Split half-lattice microacoustic filter with wide bandwidth using phase shifter
By adopting a semi-lattice filter structure in the RF filter, combining a micro-acoustic resonator and a 180° phase shifter, the existing ladder filters are solved inadequate performance under the broadband transmission standard, wide passband and good out-of-band attenuation are achieved, and the manufacturing process is simplified.
Patent Information
- Application Number
- CN201980080140.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2018-12-05
- Filing Date
- 2019-11-27
- Publication Date
- 2025-03-28
- Estimated Expiration
- 2039-11-27
AI Technical Summary
When existing ladder RF filters face the requirements of broadband transmission standards such as 5G communication standards, it is difficult to achieve wide passband and good out-of-band attenuation, and are manufactured in a complex manner.
A semi-lattice filter structure is employed, including a first signal path and a second signal path, at least one signal path includes a micro-acoustic resonator and a phase shifter configured to perform a phase shift of 180° or about 180° to achieve wide passband and good out-of-band attenuation.
It achieves wide passband, steep filter skirt and good out-of-band attenuation, meeting the requirements of broadband transmission standards such as 5G communications, while simplifying the manufacturing process.
Smart Images

Figure CN113169727B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to RF filters. In particular, the present disclosure relates to an RF filter including a microacoustic resonator. Background Art
[0002] RF filters are widely used in electronic devices to select a desired signal bandwidth from a received signal. RF filters in communication devices covering the transmission bands currently in service are usually composed of surface acoustic wave resonators or bulk acoustic wave resonators in a ladder structure.
[0003] In ladder-type RF filters, the bandwidth is primarily determined by the distance between the resonant and antiresonant frequencies, or the pole-zero distance, which is primarily determined by the coupling factor of the piezoelectric substrate of the resonator. The use of passive components such as inductors in the ladder structure may increase the bandwidth, but at the expense of overall filter performance, such as poor out-of-band attenuation. Other approaches involve scandium doping of the aluminum nitride piezoelectric layer to increase acoustic coupling, which improves the bandwidth of the ladder filter. However, the fabrication of aluminum-scandium-nitride piezoelectric layers is complex, and the amount of scandium in the aluminum nitride is physically limited to about 40%.
[0004] Other dopants of aluminum nitride or other piezoelectric materials or substrates may increase the physical bandwidth and pole-zero distance, where there are still theoretical limits.
[0005] Future broadband transmission standards such as the 5G communication standard require a wide passband of the filter with good out-of-band attenuation because other services may be close to the 5G band. For example, the RF filter for the n79 band requires a passband width of 600MHz from 4.4GHz to 5.0GHz, where the WLAN service is located just above the passband, so that when these two services should be covered by the communication device at the same time, the filter must cover a wider passband and exhibit good out-of-band attenuation. These requirements may be contradictory for the ladder filter structure, so even with the above method, it may be difficult to meet all requirements using the current ladder design. Another well-known example is the n77 band, which has a bandwidth spanning 900MHz from 3.0GHz to 4.2GHz, and also has the nearby selectivity required to achieve the common band of 4G and 5G systems. Therefore, a micro-acoustic RF filter that provides a wide passband and good out-of-band attenuation is needed. The RF filter should be less complex so that it can be directly manufactured.
[0006] The object of the present disclosure is to provide a new structure for microacoustic RF filters to achieve a wide passband, steep filter skirt and good out-of-band attenuation.
[0007] Another object of the present disclosure is to provide such a microacoustic RF filter that can be manufactured with currently available technology of surface acoustic wave (SAW) and bulk acoustic wave (BAW) or film bulk acoustic wave resonator (FBAR) type resonators.
[0008] It is yet another object of the present disclosure to provide a less complex microacoustic RF filter. Summary of the invention
[0009] According to the present disclosure, one or more of the above objects are achieved by a microacoustic RF filter comprising the features of claim 1 of the present invention.
[0010] According to an embodiment of the present disclosure, the first signal path and the second signal path are coupled between the first port and the second port. Each signal path includes a resonator. One of the first signal path and the second signal path also includes a phase shifter connected in series with the resonator. Such a circuit structure adopts a half-lattice filter. At least one resonator is a microacoustic resonator, wherein the other resonator can be a microacoustic resonator or a capacitor. In the case where the resonator includes a microacoustic resonator and a capacitor, spurious modes or other losses that may affect or contaminate the passband behavior can be reduced. When the spurious modes and / or losses of the microacoustic resonator affect the passband, the microacoustic resonator should be replaced with a capacitor.
[0011] According to an embodiment, the phase shifter is configured to perform a phase shift of 180° or about 180°. Specifically, within the passband of the filter, the phase shift may be 180° at least once. More than one 180° crossing of the phase shift function within the passband is also possible. For example, the phase shifter may perform a 180° phase shift at the center frequency of the passband bandwidth.
[0012] The phase shifter can be implemented by various embodiments. Any embodiment that presents at least one phase shift with a value of 180° within the passband of the RF filter is possible. According to an embodiment, the phase shifter may include a high-pass Pi circuit, a low-pass Pi circuit, a high-pass Tee circuit, or a low-pass Tee circuit. The phase shifter may also include its composition to additionally shape the out-of-band characteristics of the filter. The phase shifter may also include a corresponding high-order circuit, in which the number of elements is greater than 3 (three). In addition, a full-pass topology that provides good insertion loss in a wide passband and allows phase shaping may also be used. An equalizer circuit may also be used. Circuit representations of any of the above-mentioned phase shifter circuits are available to those skilled in the art.
[0013] The phase shifter may also be implemented using a so-called hybrid circuit. According to an embodiment, the phase shifter may include a 0° / 180° 3dB hybrid circuit coupled between the first signal path and the second signal path.
[0014] According to another embodiment, the phase shifter may include a 0° / 90° 3dB hybrid circuit coupled between the first signal path and the second signal path and a phase shifter configured to perform a phase shift of 90° or approximately 90°, so that in general, a phase shift of 180° or approximately 180° can be achieved. Specifically, within the passband of the RF filter, the phase shift may be 90° at least once. The 90° phase shift of the hybrid circuit and the 90° phase shift from the phase shifter circuit together form a 180° phase shift in one of the first signal path and the second signal path.
[0015] The 180° phase shifter may also be implemented as a power divider combined with a first branch and a second branch, the power divider having a phase shift of approximately 180° at least once between the first branch and the second branch within the passband of the RF filter. According to an embodiment, the phase shifter may include a power divider having a first output terminal and a second output terminal, wherein a high pass filter is connected to one of the first output terminal and the second output terminal, and a low pass filter is connected to the other of the first output terminal and the second output terminal. The use of a power combiner may require an additional phase correction circuit to provide a 0° / 180° phase shift in the two branches of the half-lattice filter.
[0016] According to an embodiment, a matching circuit may be connected between at least one of the first port and the second port and the first signal path and the second signal path. The matching circuit may include a series-connected inductor or a series-connected inductor and a parallel-connected capacitor such as a low-pass filter. The matching circuit may achieve a flat passband performance. Of course, other matching circuits mentioned as examples and familiar to those skilled in the art may also be used.
[0017] In addition, the first port and the second port of the RF filter should match the external circuit coupled thereto. The half-lattice filter structure allows the size of the impedance matching circuit to be relatively free, so that the impedance matching of the first port and the second port can achieve substantially different impedances. In one embodiment, the matching circuit can be formed so that the impedance of the first port and the second port is substantially different. The proposed half-lattice method allows impedance transformation from the input to the output of the filter.
[0018] The half-lattice filter structure allows the skirts of the stopband and passband to be modified relatively independently with respect to the passband, so that the stopband attenuation and the shape of the filter skirt do not substantially affect the passband attenuation. This is achieved by coupling other elements to the first port and / or the second port, which provide a finite transmission zero (FTZ) function. The placement of the FTZ is relatively independent, so the designer can choose to place them all at the input of the filter, all at the output of the filter, or at both the input and output of the filter. According to an embodiment, at least one or more microacoustic resonators or one or more parallel resonant circuits are connected in series between the first path and the second path and one of the first port and the second port to perform a finite transmission zero function. According to an embodiment, at least one or more microacoustic resonators or one or more series resonant circuits are connected in parallel to a node coupled between the first signal path and the second signal path and one of the first port and the second port to perform a finite transmission zero function. It must be understood that the placement of various FTZs may require additional elements such as inductors, capacitors, etc. to properly match between the FTZs. This is well known to those skilled in the art.
[0019] Resonators and phase shifters may also be cascaded in the first signal path and the second signal path to increase the degree of filtering. According to an embodiment, at least one additional resonator is cascaded in each of the first signal path and the second signal path. To extend the concept of cascading, at least one other additional resonator is cascaded in each of the first signal path and the second signal path together with an additional phase shifter, which is configured to perform a 180° phase shift at least once within the passband of the RF filter in the second signal path. All resonators involved in the cascaded half-lattice filter are allowed to have different properties, such as resonant frequency, static capacitance, zero-pole distance, manufacturing technology, etc.
[0020] According to an embodiment, for example, one or more or all microacoustic resonators are composed of an anti-series cascade of a first resonator and a second resonator or an anti-parallel cascade of a first resonator and a second resonator to improve the linearity of the filter. The first resonator and the second resonator of the anti-series and anti-parallel cascades present a piezoelectric polarity opposite to that of the piezoelectric materials that constitute them relative to the applied electric field. In the anti-series cascade, the above-mentioned first resonator and the second resonator are connected in series, so that the polarity of the piezoelectric material of the above-mentioned resonator has an opposite orientation to the signal propagating through the above-mentioned first resonator and the second resonator, for example, the voltage, the electric field at a certain timestamp T0. In the anti-parallel cascade, the above-mentioned first resonator and the second resonator are connected in parallel, so that the polarity of the piezoelectric material of the resonator has an opposite orientation to the signal propagating through the first resonator and the second resonator, for example, the voltage, the electric field at a certain timestamp T0.
[0021] In the RF filter circuit, the elements can be divided into different chips. The first chip may include a microacoustic resonator with a resonant frequency within or near the passband of the RF filter. The second chip may include a microacoustic resonator with a resonant frequency substantially higher than the upper edge of the passband, and the third chip may include a microacoustic resonator with a resonant frequency substantially lower than the lower edge of the passband. Microacoustic resonators with resonant frequencies close to the upper and lower edges of the passband or the passband of the RF filter may preferably be located on the first chip. In addition, an additional chip or filter stack may include an FTZ implemented as a series resonant circuit and / or a parallel resonant circuit, and optionally include additional inductors and / or capacitors implemented in any substrate-based technology such as stacks, passive devices on glass, low temperature co-fired ceramics (LTCC), any integrated passive device (IPD) technology, etc. If the capacitors required for the phase shifter implementation can be implemented as on-chip capacitors on a SAW substrate or a BAW substrate, they can be included in the first chip.
[0022] As an illustrative example, the RF filter can be configured to present a passband portion in the range between 4.4 GHz and 5.0 GHz that forms the n79 band. The out-of-band portion of the filter is in the frequency range below 4.4 GHz and above 5.0 GHz. The microacoustic resonator of the filter can be only a surface acoustic wave resonator or only a bulk acoustic wave resonator, or can include a surface acoustic wave resonator and a bulk acoustic wave resonator, which are implemented in various technical forms and are known by different names, such as LiTaO3, high Q temperature compensation filter (HQTCF), ThinFilm-SAW, SMR-BAW, FBAR, etc. The piezoelectric substrate can include lithium tantalate, lithium niobate, aluminum nitride or aluminum scandium nitride, or any other piezoelectric material to realize a microacoustic resonator that supports a surface acoustic wave type or a bulk acoustic wave type. MEMS-based resonators are also possible. The resonator using such piezoelectric material and half-lattice filter structure according to the above-described embodiment is able to cover the relatively wide passband required for the n79 band and exhibits a steep filter skirt and sufficient attenuation of the out-of-band portion to suppress other RF services located in the band immediately adjacent to the n79 band. Another example is given by the n77 band, which covers a 900 MHz bandwidth (3300 MHz-4200 MHz) and requires coexistence between B42 / B43 (included in the n77 band) and n79. The proposed half-lattice approach is an excellent solution that can meet customer requirements with standard microacoustic resonators.
[0023] It should be understood that both the foregoing general description and the following detailed description are merely exemplary and are intended to provide an overview or framework for understanding the nature and character of the claims. The accompanying drawings are included to provide further understanding and are incorporated in and constitute a part of this specification. The accompanying drawings illustrate one or more embodiments and, together with the description, are used to explain the principles and operation of the various embodiments. The same elements in different figures of the accompanying drawings are represented by the same reference numerals. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] In the figure:
[0025] Figure 1 shows a schematic diagram of a half-lattice type RF filter according to the principles of the present disclosure;
[0026] Figure 2 A schematic diagram showing a half-lattice type RF filter using a matching circuit;
[0027] Figure 3 A schematic diagram showing a half-lattice type RF filter using other matching circuits;
[0028] Figure 4 A principle implementation of an RF filter including a phase shifter is shown;
[0029] Figures 5A to 5I Shows Figure 4 Circuit diagrams of several embodiments of phase shifters;
[0030] Figure 6 An RF filter including a schematic diagram of yet another embodiment of a phase shifter is shown;
[0031] Figure 7 An RF filter including a further embodiment of a phase shifter is shown;
[0032] Figure 8 A schematic diagram showing another embodiment of a phase shifter is shown;
[0033] Fig.9A and 9B A principle diagram of another embodiment of a phase shifter and a corresponding schematic diagram of the phase shifter are shown;
[0034] Fig.10 shows a schematic diagram of an RF filter using a finite transmission zero circuit;
[0035] Fig.11 shows a schematic diagram of an RF filter having a cascaded half-lattice structure;
[0036] Fig. 12A and 12B shows a schematic diagram of an RF filter for the n79 frequency band and the corresponding transfer function;
[0037] Fig.13 A schematic diagram showing an RF filter using anti-series cascading; and
[0038] Fig.14 A schematic diagram of an RF filter is shown, illustrating the partitioning of the elements of the filter. DETAILED DESCRIPTION
[0039] The present disclosure will now be described more fully below with reference to the accompanying drawings showing embodiments of the present disclosure. However, the present disclosure can be implemented in many different forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to enable the present disclosure to fully convey the scope of the present disclosure to those skilled in the art. The accompanying drawings are not necessarily drawn to scale, but are configured to clearly illustrate the present disclosure.
[0040] Figure 1 A principle embodiment of an RF filter is shown. The RF filter has a half-lattice filter structure, which includes a first signal path 140 and a second signal path 150 coupled between a port 101 and another port 102. The signal path 140 includes a microacoustic resonator 141. The signal path 150 includes a series connection of another microacoustic resonator 151 and a phase shifter 152. The resonators 141, 151 can be surface acoustic wave resonators or bulk acoustic wave resonators, wherein the resonators 141, 151 can be resonators of the same type or different types. The phase shifter 152 provides a phase shift of approximately 180° or 180°. During the passband of the RF filter shown, the phase shifter 152 can have a phase shift of 180° at least once. Therefore, the phase shifter 152 can present one or more frequencies at which its phase shift for the frequency is 180°. At other frequencies within the passband of the RF filter, the phase shift may be close to 180° or may vary significantly from this, as the phase slope depends on the phase shifter implementation chosen. The 180° phase shift may occur near or at the center frequency of the passband.
[0041] Figure 1The passband of the RF filter is determined by the resonant frequencies of the resonators 141, 151. For example, the resonant frequency of the microacoustic resonator 141 may be lower than that of the microacoustic resonator 151. The lower passband corner of the passband of the RF filter is approximately at the resonant frequency of the resonator 141, while the upper passband corner is approximately at the resonant frequency of the resonator 151. It should be noted that the width of the passband of the RF filter does not depend on the coupling of the piezoelectric material of the resonator related to the distance between the resonant frequency and the anti-resonant frequency of the resonator, which is the case in conventional filters based on ladder structures. In contrast, the filter bandwidth of the semi-lattice type filter structure is determined by the resonant frequencies of the resonators used in the first signal path and the second signal path. Therefore, a filter with a larger passband bandwidth can be formed using resonators with a correspondingly larger difference between the resonant frequencies. The first signal path 140 and the second signal path 150 including the resonators 141, 151 and the phase shifter 152 form a semi-lattice type unit cell.
[0042] The phase shifter may be provided in either of the signal paths 140, 150. Figure 1 Phase shifter 152 is shown in signal path 150, but it is also possible to insert phase shifter 152 into signal path 140 connected in series with resonator 141. For practical reasons, it is useful to include a 180° phase shifter in this signal path including a resonator with a lower resonant frequency. In this case, the phase shifter is connected to a resonator with a resonant frequency at about the lower passband corner. In addition, as another alternative, the phase shifter can be located between port 101 and one of resonators 141, 151.
[0043] Figure 2 Another embodiment of the RF filter is shown. A phase shifter 232 is connected to a resonator 111 having a lower resonant frequency when compared to the resonator 121. The first signal path 110 and the second signal path 120 including the resonators 111, 121 and the phase shifter 232 form another embodiment of a half-lattice type unit cell. In addition, a matching element is connected between the first signal path 110 and the second signal path 120 and the input / output ports 101, 102. A series-connected inductor or coil 211 is connected between the port 101 and a node 221 that combines the first signal path 110 and the second signal path 120. A series-connected inductor or coil 212 is connected between the port 102 and another node 122 that couples the first signal path 110 and the second signal path 120. The inductors 111, 212 are matched to the unit cell to achieve the desired passband performance. The matching circuits 211, 212 achieve a relatively flat upper portion of the passband. The example shown is a relatively simple form of a matching circuit. More complex matching circuits are also useful.
[0044] If the microacoustic resonator provides spurious modes and / or other losses above the resonant frequency, the specific resonator (such as resonator 111) that forms the lower passband corner can be replaced with a capacitor. This can reduce the steepness of the left skirt because it has a free passband for the spurious modes. If a bulk acoustic wave resonator is used, the application of different physical implementations may be advantageous. BAW resonators that provide type II dispersion (SMR-BAW, FBAR) do have spurious modes below the series resonance frequency, making them an attractive solution for resonator 111 because this resonator is responsible for the lower passband corner of the filter. BAW resonators that provide type I dispersion (SMR-BAW, FBAR) do have spurious modes above the resonant frequency, making them an attractive solution for resonator 121 because this resonator is responsible for the upper passband corner of the filter. The resonant frequency of resonator 111 is lower than the resonant frequency of resonator 121. Resonator 111 is responsible for the lower passband corner and resonator 121 is responsible for the upper passband corner. In Figure 2 In the embodiment shown, the phase shifter 232 is connected to the resonator 121 such that the phase shifter 232 is connected to the resonator responsible for the lower passband corner. In this case, the resonator 121 may be implemented by a microacoustic resonator such as a BAW resonator or, alternatively, as a capacitor.
[0045] Figure 3 Another embodiment of an RF filter is shown including further possibilities for matching the filter passband performance. Figure 3 The RF filter of includes inductors 321, 322 connected in parallel with resonators 111 and 121, respectively. Furthermore, on the port 101 side, for additional matching purposes, a series inductor 311 and a capacitor 313 connected in parallel or in parallel are provided. The capacitor 313 is connected between the node 221 and a terminal 315 for ground potential. At the node 102, the series-connected inductor 312 is connected between the port 102 and the node 222. The node 222 is also connected to the terminal 315 for ground potential via a capacitor 314. The inductors 211, 212 ( Figure 2 ) or low-pass circuits 311, 313, 312, 314 or parallel / parallel connected inductors 321, 322 ( Figure 3 ) can be used alone or in combination with any other element to achieve the desired match of the passband of the RF filter. Other matching circuits (high pass) can also be used.
[0046] Figure 4An embodiment of an RF filter is shown, which shows a generalized two-port device 411 representing a phase shifter. The device 411 includes two ports 412, 413 that can be described by a chain matrix [■(A&B@C&D)]. One terminal of the port 412, 413 is connected to ground potential. The other terminal is connected to the signal path 110, where a phase shift of about 180° is generated between the terminals. The phase shift is such that it occurs at least once at 180° within the passband of the filter.
[0047] Figures 5A to 5I Various embodiments of a two-port phase shifter device 411 are shown. Turning now to Figure 5A , shows a low pass Pi circuit comprising an inductor 511 having its inductor terminals connected to a terminal 501 for ground potential via corresponding capacitors 512, 513. Turning now to Figure 5B , shows a low-pass Tee circuit. The low-pass Tee circuit includes two inductors 521, 522 connected in series, wherein a node 524 of the coupled inductors 521, 522 is connected to the ground terminal 501 through a capacitor 523 connected in parallel. Now turn to Figure 5C , shows a high-pass Pi circuit. The high-pass Pi circuit includes a series-connected capacitor 531, the capacitor terminals of which are connected to the ground terminal 501 through parallel-connected inductors 532, 533. Now turn to Figure 5D , shows a high-pass Tee circuit. The high-pass Tee circuit includes capacitors 541 and 542 connected in series, wherein a node 544 of the coupling capacitors 541 and 542 is connected to the ground terminal 501 through an inductor 543 connected in parallel. Now turn to Figure 5E , showing the second stage of two cascaded low-pass Pi circuits. The cascaded low-pass Pi circuit includes a cascade configuration Figure 5A Two low-pass Pi circuits. Figure 5E The circuit shown in comprises a series connection of two inductors 551, 552, wherein the terminals of the inductors 551, 552 are connected to the ground terminal 501 via parallel connected capacitors 553, 554, 555. The inner two capacitors of the cascaded low pass Pi element are compressed to one capacitor, such as capacitor 554. Fig. 5F A 4-element high-pass is shown as an example of a phase shifter having more than three elements, but still based on Figures 5A to 5D One of the four basic phase shifter cells shown. Fig. 5F The phase shifter is based on Figure 5C The high-pass Pi circuit also includes a series capacitor 561. Figure 5G and 5H The circuit of shows the basic unit cell forming an all-pass, which provides an almost flat passband, but only shapes the phase characteristics forming the all-pass. Figure 5Gand 5H The circuit is based on a high-pass Pi circuit and a low-pass Pi circuit, and also includes a capacitor 571 and an inductor 581 connected in parallel, respectively. Fig.5I An example of an equalizer circuit in a bridge-T form based on a high-pass Tee circuit is shown, in which an inductor 592 is connected in parallel with a series capacitor, and a parallel-connected capacitor 591 is provided.
[0048] Figures 5A to 5I Any of the above circuits shown can be used to implement Figure 4 A two-port phase shifter device 411. Figures 5A to 5I Two or more of the phase shifter circuits can be cascaded to form a second-order or higher-order phase shifter, depending on the desired performance of the RF filter. A higher-order phase shifter may have better broadband performance of the phase shifter, thereby improving the S21 parameter of the RF filter. Figures 5A to 5D The cascading of the circuits is equivalent to providing more than three elements. As an example, Fig. 5F The circuit is achieved by adding an additional series capacitor 561 to Figure 5C The expansion of the circuit.
[0049] Figure 6 An RF filter is shown that includes a phase shifter device 611 implemented as a four-port circuit with ports 612, 613, 614, 615. Port 615 can be connected to the ground terminal 501 through a terminal device 616 such as a resistor. Resistor 616 can be omitted so that port 615 is directly connected to the ground terminal 501. Element 616 can also be a complex impedance. Alternatively, a three-port device including ports 612, ..., 614 and omitting port 615 can be used. Circuit 611 establishes a 180° phase shift between ports 612 and 613.
[0050] Circuit 611 can be implemented using several examples. For example, circuit 611 can be implemented as a balun circuit that converts a balanced signal (two signal lines include a 0° signal line and a 180° signal line, both referenced to ground potential GND) into a single-ended signal (one signal line, such as a 0° signal, referenced to ground potential GND). The balun circuit can also be implemented using a 0° / 90° 3dB hybrid circuit and an additional 90° phase shifter. If the circuit system around the RF filter already involves an orthogonal hybrid circuit, a 3dB hybrid circuit can be used, so that the phase shifter 611 can use such a circuit that already exists in other circuits available on the circuit board. This may be the case with a power amplifier circuit. Other possibilities may use a transition from low pass to high pass, which includes a 180° phase shift, to which a power divider, such as a 3dB Wilkinson power divider, can be added. Possible implementations of circuit 611 that employ one or more of the above concepts are described in more detail below.
[0051] Figure 7 An RF filter using a 3dB (0° / 90°) hybrid circuit 711 is shown. Circuit 711 generates a 90° phase shift between its terminals 1-4, so an additional phase shifter 721 is required to achieve an overall phase shift of 180°. Phase shifter 721 performs a 90° phase shift at least once within the passband of the RF filter shown. As an alternative embodiment (not shown), circuit 711 can be implemented as a 180° hybrid circuit such as a balun circuit, so that the 90° phase shifter 721 can be omitted.
[0052] Figure 8 Shows Figure 6 Another implementation of circuit 611. Figure 8 The circuit in includes a port 820, to which two chain matrix circuits 811, 812 having different chain matrices [■(A&B@C&D)] are connected, and the two chain matrix circuits 811, 812 are coupled to separate ports 821, 822. A phase shift of approximately 180° is provided between the ports 821 and 822. The circuit 811 can be implemented as a low-pass filter, and the circuit 812 can be implemented as a high-pass filter, so that a phase shift of 180° can be achieved between the ports 821 and 822.
[0053] Fig.9A and 9B The circuit shown is further described in detail Figure 8 Other examples of the concepts shown. Fig.9AA power divider 931, 932 represented by a corresponding [■(A&B@C&D)] matrix device coupled to port 920 is included. The power divider 931, 932 (such as a Wilkinson power divider) or the like can be implemented as a 3dB power dividing device. A low pass filter 911 is provided in one of the split signal paths, and a high pass filter 912 is provided in the other split signal path. The output ports 921, 922 connected to the low pass and high pass filters 911, 912 present a 180° phase shift between them.
[0054] Fig. 9B Shows Fig.9A Schematic circuit implementation of the general phase shifter concept of the Wilkinson power divider. The power dividers 931, 932 are implemented by a low-pass Tee circuit, which includes two inductors connected in series and a capacitor connected in parallel, and the capacitor is coupled to the node between the two inductors. The low-pass circuit 911 includes a number of inductors connected in series, wherein the node between the inductors is coupled to the ground potential through the capacitors connected in parallel. The high-pass filter 912 includes at least three capacitors connected in series, wherein the node between the capacitors is connected to the ground potential through the inductors connected in parallel. The resistor 941 is used to match the signal line required by the Wilkinson power divider circuit system.
[0055] Figure 6 Other implementations of the phase shifter 611 are possible, such as an all-pass filter that generates a phase shift of 180°. A Marchand-balun may also be used, which includes a transmission line that implements broadband balun operation. However, the size of the Marchand-balun may be large.
[0056] Fig.10 Several possible RF circuits including elements for performing finite transmission zero functions are shown. In general, finite transmission zero (FTZ) circuit elements can be used to shape filter performance in the skirt of the passband and in the stopband region outside the passband. FTZ elements can be used to insert transmission poles and / or make the skirts steeper at specific frequency locations. FTZ elements can be added between half-lattice cells and input / output ports. When compared to traditional filter structures (such as ladder filters), FTZ elements can shape filter performance regardless of the half-lattice cells and the passband of the RF filter. The filter designer can relatively freely choose the number of FTZ elements and the type of FTZ elements according to the circuit requirements. Fig.10 The circuit in shows several possibilities of FTZ elements in one circuit. It should be noted that one or more of the shown elements may be omitted based on performance requirements.
[0057] exist Fig.10In the central part of the RF filter, a half-lattice unit cell 1010 is shown. On the left side of the half-lattice unit cell 1010, an FTZ resonator 1021 is connected in series between the terminal 101 and the half-lattice unit 1010. A series resonant element 1022 connected in parallel is connected between the microacoustic resonator 1021 and the ground potential 501. A series-connected parallel resonant circuit 1023 is connected in series between the port 101 and the resonator 1021. A series inductor 1050 represents a matching inductor. On the right side of the half-lattice unit 1010, a parallel-connected microacoustic resonator 1031 is provided, which is connected between the unit cell 1010 and the ground potential. A series-connected parallel resonant circuit 1032 is connected between the unit cell 1010 and the port 102. A parallel-connected series resonant circuit 1033 is connected between the parallel resonant circuit 1032 and the ground potential. An inductor 1051 represents a matching circuit directly connected to the port 102. As shown Fig.10 As shown, one of the FTZ circuits can be connected in series, where the finite transmission zero function of the resonator is determined by the anti-resonant frequency of the resonator. Another FTZ circuit can be connected in parallel or in series, where the finite transmission zero function of the resonator is determined by the series resonant frequency of the resonator. FTZ circuits can also be connected in a combination of series and parallel like a ladder. The number of FTZ circuits is theoretically unlimited, so the designer can place as many FTZ circuits as the specifications of the RF filter require. As shown in FIG. Fig. 12A As shown, it is also possible to connect the FTZs only in series or only in parallel.
[0058] Typically, the passband operation of the RF filter is primarily determined by the half-lattice unit 1010 including the matching circuit, and the out-of-band performance is primarily determined by the selection of the FTZ element and the phase shifter topology, where both the passband and out-of-band functions can be combined using Fig.10 The concepts described are relatively independent and freely designed from each other. This capability is different from conventional ladder filters. The FTZ element can be implemented with a microacoustic resonator such as resonator 1021, 1031. If the frequency of the FTZ is too high or too low, it can be implemented with an LC resonant circuit connected to unit 1010. When the frequency spread of the resonant frequency of the FTZ element is so wide that the individual resonators can no longer be implemented on one chip, multiple microacoustic chips can be used, or some FTZ resonators can be implemented with substrate-based technologies such as low temperature co-fired ceramics (LTCC), laminates, high temperature co-fired ceramics (HTCC), passive devices on glass (POG) or integrated passive devices (IPD).
[0059] Fig.11An RF filter including a cascaded half-lattice unit cell 1110 is shown. The second stage cascaded half-lattice unit cell 1110 includes two cascaded first stage half-lattice unit cells 1120, 1130. The first unit cell 1120 includes resonators 1121, 1122 in the first signal path and the second signal path, and a 180° phase shifter 1122. The second unit cell 1130 includes resonators 1131, 1132 in the first signal path and the second signal path and a 180° phase shifter 1132 belonging to the unit cell. As shown, the phase shifters 1122, 1132 from the two unit cells 1120, 1130 are combined into a single phase shifter 1142. Careful design allows a circuit to be designed that still provides approximately a 180° phase shift element 1142 in combination with the four resonators 1121, 1131, 1122 and 1132. Additional unit cells (not shown) may also be cascaded. The number of phase shifters required is basically half the number of cascaded unit cells. That is, in the non-cascaded case, one unit cell requires one phase shifter. When two unit cells are cascaded, one phase shifter is still required. When three unit cells are cascaded, two phase shifters are required. When four unit cells are cascaded, two phase shifters are required, and so on. The characteristics of the individual resonators 1121, 1131, 1122 and 1132 can be different. If the loss and / or spurious modes have too much impact on the passband, one or more resonators can also be replaced with capacitors (as described above).
[0060] Fig. 12A , 12B A detailed circuit diagram of an exemplary n79 filter for 5G communication services and corresponding transmission functions is shown. The basic unit cell of the n79 filter includes half-lattice type resonators 1211, 1212. A phase shifter 1213 is connected in series to the resonator 1211. Similar to Figure 5B , the phase shifter 1213 is implemented as a second-order low-pass Tee circuit. Figure 5D A second-order high-pass Tee circuit or similar Fig. 5F The circuit of is also useful. Port 1201 is connected to the unit cell 1210 through a matching inductor 1202 and a series connection of microacoustic resonators 1203 and 1204 constituting two FTZs. The other port 1202 is connected to the unit cell 1210 through a matching inductor 1205, a parallel-connected FTZ microacoustic resonator 1206, and a series-connected FTZ resonator 1207. Fig. 12B Shows Fig. 12AThe figure shows the transfer function of the filter. The figure shows a passband between 4.4 GHz and 5.0 GHz, with a relatively flat top shape of the passband. The passband spans between the lower passband corner 1221 and the upper passband corner 1222. On the right side of the passband (beyond the upper skirt of the passband and the upper passband corner 1222), sufficient attenuation is provided to suppress WLAN services within this frequency range.
[0061] It should be noted that the matching circuits connected to the input / output ports 1201, 1202 can be configured quite freely so that an impedance conversion from one impedance level to another substantially different impedance level can be performed. For example, the impedance to be provided at port 1201 may be 50 ohms, and the impedance to be provided at port 1202 may be 200 ohms. In this case, the matching inductor 1205 must be reconfigured to achieve an impedance of 200 ohms ( Fig. 12A 12), which can be easily implemented by a skilled circuit designer. Since the filter passband is implemented by the half-lattice unit 1210, the circuit designer is relatively free to size the impedance of the input / output port to a desired level. Specifically, the impedance at one port can be converted to another substantially different impedance at another port, wherein the S21 parameter of the resulting filter remains at almost the same level (almost unchanged).
[0062] Fig.13 A schematic diagram of an RF filter circuit is shown, in which, for example, resonators are cascaded in an anti-serial cascade to improve linearity. In a half-lattice unit 1310, resonators 1311a, 1311b are connected in series, wherein the polarity of the piezoelectric material included in the above resonators has an anti-serial orientation, which is shown with corresponding arrows. Compared with resonator 1311a, the arrow of resonator 1311a is shown from left to right, and the arrow of resonator 1311b is shown from right to left, i.e., in the opposite direction. In fact, the opposite polarity orientation of the piezoelectric material can be easily selected during the manufacturing process of the above resonator. When a signal propagates from resonator 1311a to resonator 1311b through the circuit, the signal experiences the opposite polarity of the piezoelectric material, so that the contraction of the piezoelectric material in resonator 1311a occurs simultaneously with the expansion of the piezoelectric material of resonator 1311b. And vice versa. The same principle of the anti-serial cascade of resonators has been applied to FTZ elements 1303 and 1304.
[0063] As another possibility ( Fig.13 (not shown), the resonators may be combined in an anti-parallel cascade connection, wherein the piezoelectric material of one resonator is oriented opposite to the piezoelectric material of another resonator in the parallel connection of the resonators. Fig.13The circuit shows a first order anti-series cascade of resonators, but second and higher orders of anti-series and anti-parallel cascades of resonators are also possible. Anti-series or anti-parallel resonator cascading schemes can be applied to improve the linearity of the filter. The improved linearity improves filter performance and reduces distortion in the filter signal. Resonator cascading also increases the effective area of the resonators, so the filter can handle more power, thereby improving the power endurance of the filter.
[0064] Fig.14 The division of devices within the filter in terms of manufacturing technology is shown. Generally, substrate-based technology and microacoustic resonator technology are available. Substrate-based technologies include low-temperature co-fired ceramics, laminates, high-temperature co-fired ceramics, passive devices on glass (POG), integrated passive devices (IPD). Microacoustic resonator technology includes thin film surface acoustic wave resonators, bulk acoustic wave resonators, substrates such as lithium tantalate (LiTaO3) with different cutting angles such as 42°, lithium niobate (LiNbO3), aluminum scandium nitride (AlScN), aluminum nitride (AlN) or any other piezoelectric material. High Q temperature compensated filters (HQTCF) are also possible. Lamb wave resonators (SAW on AlN or AlScN), ThinFilm-SAW or other microacoustic resonator technologies can be used for this specific filter design concept.
[0065] Elements close to the input / output ports (such as elements B1, B9) can be implemented as lumped elements using discrete devices such as inductors and / or capacitors to implement a matching network. A combination of one of the substrate-based technologies and lumped elements is also possible. Therefore, elements B1, B9 can be implemented on a circuit substrate 1410. Elements B2, B8 can be FTZ elements whose resonant frequencies are far away from the passband of the RF filter, so that they can be implemented as one microacoustic chip 1420. Alternatively, they can be implemented within a substrate-based technology. The resonant frequencies of FTZ elements B3, B7 are assumed to be close to the passband of the RF filter, so they can be implemented on another microacoustic chip 1430 different from chip 1420. Resonators B4, B5 and phase shifters B6 determine the passband of the filter so that they are also implemented on chip 1430 together with nearby FTZ elements B3, B7. Note that the capacitors of distant FTZ elements B2, B8 can be implemented on microacoustic chip 1430 together with nearby elements. In addition, the capacitor used to implement the phase shifter B6 can be placed on the microacoustic chip 1430. In the case where spurious modes affect the passband of the filter, the resonator B4 can be implemented by a capacitor. In the case of using bulk acoustic wave devices, the resonator B4 can be implemented by providing a resonator of dispersion type II, and the resonator B5 can be implemented by providing a resonator of dispersion type I. If the manufacturing technology used does not allow resonators B4 and B5 to be implemented on the same chip, they can also be implemented on different chips. This is especially the case if a mixture of resonator technologies is used.
[0066] In one embodiment, the resonant frequency of resonator B4 is lower than that of resonator B5, such that resonator B4 is responsible for the lower passband corner 1221 and resonator B5 is responsible for the upper passband corner 1222. Resonator B4 may be implemented by a resonator technology such as a BAW technology that provides loss and / or spurious modes below the resonant frequency of resonator B4. Resonator B5 may be implemented by another resonator technology such as another BAW technology that provides loss and / or spurious modes above the resonant frequency of resonator B5.
[0067] In summary, conventional ladder topologies do not provide a suitable solution for next-generation filters such as 5G communication services, as available piezoelectric materials may not allow the required wide bandwidth. The half-lattice unit cell approach can achieve wide passband bandwidth and allows flexibility in optimizing passband and stopband performance. The half-lattice filter structure includes a 180° phase shifter for which various design approaches can be used.
[0068] It will be clear to those skilled in the art that various modifications and variations may be made without departing from the spirit or scope of the present disclosure as defined in the appended claims. Since modifications, combinations, sub-combinations and variations of the disclosed embodiments that incorporate the spirit and substance of the present disclosure are conceivable to those skilled in the art, the present disclosure should be interpreted as including all contents within the scope of the appended claims.
Claims
1. A micro-acoustic RF filter, comprising: A first port (101) and a second port (102); a first node coupled to the first port; a second node coupled to the second port; a first signal path (120) coupled between the first node and the second node, the first signal path comprising a first resonator (121); a second signal path (110) coupled between the first node and the second node, the second signal path comprising a second resonator (111) and a phase shifter (232) connected in series with the second resonator, wherein at least one of the first resonator and the second resonator (111, 121) comprises a microacoustic resonator; a third resonator, wherein the third resonator comprises an additional microacoustic resonator, a first end of the additional microacoustic resonator is coupled to the second node, and a second end of the additional microacoustic resonator is coupled to a reference node; as well as A fourth resonator is coupled between the first port and the first node.
2. The microacoustic RF filter of claim 1, wherein the phase shifter (232) is configured to perform a phase shift of 180° at least once within a passband of the RF filter.
3. The microacoustic RF filter according to claim 1 or 2, wherein the phase shifter (411) comprises at least one of the following: a high-pass Pi circuit, a low-pass Pi circuit, a high-pass Tee circuit, a low-pass Tee circuit, an all-pass circuit and a combination thereof, so as to realize a phase shifter composed of at least three elements arranged in the second signal path.
4. The microacoustic RF filter according to claim 1 or 2, wherein the phase shifter comprises a 0° / 180° -3dB hybrid circuit (611) coupled between the first signal path and the second signal path.
5. A microacoustic RF filter according to claim 1 or 2, wherein the phase shifter includes a 0° / 90° -3dB hybrid circuit (711) coupled between the first signal path and the second signal path, and one of the first signal path and the second signal path includes a phase shifter (721) configured to perform a 90° phase shift at least once within the passband of the RF filter.
6. The microacoustic RF filter according to claim 1 or 2, wherein the phase shifter comprises a power divider (931, 932) having a first output terminal and a second output terminal, and a high-pass filter (912) connected to one of the first output terminal and the second output terminal and a low-pass filter (911) connected to the other of the first output terminal and the second output terminal.
7. The microacoustic RF filter according to claim 1 or 2, further comprising a matching circuit (211, 212), wherein the matching circuit (211, 212) is connected between at least one of the first port and the second port (101, 102) and the first signal path and the second signal path (110, 120) to provide passband matching.
8. The microacoustic RF filter according to claim 7, wherein the matching circuit comprises at least one of a series-connected inductor (311, 312) and a parallel-connected capacitor (313, 314).
9. The microacoustic RF filter according to claim 1 or 2, further comprising at least one or more of a microacoustic resonator (1021) and a parallel resonant circuit (1023, 1032) connected in series between the first signal path and the second signal path and one of the first port and the second port (101, 102) to perform a finite transmission zero function.
10. The microacoustic RF filter according to claim 1 or 2, further comprising at least one or more of a microacoustic resonator (1031) and a series resonant circuit (1022, 1033) connected in parallel to a node coupled between the first signal path and the second signal path and a terminal (501) for ground potential to perform a finite transmission zero function.
11. The microacoustic RF filter according to claim 1 or 2, wherein at least one additional resonator (1131, 1132) is cascaded in each of the first signal path and the second signal path.
12. The microacoustic RF filter according to claim 1 or 2, further comprising at least one other additional resonator connected in cascade in each of the first signal path and the second signal path, and further comprising an additional phase shifter connected in cascade in the second signal path and configured to perform a phase shift of 180° at least once within the passband of the RF filter.
13. The microacoustic RF filter according to claim 1 or 2, wherein one or more of the microacoustic resonators (1311a, 1311b, 1312a, 1312b, 1303, 1304) is composed of an anti-series cascade or an anti-parallel cascade of the first resonator and the second resonator having opposite piezoelectric polarities.
14. The microacoustic RF filter according to claim 1 or 2 comprises a first chip and a second chip (1420, 1430) configured to implement a microacoustic resonator, wherein the first chip (1430) comprises a microacoustic resonator having a resonant frequency within a passband of the RF filter, and the second chip (1420) comprises a microacoustic resonator having a resonant frequency substantially higher than an upper edge of the passband or substantially lower than a lower edge of the passband.
15. The microacoustic RF filter according to claim 1 or 2, wherein the first resonator and the second resonator (121, 111) of the first signal path and the second signal path (120, 110) each exhibit a resonant frequency, wherein the resonant frequency of the second resonator (111) of the second signal path (110) is lower than the resonant frequency of the first resonator (121) of the first signal path (120), and wherein the phase shifter (232) is connected to the second resonator (111) of the second signal path (110).
16. A microacoustic RF filter according to claim 15, wherein the second resonator (111) of the second signal path is implemented by a microacoustic resonator technology that provides losses and / or spurious modes below the resonant frequency of the second resonator (111), and wherein the first resonator (121) of the first signal path is implemented by a microacoustic resonator technology that provides losses and / or spurious modes above the resonant frequency of the first resonator (121).
17. A microacoustic RF filter according to claim 1 or 2, wherein the RF filter is configured to present a passband having a lower passband corner and an upper passband corner, wherein the second resonator is configured to form the lower passband corner and the first resonator is configured to form the upper passband corner, wherein the phase shifter (232) is connected to the second resonator configured to form the lower passband corner.
18. The microacoustic RF filter of claim 17, wherein the second resonator configured to form the lower passband corner comprises a bulk acoustic wave resonator exhibiting dispersion type II, and the first resonator configured to form the upper passband corner comprises a bulk acoustic wave resonator exhibiting dispersion type I.
19. The microacoustic RF filter according to claim 17, wherein the second resonator configured to form the lower passband corner is implemented by a capacitor, and the first resonator configured to form the upper passband corner is implemented by a microacoustic resonator.
20. The microacoustic RF filter of claim 19, wherein the phase shifter (232) is connected to the capacitor.
21. A microacoustic RF filter according to claim 1 or 2, wherein the RF filter is configured to present a passband portion and an out-of-band portion, wherein the passband portion is within the n79 band or between 4.4 GHz and 5.0 GHz, and the out-of-band portion is outside the n79 band or within a frequency range of less than 4.4 GHz and greater than 5.0 GHz, wherein the first resonator and the second resonator are one of a surface acoustic wave resonator or a bulk acoustic wave resonator of a piezoelectric substrate comprising at least one of lithium tantalate, lithium niobate and scandium aluminum nitride.
Citation Information
Patent Citations
Filter structure comprising piezoelectric resonators
US20030042995A1