of this embodiment. As shown in Figure 10, the dashed line is the fluctuation range of the traditional filter, and the solid line is the fluctuation of the filter designed by the method of this embodiment. The fluctuation range of the broken line is about 20 ns, and the fluctuation range of the solid line is about 6 ns. It can be seen that the group delay fluctuation of the filter designed by the method of this embodiment is significantly improved compared with the prior art.

FIG. 11 is a comparison diagram of the passband insertion loss of the traditional filter and the filter designed by the method of this embodiment. As shown in FIG. 11, the solid line is the insertion loss of the filter designed by the method of this embodiment, and the dashed line is the traditional The insertion loss of the filter, the insertion loss of some frequency bands in the passband of the filter designed by the method of this embodiment is slightly deteriorated. FIG. 12 is a diagram of out-of-band suppression of the filter designed by the method of this embodiment in the range of 0-3 GHz. As shown in FIG. 12, the out-of-band suppression of the filter is greater than 40 dB.

From this, it can be seen that the filter obtained by the method of this embodiment can significantly improve its group delay fluctuation under the condition that the performance is not severely deteriorated.

The foregoing specific implementations do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that, depending on design requirements and other factors, various modifications, combinations, sub-combinations, and substitutions can occur. Any modification, equivalent replacement and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

A design method of low group delay fluctuation filter is characterized in that it comprises the following steps:

Step 1: According to the filter index, design a Gaussian low-pass filter to obtain the inductance and capacitance values of the Gaussian low-pass filter;

Step 2: Convert the low-pass filter into a band-pass filter, where the inductance in the low-pass filter is transformed into an LC series resonant circuit in the band-pass filter, and the capacitor in the low-pass filter is transformed into a band-pass filter The LC parallel resonant circuit in

Step 3: Connect a capacitor in parallel with each LC series resonance circuit to form a BVD circuit model; connect a capacitor in series with each LC parallel resonance circuit to convert the parallel branch into a BVD circuit model;

Step 4: Take the capacitance value and inductance value in the BVD circuit model as the initial value of the objective optimization function, and through iterative methods, make the objective optimization function approach zero or reach the minimum value, and record the capacitance and inductance values at this time;

Step 5: Calculate each resonator parameter based on the capacitance value and inductance value obtained in step 4, and replace the BVD circuit model with a bulk acoustic wave resonator with this parameter, thereby constructing a low group delay fluctuation filter.
The design method of a low group delay fluctuation filter according to claim 1, wherein, in the step 1:

Obtain the inductance and capacitance values of the Gaussian low-pass filter through the roots of the transfer function. The transfer function is:

Among them, ω c is the cutoff frequency.
The design method of a low group delay fluctuation filter according to claim 1, wherein, in the step 2:

The conversion formula for converting a low-pass filter into a band-pass filter is:

Among them, ω 0 is the center frequency, and Δω is the filter bandwidth.
The method for designing a low group delay fluctuation filter according to claim 1, wherein in the step 3:

The conversion formula to convert the parallel branch into the BVD circuit model is:

Li1=Li(1+β i ) 2 i=6,...9

Ci1=Ci/(1+β i ) i=6,...9

C0i1=C0i/(1+β i ) i=6,...9

Among them, C0i is the capacitor in parallel on the LC series resonant circuit or the capacitor in series on the LC parallel resonant circuit; Ci is the capacitor in the LC series resonant circuit or the capacitor in the LC parallel resonant circuit; Li is the inductance or in the LC series resonant circuit The inductance in the LC parallel resonant circuit; Li1 is the inductance in the parallel branch after being transformed into the BVD circuit model; Ci1 is the parallel capacitance in the parallel branch after being transformed into the BVD circuit model; C0i1 is the parallel branch after being transformed into the BVD circuit model Series capacitors in.
The method for designing a low group delay fluctuation filter according to claim 1, wherein in the step 4:

The objective optimization function is:

S(ω)=A[H(ω)-H d (ω)] 2 +B[D(ω)-D d (ω)] 2

Among them, H(ω) is the amplitude-frequency function of the filter, D(ω) is the group delay function of the filter; H d (ω) is the amplitude-frequency objective function of the filter, and D d (ω) is the filter’s Group delay objective function.
The method for designing a low group delay fluctuation filter according to claim 1, wherein in the step 4:

Through Newton iteration or genetic algorithm methods, the objective optimization function is approached to zero or reaches the minimum value.
The design method of a low group delay fluctuation filter according to claim 1, wherein in the step 5:

Resonator parameters include effective electromechanical coupling coefficient
Resonator area A, series resonance frequency f s .