CN117242700A

CN117242700A - Transverse excitation thin film bulk acoustic resonator matrix filter with switches in parallel with sub-filter shunt capacitors

Info

Publication number: CN117242700A
Application number: CN202280030881.4A
Authority: CN
Inventors: 安德鲁·盖耶特; 尼尔·芬齐; 迈克尔·艾迪; 布莱恩特·加西亚
Original assignee: Murata Manufacturing Co Ltd
Current assignee: Murata Manufacturing Co Ltd
Priority date: 2021-04-27
Filing date: 2022-04-14
Publication date: 2023-12-15

Abstract

Matrix filters having input ports and sub-filters connected between the input ports and corresponding output ports are disclosed. Each sub-filter comprises a ladder circuit with n laterally excited thin film bulk acoustic resonator (XBAR) series elements and n-1 capacitor parallel elements, where n (the order of the sub-filter) is an integer greater than 2. Each sub-filter also has a first switch in parallel with the first capacitor parallel element and a second switch in parallel with the last capacitor parallel element.

Description

Transverse excitation thin film bulk acoustic resonator matrix filter with switches in parallel with sub-filter shunt capacitors

Technical Field

The present disclosure relates to radio frequency filters using acoustic wave resonators, and in particular, to filters for use in communication devices.

Background

A Radio Frequency (RF) filter is a dual port device configured to pass certain frequencies and stop other frequencies, where "pass" means to transmit with relatively low signal loss and "stop" means to prevent or substantially attenuate. The range of frequencies through which a filter passes is referred to as the "passband" of the filter. The range of frequencies at which such a filter stops is referred to as the "stop band" of the filter. A typical RF filter has at least one pass band and at least one stop band. The specific requirements for either the pass band or the stop band depend on the specific application. For example, a "passband" may be defined as a frequency range where the insertion loss of the filter is better than a defined value (e.g., 1dB, 2dB, or 3 dB). A "stop band" may be defined as a frequency range where the rejection of the filter is greater than a defined value such as 20dB, 30dB, 40dB or more (depending on the application).

RF filters are used in communication systems that transmit information over a wireless link. For example, RF filters may be found in RF front ends of cellular base stations, mobile phones and computing devices, satellite transceivers and ground stations, ioT (internet of things) devices, laptop and tablet computers, fixed point radio links, and other communication systems. RF filters are also used in radar, as well as in electronic and information warfare systems.

RF filters typically require many design tradeoffs to achieve the best tradeoff between performance parameters (e.g., insertion loss, rejection, isolation, power handling, linearity, size, and cost) for each particular application. Particular designs and methods of manufacture and enhancements may benefit from one or more of these requirements simultaneously.

Performance enhancement of RF filters in wireless systems can have a wide impact on system performance. Improvements in RF filters may be used to provide system performance improvements such as larger cell sizes, longer battery life, higher data rates, greater network capacity, lower cost, enhanced security, higher reliability, etc. These improvements may be implemented individually and in combination at multiple levels of the wireless system (e.g., at the RF module, RF transceiver, mobile or fixed subsystem, or network level).

High performance RF filters for current communication systems typically contain acoustic wave resonators including Surface Acoustic Wave (SAW) resonators, bulk Acoustic Wave (BAW) resonators, film bulk acoustic wave resonators (FBAR), and other types of acoustic resonators. However, these prior art techniques are not well suited for use at the higher frequencies and bandwidths proposed for future communication networks.

The desire for a wider communication channel bandwidth will inevitably lead to the use of a higher frequency communication band. Radio access technologies for mobile telephone networks have been standardized by 3GPP (third generation partnership project). The radio access technology for the 5 th generation mobile network is defined in the 5G NR (new radio) standard. The 5G NR standard defines several new communication bands. Two of these new communication bands are: n77 using a frequency range from 3300MHz to 4200 MHz; and n79 using a frequency range from 4400MHz to 5000 MHz. Both frequency band n77 and frequency band n79 use time division multiplexing (TDD) such that communication devices operating in frequency band n77 and/or frequency band n79 use the same frequency for uplink and downlink transmissions. The band pass filters for the frequency bands n77 and n79 must be able to handle the transmit power of the communication device. The WiFi bands at 5GHz and 6GHz also require high frequencies and wide bandwidths. The 5G NR standard also defines a millimeter wave communication band having frequencies between 24.25GHz and 40 GHz.

A laterally excited thin film bulk acoustic resonator (XBAR) is an acoustic resonator structure for use in microwave filters. XBAR is described in patent US10,491,291 entitled "TRANSVERSELY EXCITED FILM BULK ACOUSTIC RESONATOR (transverse excited film bulk Acoustic resonator)". An XBAR resonator comprises an interdigital transducer (IDT) formed on a thin floating layer or diaphragm of monocrystalline piezoelectric material. The IDT includes a first set of parallel fingers extending from a first bus bar and a second set of parallel fingers extending from a second bus bar. The first set of parallel fingers and the second set of parallel fingers are staggered. A microwave signal applied to the IDT excites a shear primary acoustic wave in the piezoelectric diaphragm. The XBAR resonator provides very high electromechanical coupling and high frequency capability. The XBAR resonator may be used in various RF filters including band reject filters, band pass filters, diplexers and multiplexers. XBAR is well suited for use in filters for communication bands with frequencies above 3 GHz. The matrix XBAR filter is also suitable for frequencies between 1GHz and 3 GHz.

Drawings

Fig. 1 includes a schematic plan view, two schematic cross-sectional views, and a detailed cross-sectional view of a laterally excited thin film bulk acoustic resonator (XBAR).

Fig. 2A is an equivalent circuit model of an acoustic resonator.

Fig. 2B is a graph of the admittance of an ideal acoustic resonator.

Fig. 2C is a circuit symbol for an acoustic resonator.

Fig. 3A is a schematic diagram of a matrix filter using acoustic resonators.

Fig. 3B is a schematic diagram of the sub-filter of fig. 3A.

Fig. 4 is a graph of performance of an embodiment of the filter of fig. 3A, showing the resonant frequencies of the sub-filters.

Fig. 5 is a graph of the performance of an embodiment of the filter of fig. 3A, showing the passband frequencies of the sub-filters.

Fig. 6 is a schematic diagram of a matrix duplexer using acoustic resonators.

Fig. 7 is a graph of the input-output transfer function of the embodiment of the switching diplexer of fig. 6.

Fig. 8 is a schematic diagram of a matrix triplexer using acoustic resonators.

Fig. 9 is a graph of the input-output transfer function of the embodiment of the triplexer of fig. 8.

Fig. 10A is a schematic diagram of a reconfigurable switching matrix filter using acoustic resonators.

Fig. 10B is a schematic diagram of the sub-filter and switch module of fig. 10A.

Fig. 11 is a graph of input-output transfer functions for two configurations of the embodiment of the reconfigurable switching matrix filter of fig. 10A.

Fig. 12 is a block diagram of a tri-band time division duplex radio using a switching matrix triplexer.

Fig. 13 is a block diagram of a tri-band diversity receiver using a matrix triplexer.

Throughout this specification, elements appearing in the figures are assigned a three-digit or four-digit reference number in which the two least significant digits are specific for the element and one or two most significant digits are the drawing number in which the element is first introduced. Elements not described in conjunction with the figures may be assumed to have the same characteristics and functions as elements previously described with the same reference numerals.

Detailed Description

Description of the device

Fig. 1 shows a simplified schematic top view, orthogonal cross-sectional view and detailed cross-sectional view of a laterally excited thin film bulk acoustic resonator (XBAR) 100. An XBAR resonator such as resonator 100 may be used in a variety of RF filters including band reject filters, bandpass filters, diplexers, and multiplexers. XBAR is particularly suitable for use in filters for communication bands having frequencies above 3 GHz. The matrix XBAR filter described in this patent is also suitable for frequencies above 1 GHz.

The XBAR 100 is constituted by a thin film conductor pattern formed on the surface of a piezoelectric plate 110 having a front surface 112 and a rear surface 114, respectively, which are parallel. The piezoelectric plate is a thin single crystal layer of piezoelectric material such as lithium niobate, lithium tantalate, lanthanum gallium silicate, gallium nitride, or aluminum nitride. The piezoelectric plate is diced such that the orientations of the X, Y and Z crystal axes relative to the front and back surfaces are known and consistent. The piezoelectric plate may be Z-cut (that is, the Z-axis is perpendicular to the front and back surfaces 112, 114), rotary Z-cut, or rotary YX-cut. XBAR can be fabricated on piezoelectric plates with other crystal orientations.

The rear surface 114 of the piezoelectric plate 110 is attached to the surface of the substrate 120 except for the portion of the piezoelectric plate 110 that forms the diaphragm 115 that spans the cavity 140 formed in the substrate. The portion of the piezoelectric plate that spans the cavity is referred to herein as the "diaphragm" 115 because it is physically similar to the diaphragm of a microphone. As shown in fig. 1, the diaphragm 115 is continuous with the remainder of the piezoelectric plate 110 around the entire perimeter 145 of the cavity 140. In this context, "continuous" means "continuously linked without any intermediate". In other configurations, the diaphragm 115 may be continuous with the piezoelectric plate around at least 50% of the perimeter 145 of the cavity 140.

The substrate 120 provides mechanical support to the piezoelectric plate 110. The substrate 120 may be, for example, silicon, sapphire, quartz, or some other material, or combination of materials. The rear surface 114 of the piezoelectric plate 110 may be bonded to the substrate 120 using a wafer bonding process. Alternatively, the piezoelectric plate 110 may be grown on the substrate 120, or attached to the substrate in some other manner. The piezoelectric plate 110 may be attached directly to the substrate, or may be attached to the substrate 120 via one or more intermediate material layers (not shown in fig. 1).

The conventional meaning of "cavity" is "empty space within an entity". The cavity 140 may be a hole (as shown in sections A-A and B-B) completely through the substrate 120, or a recess in the substrate 120 below the diaphragm 115. For example, the cavity 140 may be formed by selective etching of the substrate 120 before or after attaching the piezoelectric plate 110 and the substrate 120.

The conductor pattern of the XBAR 100 includes an interdigital transducer (IDT) 130.IDT 130 includes a first plurality of parallel fingers (e.g., fingers 136) extending from a first bus bar 132 and a second plurality of fingers extending from a second bus bar 134. The first plurality of parallel fingers and the second plurality of parallel fingers are interleaved. The interleaved fingers overlap by a distance AP, which is commonly referred to as the "aperture" of the IDT. The center-to-center distance L between the outermost fingers of IDT 130 is the "length" of the IDT.

The first bus bar 132 and the second bus bar 134 serve as terminals of the XBAR 100. The radio frequency or microwave signal applied between the two busbars 132, 134 of the IDT 130 excites a dominant acoustic mode within the piezoelectric plate 110. The dominant acoustic mode of XBAR is the bulk shear mode, in which acoustic energy propagates in a direction substantially orthogonal to the surface of the piezoelectric plate 110, which is also perpendicular or transverse to the direction of the electric field generated by the IDT fingers. Thus, XBAR is considered to be a laterally excited thin film bulk wave resonator.

The IDT 130 is located on the piezoelectric plate 110 such that at least the fingers of the IDT 130 are disposed on the piezoelectric plate across the cavity 140 or on the diaphragm 115 suspended above the cavity 140. As shown in fig. 1, the cavity 140 has a rectangular shape whose range is larger than the aperture AP and the length L of the IDT 130. The cavity of the XBAR may have a different shape (e.g., regular or irregular polygons). The cavity of the XBAR may have more or less than four sides, which sides may be straight or curved.

A detailed cross-sectional view (detail C) shows two IDT fingers 136a, 136b on the surface of the piezoelectric plate 110. The dimension p is the "pitch" of the IDTs, and the dimension w is the width or "mark" of the IDTs. The dielectric layer 150 may be formed between the IDT fingers and optionally over the IDT fingers (see IDT finger 136 a). Dielectric layer 150 may be a non-piezoelectric dielectric material (e.g., silicon dioxide or silicon nitride). The dielectric layer 150 may be formed of multiple layers of two or more materials. IDT fingers 136a and 136b can be aluminum, copper, beryllium, gold, tungsten, molybdenum, alloys thereof, combinations thereof, or some other conductive material. Thin (relative to the total thickness of the conductor) layers of other metals (e.g., chromium or titanium) may be formed below and/or above the fingers, and/or as layers within the fingers, to improve adhesion between the fingers and the piezoelectric plate 110, and/or to passivate or encapsulate the fingers, and/or to improve power handling. The bus bars of IDT 130 can be made of the same or different materials as the fingers.

For ease of presentation in fig. 1, the geometric spacing and width of IDT fingers is greatly exaggerated relative to the length of XBAR (dimension L) and aperture (dimension AP). A typical XBAR has more than ten parallel fingers in IDT 110. The XBAR may have hundreds of parallel fingers in IDT 110. Similarly, the thickness of the finger in the cross-section is greatly exaggerated.

XBAR based on shear acoustic resonance can achieve better performance than current state-of-the-art Surface Acoustic Wave (SAW), film Bulk Acoustic Resonator (FBAR) and solid state mounted resonator bulk acoustic wave (SMR BAW) devices. In particular, the piezoelectric coupling for shear wave XBAR resonance may be higher (> 20%) compared to other acoustic resonators. The high voltage electrical coupling enables the design and implementation of various types of microwave and millimeter wave filters with considerable bandwidth.

The basic behavior of an acoustic resonator comprising XBAR is typically described using the butterworth-dyke (BVD) circuit model as shown in fig. 2A. The BVD circuit model consists of a dynamic arm and a static arm. The dynamic arm comprises a dynamic inductance L _m Dynamic capacitance C _m And resistance R _m . The static arm comprises a static capacitor C ₀ And resistance R ₀ . Although the BVD model does not fully describe the behavior of acoustic resonators, it models well the two main resonances used to design bandpass filters, diplexers, and multiplexers (a multiplexer is a filter with more than 2 input ports or output ports with multiple pass bands).

The first main resonance of the BVD model is represented by dynamic inductance L _m And dynamic capacitance C _m Dynamic resonance caused by the series combination. The second main resonance of the BVD model is represented by dynamic inductance L _m Dynamic capacitance C _m And static capacitance C ₀ Is a combination of the above. In a non-destructive resonator (R) _m ＝R ₀ =0), the frequency F of the dynamic resonance is given by the following equation _r

The frequency F of antiresonance is given by the following equation _a

Wherein γ=c ₀ /C _m Depending on the resonator structure and the type and orientation of the crystal axis of the piezoelectric material.

Fig. 2B is a graph 200 of the admittance magnitude of a theoretical lossless acoustic resonator. The acoustic resonator has a resonance 212 at a resonance frequency where the admittance of the resonator approaches infinity. Resonance is due to dynamic inductance L in the BVD model of FIG. 2A _m And dynamic capacitance C _m Caused by a series combination of (a) and (b). The acoustic resonator also exhibits antiresonance 214, wherein the admittance of the resonator is near zero. The antiresonance is formed by dynamic inductance L _m Dynamic capacitance C _m And static capacitance C ₀ Caused by a combination of (a) and (b). In a non-destructive resonator (R) _m ＝R ₀ =0), the frequency F of resonance is given by the following equation _r

The frequency F of antiresonance is given by the following equation _a

In short, a lossless acoustic resonator may be considered as a short circuit at resonant frequency 212 and an open circuit at anti-resonant frequency 214. The resonant and antiresonant frequencies in fig. 2B are representative, and the acoustic resonator may be designed for other frequencies.

Fig. 2C shows a circuit symbol for an acoustic resonator such as XBAR. This notation will be used in subsequent figures to designate each acoustic resonator in the filter schematic.

Fig. 3A is a schematic diagram of a matrix filter 300 using acoustic resonators. The matrix filter 300 comprises an array 310 of n sub-filters 320-1, 320-2, 320-n connected in parallel between a first filter port (FP 1) and a second filter port (FP 2), where n is an integer greater than one. The sub-filters 320-1, 320-2, 320-n may have successive pass bands such that the bandwidth of the matrix filter 300 is equal to the sum of the bandwidths that make up the sub-filters. An example of a SUB-filter with a continuous passband is described in U.S. patent application Ser. No. 17/362,727 entitled "TRANSVERSELY-EXCITED FILM BULK ACOUSTIC RESONATOR MATRIX FILTERS WITH SPLIT DIE SUB-FILTERS," filed on Jade 6, 29 of 2021, which is incorporated herein by reference. The sub-filters 320-1, 320-2, 320-n may have discontinuous pass bands such that the bandwidth of the matrix filter 300 is not equal to the sum of the bandwidths that make up the sub-filters, but have separate and independent pass bands separated by a stop band that exists at an input-output transfer function of the matrix filter 300 of less than-20 dB. In the subsequent examples of this patent, the sub-filters have discontinuous pass bands, and n=3. n may be less than or greater than 3 as desired to provide a desired discontinuous passband for the matrix filter 300. In some cases, the n sub-filters 320-1, 320-2, 320-n may include one or more XBARs. The filter 300 and/or the sub-filter may be an RF filter that passes a frequency band defined by the 5G NR standard.

The array of sub-filters 310 terminates at the FP1 end in acoustic resonators XL1 and XH1, which acoustic resonators XL1 and XH1 are preferably, but not necessarily, XBAR. The array of sub-filters 310 terminates at the FP2 end in acoustic resonators XL2 and XH2, which acoustic resonators XL2 and XH2 are preferably, but not necessarily, XBAR. Acoustic resonators XL1, XL2, XH1 and XH2 create "transmission zeros" at their respective resonant frequencies. The "transmission zero" is the frequency at which the input-output transfer function of the filter 300 is very low (and would be zero if the acoustic resonators XL1, XL2, XH1 and XH2 were lossless). The zero transmission may be caused by one or more of the acoustic resonators that create a very low ground impedance and, thus, in this configuration, the sub-filter is removed as a filtering component due to the substantial shorting of the acoustic resonators to ground, such that the sub-filter has no effect on the filter 300 during transmission of the zero frequency. The resonant frequencies of XL1 and XL2 are typically, but not necessarily, equal, and the resonant frequencies of XH1 and XH2 are typically, but not necessarily, equal. The resonant frequencies of the acoustic resonators XL1, XL2 are selected to provide a transmission zero adjacent the lower edge of the filter passband. XL1 and XL2 may be referred to as "low-edge resonators" because of their resonant frequency near the lower edge of the filter passband. Acoustic resonators XL1 and XL2 also act as parallel inductances to help match the impedance at the ports of the filter to the desired impedance value. In the subsequent examples of this patent, the impedance at all ports of the filter is matched to 50 ohms. The impedance may be another value (e.g., 20, 100, or 1000 ohms) if desired. The resonant frequencies of the acoustic resonators XH1, XH2 are selected to provide transmission zeroes at or above the high edges of the filter passband. XH1 and XH2 may be referred to as "high edge resonators" because of their resonant frequency near the high edge of the filter passband. High-edge resonators XH1 and XH2 are not required in all matrix filters, e.g. filters with sub-filters that will not pass the relative amplitudes of the signals at these high edge frequencies.

Fig. 3B is a schematic diagram of a sub-filter 350 suitable for each of sub-filters 320-1, 320-2, and 320-n of filter 300. The sub-filter 350 comprises three acoustic resonators XA, XB, XC connected in series between a first sub-filter port (SP 1) connectable to FP1 and a second sub-filter port (SP 2) connectable to FP 2. The acoustic resonators XA, XB, XC are preferably, but not necessarily, XBAR. The sub-filter 350 comprises two coupling capacitors CA, CB, each of which is connected between ground and a respective node between two of the acoustic resonators. The inclusion of three acoustic resonators in sub-filter 350 is exemplary. The sub-filter may have m acoustic resonators, where m is an integer greater than 1. A sub-filter with m acoustic resonators includes m-1 coupling capacitors. The m acoustic resonators of the sub-filter are connected in series between the two ports SP1 and SP2 of the sub-filter, and each of the m-1 coupling capacitors is connected between ground and a node between a corresponding pair of the m acoustic resonators.

Compared to other types of acoustic resonators, XBAR has a very high electromechanical coupling (which results in a large difference between the resonant frequency and the antiresonant frequency), but a low capacitance per unit area. As shown in fig. 3A and 3B, the matrix filter architecture takes advantage of the high electromechanical coupling of XBAR without requiring high resonator capacitance.

Fig. 4 is a graph 400 of performance 405 of an exemplary embodiment of a matrix filter implemented using XBAR for all acoustic resonators. The performance 405 in graph 400 may be the performance of the transfer function S21 of the filter 300 with 3 discrete pass-band sub-filters 1, 2 and 3. Specifically, the performance 405 includes a solid line 410, a dashed line 420, and a dotted line 430, the solid line 410, the dashed line 420, and the dotted line 430 being graphs of S21 (FP 1 to FP2 transfer functions) of the filter according to frequency, wherein each of the lines 410, 420, and 430 is for discontinuous passband sub-filters 1, 2, and 3, respectively. That is, the solid line 410 is a graph of FP1 to FP2 individual transfer functions S21 of the sub-filter 1 of the filter according to frequency and having the individual resonance frequency SF 1. The dashed line 420 is a graph of FP 1-FP 2 individual transfer functions S21 of sub-filter 2 according to frequency and having a filter of individual resonance frequency SF 2. The dotted line 430 is a graph of FP 1-FP 2 individual transfer functions S21 of sub-filter 3 according to frequency and having a filter of individual resonance frequency SF 3. Since the exemplary filter is symmetrical, the solid line 410, the dashed line 420, and the dotted line 430 are also graphs of S12.

Fig. 5 is a graph 500 of passband frequencies of sub-filters within the exemplary matrix filter of the filter of fig. 3A (the performance of which is shown in fig. 4). The example of graph 500 may be for receive frequencies (from low frequency to high frequency) of LTE bands 3, 1, and 7 having a passband defined above-3 dB. Specifically, P1, P2, and P3 are passband frequencies with magnitudes of input-output transfer functions of the sub-filter 1, the sub-filter 2, and the sub-filter 3, respectively, higher than-3 dB. The pass bands P1, P2 and P3 are discontinuous because each pair of adjacent pass bands are separated by a stop band in which the input-output transfer function of the matrix filter is less than-20 dB. For example, the pair of pass bands are discontinuous because pass bands P1 and P2 are separated by a stop band SB1 that exists at an input-output transfer function S21 of matrix filter 300 that is less than-20 dB. Furthermore, the pair of pass bands is discontinuous because the pass bands P2 and P3 are separated by the stop band SB2 that exists at an input-output transfer function S21 of the matrix filter 300 that is less than-20 dB. As described above, the present application also contemplates embodiments in which the sub-filters 320-1, 320-2, 320-n may have a continuous passband.

The matrix filter for fig. 4 and 5 includes 3 sub-filters with connections between input and output ports that can be switched in and out to provide multiple pass bands for input and output RF communication signals. Each sub-filter may include three XBARs as shown in fig. 3A and 3B. In other cases, there may be two, four, five or up to ten sub-filters. Furthermore, each sub-filter may comprise more than three XBARs and two coupling capacitors. Some of the sub-filters may have: m acoustic resonators (where m=4, 5, 6 or up to 10); and corresponding m-1 coupling capacitors as shown in fig. 3B. In this and all subsequent examples, filter performance was determined by modeling the filter using the BVD model (fig. 2A) for XBAR. It is to be understood that the concepts herein with respect to 3 sub-filters can be extended to just two, or up to four, five, or up to any number determined by size and routing complexity considerations.

The input-output transfer function of an exemplary filter (as shown in fig. 4) is the vector sum of the input-output transfer functions of three sub-filters having discrete pass bands. The discontinuous pass band may indicate that a single sub-filter input-output transfer function does not cross another sub-filter input-output transfer function at frequencies where the two filters S21 transfer functions are above-20 dB. For this purpose, the input-output transfer functions of the sub-filter 1 and the sub-filter 2 cross at a frequency slightly below 2GHz, at which (a) S21 of the two filters is not higher than-20 dB, and (b) the phases of the input-output transfer functions of the two filters are substantially equal. In this context, "not above" means sufficiently below to not cause an objectionable change in the transfer function of any one sub-filter due to the transfer function of a different sub-filter of the matrix filter within the filter passband range (in this case 1.5 to 3 GHz). The quantitative value of "no higher" may be different for different filter applications. Similar requirements apply for sub-filter 2 and sub-filter 3. In a matrix filter with more than three sub-filters, similar requirements apply for each pair of adjacent (in frequency) filters in the sub-filters.

In some cases, a "continuous" passband matrix filter describes a matrix filter having a passband that is the sum of the passbands of more than one sub-filter, while a "discontinuous" passband matrix filter describes a matrix filter in which each passband is the passband of only one sub-filter. For some switched matrix filters, the pass bands of the "discontinuous" sub-filters are not adjacent or overlapping above-20 dB. The matrix filter may also have some sub-filters that are continuous and other sub-filters that are discontinuous. For example, the matrix filter may be a filter having at least one stop band between the pass bands of at least one pair of adjacent sub-filters.

In one example, the lowest passband (discontinuous passband sub-filter 1) is the LTE band Rx 3 and has 3 resonators and 2 coupling capacitors. Here, the intermediate passband (discontinuous passband sub-filter 2) is the LTE band Rx 1 and has 5 resonators and 4 coupling capacitors. The highest pass band (discontinuous pass band sub-filter 3) is the LTE band Rx 7 and the sub-filter has 4 resonators and 3 coupling capacitors. The filter may have one or more XL resonators, and zero or more XH resonators.

The exemplary matrix filter is symmetrical because the impedance at both port 1 and port 2 is equal to 50 ohms. The matrix filter may also be designed to have significantly different impedances at port 1 and port 2, in which case the internal circuit will be asymmetric. The vertical dash-dot line identifies the resonant frequency of the XBAR within the exemplary matrix filter. The line labeled "XL" identifies the resonant frequency of resonators XL1 and XL2, which is adjacent to the lower edge of the filter passband. Similarly, the line labeled "XH" identifies the resonant frequencies of resonators XH1 and XH2, which are adjacent to the upper edge of the filter passband. The two lines labeled "SF1" in fig. 4 identify the resonant frequencies of the XBARs within the individual sub-filters 1. The two lines labeled "PBF1" in fig. 5 identify the pass band frequencies of the XBAR within the individual sub-filters 1. Note that both resonant frequencies are lower than the center of the passband. This is because the resonance frequency of the series resonator and the capacitor is higher than that of the individual resonator. Similarly, the two lines labeled "SF2" identify the resonant frequencies of the XBARs within the sub-filter 2, and the two lines labeled "SF3" identify the resonant frequencies of the XBARs within the sub-filter 3. Similarly, the two lines labeled "SF2" in fig. 4 identify the resonant frequencies, and the two lines labeled "PBF2" in fig. 5 identify the passband frequencies of the XBAR within the sub-filter 2. Finally, the two lines labeled "SF3" in fig. 4 identify the resonant frequencies, and the two lines labeled "PBF3" in fig. 5 identify the pass band frequencies of the XBAR within the sub-filter 3.

Fig. 6 is a schematic diagram of a matrix filter 600 configured as a diplexer. Matrix filter 600 includes an array 610 of three sub-filters 620-1, 620-2, 620-n. The sub-filter 1620-1 is connected between the first filter port (FP 1) and the second filter port (FP 2). Sub-filter 2 620-2 and sub-filter 3 620-3 are connected in parallel between FP1 and the third filter port (FP 3). FP1 is the common port or input port of the diplexer, and FP2 and FP3 are branch ports or output ports. The array of sub-filters 610 terminates at both ends in XBAR XL and XH as previously described.

Fig. 7 is a graph 700 of performance 705 of an example of matrix filter duplexer 600 of fig. 6. In this example, XL, XH and the three sub-filters are identical to the corresponding elements of matrix filter 300 of FIG. 3A. In fig. 7, a solid line 410 under 710 is a graph of S21 (FP 1 to FP2 transfer functions) according to frequency. The dashed line 420 and the dotted line 430 below 720 are graphs of S31 (FP 1 to FP3 transfer functions) according to frequency. Since the exemplary filter is symmetrical, the solid line 410 below 710, and the dashed and dotted lines 420 and 430 below 720 are also graphs of S12 and S13, respectively. The switched matrix filter 600 is exemplary. In most applications, the diplexer will have the same number (two, three or more) of sub-filters in parallel between the common port and the two branch ports.

FP1 may be considered a common port of matrix filter duplexer 600. FP2 may be considered a "low band" port and FP3 may be considered a "high band" port. When the matrix filter duplexer is used in a frequency division duplex radio, one of FP2 and FP3 may be a reception port of the duplexer and the other of FP2 and FP3 may be a transmission port of the duplexer, depending on frequencies allocated for reception and transmission.

In a second diplexer configuration, which is a variation of filter 600, sub-filter 1 620-1 and sub-filter 2620-2 are connected in parallel between FP1 and FP 2. Here, sub-filter 3620-3 is connected between FP1 and FP 3. In this case, the graph of the performance of the example of the matrix filter duplexer has a solid line 410 and a broken line 420 as a graph of S21 according to frequency; and a dotted line 430 as a graph of S31 according to frequency.

In the third duplexer configuration as a modification of the filter 600, the sub-filter 1 620-1 and the sub-filter 3620-3 are connected in parallel between FP1 and FP 2. Here, the sub-filter 2620-2 is connected between FP1 and FP 3. In this case, the graph of the performance of the example of the switching matrix filter duplexer has a solid line 410 and a dotted line 430 as a graph of S21 according to frequency; and a dotted line 420 as a graph of S31 according to frequency.

Because either of the branch ports FP2 or FP3 may be selected or switched to the output of the filter, the diplexer filter 600 and both variants are switched matrix filters. For example, a sub-filter connection between an input port and an output port may be switched in and out to provide multiple pass bands for input and output RF communication signals.

Fig. 8 is a schematic diagram of a matrix triplexer filter 800 using acoustic resonators. Matrix filter 800 includes an array 810 of three sub-filters 820-1, 820-2, 820-n. Sub-filter 1 820-1 is connected between first filter port (FP 1) and second filter port (FP 2). Sub-filter 2 820-2 is connected between FP1 and third filter port (FP 3). Sub-filter 3 820-3 is connected between FP1 and fourth filter port (FP 4). The array of sub-filters 810 terminates at both ends in XBAR XL and XH as previously described. FP1 is the common port or input port of the multiplexer, and FP2, FP3 and FP4 are the branch ports or output ports of the multiplexer. The multiplexer may have more than three branch ports. A multiplexer with two branch ports is commonly referred to as a "diplexer" and a multiplexer with three branch ports may be referred to as a "triplexer".

Fig. 9 is a graph 900 of performance of an example of the function of the embodiment of triplexer filter 800 of fig. 8. In this example, XL, XH and the three sub-filters are identical to the corresponding elements of matrix filter 300 of FIG. 3A. In fig. 9, the solid line below 910 is a graph of S21 (FP 1 to FP2 transfer function) according to frequency. The dashed line below 920 is a graph of S31 (FP 1 to FP3 transfer function) according to frequency. The dotted line below 930 is a graph of S41 (FP 1 to FP4 transfer function) according to frequency. Since the exemplary filter is symmetrical, the solid line below 910, the dashed line below 920, and the dotted line below 930 are also graphs of S12, S13, and S14, respectively.

FP1 may be considered a common port of the matrix filter. FP2 may be considered a "low band" port, FP3 may be considered a "mid band" port, and FP4 may be considered a "high band" port. When the matrix filter is used in a Frequency Division Duplex (FDD) radio, one of FP2, FP3, and FP4 may be a receiving port and the other one of FP2, FP3, and FP4 may be a transmitting port, depending on the frequencies allocated for reception and transmission. In other cases, in FDD radio, two of FP2, FP3, FP4 may be receiving ports and another one of FP2, FP3, FP4 may be transmitting ports; or vice versa.

In an additional multiplexer configuration that is a variation of filter 800, any one or more of sub-filter 1 820-1, sub-filter 2 820-2, and sub-filter 3 820-3 may be connected in parallel between FP1 and FP2, FP3, and/or FP 4. In this case, the graph of the performance of the example of the matrix filter duplexer has corresponding lines among the solid line 410, the broken line 420, and/or the dotted line 430 as graphs of S21, S31, and/or S41 according to frequencies.

Because any one or more of ports FP2, FP3, and FP4 may be selected or switched to the output of the filter, multiplexer filter 800 and both variants may be switched matrix filters. For example, a sub-filter connection between an input port and an output port may be switched in and out to provide multiple pass bands for input and output RF communication signals. In one example, a switched XBAR matrix filter with 3 sub-filters for LTE bands 3, 1, and 7 provides a multi-passband reconfigurable filter that is configurable for all 7 possible states: only 1, only 3, only 7, 1+3, 1+7, 3+7, and 1+3+7. The filter has low loss due to its matrix architecture (e.g. due to the position of the switches, and because the filter does not require inductors). The filter also has an output impedance that is matched to the LNA, so no external impedance matching is required.

For example, fig. 10A is a schematic diagram of a reconfigurable switching matrix filter 1000 using XBAR. Reconfigurable switching matrix filter 1000 includes an array 1010 of n sub-filters/switching circuits 1020-1, 1020-2, 1020-n connected in parallel between a first filter port (FP 1) and a second filter port (FP 2), where n is an integer greater than one. In the following example, n=3. In other cases, n may be greater than 3 as needed to provide the desired bandwidth for the reconfigurable matrix filter 1000. Each sub-filter/switch circuit acts as a discontinuous band-pass filter that can be selectively enabled (i.e., connected between FP1 and FP 2) or disabled (i.e., not connected between FP1 and FP 2). The array of sub-filters/switching circuits 1010 terminate at both ends with XBAR XL and XBAR XH as previously described.

The sub-filters/switch circuits 1020-1, 1020-2, 1020-n have discontinuous pass bands such that when all sub-filters/switch modules are enabled, the bandwidth of the matrix filter 1000 is not equal to the sum of the bandwidths that make up the sub-filters, but rather has separate and independent pass bands separated by a stop band that exists at an input-output transfer function of the matrix filter 300 of less than-20 dB. One or more of the sub-filters/switching circuits may be disabled to customize the matrix filter bandwidth, or to insert a notch or stop band within the entire passband (e.g., to provide a desired discontinuous passband for the matrix filter). The filter 300 and/or the sub-filter may be an RF filter that passes a frequency band defined by the 5G NR standard.

Fig. 10B is a schematic diagram of a sub-filter/switch circuit 1050 suitable for sub-filter/switch circuits 1020-1, 1020-2, and 1020-n in fig. 10A. The sub-filter/switch circuit 1050 includes three acoustic resonators X1, X2, X3 connected in series between the first sub-filter port (SP 1) and the second sub-filter port (SP 2), and coupling capacitors C1, C2 connected to ground from a junction between adjacent acoustic resonators. The inclusion of three acoustic resonators in the sub-filter/switch circuit 1050 is exemplary, and the sub-filter/switch circuit may have more than three acoustic resonators. When the sub-filter/switch circuit comprises more than three acoustic resonators, the number of coupling capacitors will be one less than the number of acoustic resonators. The acoustic resonators X1, X2, X3 are preferably, but not necessarily, XBAR.

The sub-filter/switch circuit 1050 includes a switch SW1 in parallel with the first parallel capacitor C1, and a switch SW2 in parallel with the last parallel capacitor C2. When switches SW1 and SW2 are open, the sub-filter/switch circuit operates as a sub-filter suitable for use in any of the previous examples. In this case, the sub-filter/switch circuit connection between the input port and the output port is switched in to provide the pass band of the sub-filter for the input and output RF communication signals. When switches SW1 and SW2 are closed, the sub-filter/switch circuit presents the appropriate impedance to SP1 and SP2, but with an open input-output transfer function. In this case, the sub-filter/switch circuit connection between the input port and the output port is cut out and the pass band of the sub-filter is not provided for the input and output RF communication signals.

When the sub-filter/switch circuit comprises more than two parallel capacitors, the switch is connected in parallel with two parallel capacitors next to the acoustic resonator connected to the two sub-filter ports. In other words, the switch is connected in parallel with the "first parallel capacitor" and the "last parallel capacitor", which are not in the middle of the resonator string, but only inside the two "end acoustic resonators" at both ends of the string. In some cases, filter 1000 may be described as having its respective output ports SP2 connected to all of its sub-filters of common output port FP 2. For example, the first switch is connected in parallel with a first capacitor parallel element between an XBAR series element immediately adjacent to the filter input port and an XBAR series element further from the input port; and wherein the second switch is in parallel with a last capacitor parallel element between an XBAR series element immediately adjacent to the output port of the filter and an XBAR series further from the output port.

Fig. 11 is a graph 1100 of performance 1105 of an example of the reconfigurable switching matrix filter duplexer 1000 of fig. 10. In this example, the components within XL, XH and the three sub-filters/switching circuits are identical to the corresponding elements of matrix filter 300 of FIG. 3A. In fig. 11, the solid line below 1110 is a graph of S21 (port 1 to port 2 transfer function) of the filter according to frequency when the sub-filter/switch circuit 1 is enabled and the sub-filters/switch circuits 2 and 3 are disabled. The dashed line below 1120 is a plot of S21 as a function of frequency when sub-filter/switch circuit 2 is enabled and sub-filters/switch circuits 1 and 3 are disabled. The dashed line below 1130 is a plot of S21 as a function of frequency when sub-filter/switch circuit 3 is enabled and sub-filters/switch circuits 1 and 2 are disabled. The sum of the two curves below 1110 and 1130 (not shown but easily imagined) is the port 1 to port 2 transfer function as a function of frequency when the sub-filters/switching circuits 1 and 3 are enabled and the sub-filter/switching circuit 2 is disabled. The sum of the two curves under 1110 and 1120 is the port 1 to port 2 transfer function as a function of frequency when sub-filters/switching circuits 1 and 2 are enabled and sub-filter/switching circuit 3 is disabled. The sum of the two curves under 1110 and 1130 is the port 1 to port 2 transfer function as a function of frequency when sub-filters/switching circuits 1 and 3 are enabled and sub-filter/switching circuit 2 is disabled. The sum of the two curves below 1120 and 1130 is the port 1 to port 2 transfer function as a function of frequency when sub-filters/switching circuits 2 and 3 are enabled and sub-filter/switching circuit 1 is disabled. The sum of the three curves under 1110, 1120 and 1130 is the port 1 to port 2 transfer function as a function of frequency when the sub-filters/switching circuits 1, 2 and 3 are enabled. The open-circuited input-output transfer function is a port 1 to port 2 transfer function as a function of frequency when all of the sub-filters/switching circuits 1, 2 and 3 are disabled. By enabling various combinations of three sub-filters/switching circuits, a total of eight different filter configurations are possible.

Fig. 12 is a schematic block diagram of a tri-band Time Division Duplex (TDD) radio 1200 using a switching matrix triplexer. TDD radios transmit and receive in the same frequency channel within a specified communication band. The radio 1200 includes a switching matrix triplexer 1210 having a first filter port FP1 configured to connect to an antenna 1205 and a second filter port FP2 coupled to a transmit/receive (T/R) switch 1215. For example, the switch matrix band pass filter 1210 may be the switch matrix band pass filter of fig. 10A. The T/R switch 1215 connects the second port of the matrix bandpass filter 1210 to the output of the transmitter 1220 or to the input of the receiver 1225. The T/R switch 1215, transmitter 1220 and receiver 1225 are governed by a processor 1230 that performs media access control functions. In particular, processor 1230 controls the operation of T/R switch 1215 and switching matrix bandpass filter 1210. When the switching matrix bandpass filter 1210 is reconfigurable, the processor 1230 may control the operation of the switches within the bandpass filter. The antenna 1205 may be part of the radio 1200 or external to the radio 1200.

The radio 1200 is configured for operation in three designated communication bands. The switch matrix bandpass filter 1210 has internal switches that allow selection of one of three pass bands containing a specified communication band, and one or more stop bands to block specified frequencies outside the specified communication band. Preferably, the switched bandpass filter 1210 has low loss in its passband and high rejection in its stopband. Furthermore, the switching band pass filter 1210 must be compatible with TDD operation, that is, stable and reliable in delivering RF power generated by the transmitter 1220. The switching matrix band pass filter 1210 may be the switching matrix filter 300 of fig. 3A or the reconfigurable switching matrix filter 1000 of fig. 10A implemented using acoustic resonators, which may be XBAR.

The switching matrix band pass filter 1210 may be a reconfigurable matrix filter as shown in fig. 10A. The use of a reconfigurable filter will allow the bandwidth of the filter to be set to a single LTE or 5G NR band, allowing the same transmitter and receiver to be used for three separate bands. The switches within the sub-filters may be controlled by processor 1230.

Fig. 13 is a schematic block diagram of a tri-band diversity receiver 1300 using a matrix triplexer. The tri-band diversity receiver radio receives in three different frequency ranges corresponding to the three communication bands. The receiver 1300 includes an antenna 1305, a matrix filter triplexer 1310, the matrix filter triplexer 1310 having a common filter port FP1 configured to connect to the antenna 1305, a first receiver filter port FP2 coupled to an input of the receiver 1320, a second receiver filter port FP3 coupled to an input of the receiver 1325, and a third receiver filter port FP4 coupled to an input of the receiver 1330.

The receiver 1300 is configured to operate in a designated communication band. Matrix filter triplexer 1310 includes a receive filter coupled between each of: FP1 and FP2; FP1 and FP3; and FP1 and F4. The receive filter includes a discontinuous passband receive sub-filter. Matrix filter triplexer 1310 may be implemented using an acoustic resonator, which may be an XBAR.

Matrix filter duplexer 1310 may be matrix triplexer 800 of fig. 8. The FP1, FP2, FP3, and FP4 ports of matrix filter duplexer 1310 may be the FP1, FP2, FP3, and FP4 ports of matrix multiplexer 800.

In another case, matrix filter duplexer 1310 may be similar to reconfigurable switched filter 1000 of fig. 10A, with the same number of sub-filters in the transmit and receive filters. The FP1 port of matrix filter duplexer 1310 may be FP1 of reconfigurable switching filter 1000; and the FP2, FP3, and FP4 ports of matrix filter duplexer 1310 may be FP2 of reconfigurable switching filter 1000.

The acoustic resonator matrix filter topologies herein (e.g., the acoustic resonator matrix filter topologies of filters 300, 600, 800, and/or 1000) may reduce the size of the resonators in the filter, thus: the component cost for the filter and the manufacturing cost of the filter are reduced; a filter is provided having a passband that is very insensitive to switching losses; a filter is provided having an achievable impedance transformation for matching the impedance at the input and output of the filter; and providing a filter that matches the minimum noise figure of the output connected to the LNA without any matching inductor. These topologies allow multiple pass bands to be switched in and out for input and output RF communication signals without the need for inductors (e.g., between coupling capacitors and ground), in one example, an XBAR matrix filter with 3 sub-filters for LTE bands 3, 1, and 7 provides a multi-pass band reconfigurable filter configurable for all 7 possible states (only 1, only 3, only 7, 1+3, 1+7, 3+7, and 1+3+7). The filter has low loss due to its matrix architecture (e.g. due to the position of the switches, and because the filter does not require inductors). The filter also has an output impedance that is matched to the LNA, so no external impedance matching is required.

Ending comments

Throughout this specification, the embodiments and examples shown should be considered as exemplars, rather than limitations on the apparatus and processes disclosed or claimed. While many of the examples presented herein relate to specific combinations of method acts or system elements, it should be understood that these acts and these elements may be combined in other ways to achieve the same objectives. With respect to the flowcharts, additional and fewer steps may be taken, and the steps shown may be combined or further refined to implement the methods described herein. Acts, elements and features discussed only in connection with one embodiment are not intended to be excluded from a similar role in other embodiments.

As used herein, "plurality" means two or more. As used herein, a "collection" of items may include one or more such items. As used herein, the terms "comprising," "including," "carrying," "having," "containing," "involving," and the like, whether in the written description or the claims, are to be construed as open-ended, i.e., to mean including but not limited to. Only the transitional phrases "consisting of … …" and "consisting essentially of … …" are transitional phrases with respect to the closed or semi-closed state of the claims, respectively. Use of ordinal terms such as "first," "second," "third," etc., in the claims to modify a claim element does not by itself connote any priority, precedence, or order of one claim element over another or the temporal order in which acts of a method are performed, but are used merely as labels to distinguish one claim element having a certain name from another element having a same name (but for use of the ordinal term) to distinguish the claim elements. As used herein, "and/or" means that the listed items are alternatives, but alternatives also include any combination of the listed items.

Claims

1. A matrix filter, comprising:

a filter input port; and

two or more sub-filters connected between the filter input port and the respective filter output port, each sub-filter comprising a ladder circuit having n laterally excited thin film bulk acoustic resonator, XBAR, series elements and n-1 capacitor parallel elements, wherein the order n of the sub-filters is an integer greater than 2,

wherein each sub-filter further comprises a first switch connected in parallel with the first capacitor parallel element and a second switch connected in parallel with the last capacitor parallel element.

2. The filter of claim 1, wherein the two or more sub-filters have discontinuous pass bands; and wherein each of the discrete pass bands is the pass band of only one sub-filter.

3. The filter of claim 2, wherein the two or more sub-filters have a continuous passband.

4. The filter of claim 1, wherein only one of the two or more sub-filters is selected to be connected between the filter input port and a respective filter output port; and wherein there is no crossing of the input-output transfer function of a single sub-filter with the input-output transfer function of the other sub-filter at frequencies where the transfer functions of the two filters are higher than-20 dB.

5. The filter of claim 1, wherein each of the two or more sub-filters has a discontinuous passband separated by a stopband, the stopband being present at an input-output transfer function of the matrix filter of less than-20 dB; and wherein sub-filter connections between the filter input ports and the respective filter output ports can be switched to select one or more of the discontinuous pass bands.

6. The filter of claim 1, wherein the first capacitor shunt element is between an XBAR series element immediately adjacent the filter input port and an XBAR series element further from the filter input port; and wherein the last capacitor shunt element is between an XBAR series element immediately adjacent to the respective filter output port and an XBAR series element further from the respective filter output.

7. The filter according to claim 6, wherein,

the XBAR series element includes a first end XBAR connected to the first sub-filter port, a second end XBAR connected to the second sub-filter port, and one or more intermediate XBARs connected between the first end acoustic resonator and the second end acoustic resonator.

8. The filter of claim 1, wherein each respective filter output port is connected to a common output port.

9. The filter according to claim 1, wherein,

each of the XBAR series elements is connected in series between a first sub-filter port and a second sub-filter port;

each of the shunt capacitors is connected between ground and a node between a respective pair of XBAR series elements; and

each of the first switch and the second switch is connected between ground and the node between the respective pair of XBAR series elements.

10. The filter of claim 1, further comprising:

a first XBAR connected between the first filter port and ground; and

a second XBAR connected between the second filter port and ground, wherein,

the first XBAR and the second XBAR are configured to produce respective transmission zeros adjacent a lower edge of a passband of the filter.

11. A filter, comprising:

a first filter port and a second filter port;

n sub-filters, where n is an integer greater than 1, each of the n sub-filters having a first sub-filter port connected to the first filter port and a second sub-filter port connected to the second filter port; wherein each sub-filter comprises a ladder circuit having at least three laterally excited thin film bulk acoustic resonator, XBAR, series elements and at least two capacitor parallel elements;

Wherein each of the n sub-filters further comprises a first switch connected in parallel with the first capacitor parallel element and a second switch connected in parallel with the last capacitor parallel element.

12. The filter of claim 11, wherein the n sub-filters have discontinuous pass bands; and wherein each of the discrete pass bands is the pass band of only one sub-filter.

13. The filter of claim 11, wherein each of the n sub-filters has a discontinuous passband separated from the passbands of all other of the n sub-filters by a stopband, the stopband being present at an input-output transfer function of the matrix filter of less than-20 dB.

14. The filter of claim 11, wherein only one of the n sub-filters is selected to connect between the first filter port and the second filter port; and wherein there is no crossing of the input-output transfer function of a single sub-filter with the input-output transfer function of the other sub-filter at frequencies where the transfer functions of the two filters are higher than-20 dB.

15. The filter of claim 11, further comprising:

A first XBAR connected between the first filter port and ground; and

a second XBAR connected between the second filter port and ground,

wherein the first and second XBARs are configured to produce respective transmission zeroes adjacent a lower edge of a passband of the filter, wherein the first and second acoustic resonators are laterally excited thin film bulk acoustic resonators XBARs.

16. The filter of claim 11, wherein,

the pass band of the filter is selected to be equal to only one of the discrete pass bands of the n sub-filters.

17. The filter of claim 11, further comprising:

a third XBAR connected between the first filter port and ground; and

a fourth XBAR connected between the second filter port and ground,

wherein the third XBAR and the fourth XBAR are configured to produce a transmission zero adjacent an upper edge of a passband of the filter.

18. The filter of claim 11, wherein,

each of the XBAR series elements is connected in series between the first sub-filter port and the second sub-filter port;

19. The filter of claim 18, wherein,

the XBAR series element includes a first end XBAR connected to the first sub-filter port, a second end XBAR connected to the second sub-filter port, and one or more intermediate XBARs connected between the first end acoustic resonator and the second end acoustic resonator, and

the first capacitor parallel element is between an XBAR series element immediately adjacent to the first sub-filter port and an XBAR series element further from the first sub-filter port; and wherein the last capacitor shunt element is between an XBAR series element immediately adjacent to the second sub-filter port and an XBAR series element further from the second sub-filter port.

20. A tri-band diversity receiver comprising:

a matrix triplexer coupled between an antenna and three receivers, the triplexer comprising:

a first sub-filter coupled between the first filter port and a second filter port coupled to the first receiver;

A second sub-filter coupled between the first filter port and a third filter port coupled to a second receiver;

a third sub-filter coupled between the first filter port and a fourth filter port coupled to a third receiver;

wherein each sub-filter comprises a ladder circuit having at least three laterally excited thin film bulk acoustic resonator, XBAR, series elements and at least two capacitor parallel elements;

21. The filter of claim 20, wherein the first, second, and third sub-filters each have a discontinuous passband separated from the passbands of all other of the first, second, and third sub-filters by a stopband, the stopband being present at an input-output transfer function of the matrix filter of less than-20 dB.

22. The filter of claim 20, wherein,

23. The filter of claim 20, wherein,

for the first sub-filter, the first capacitor parallel element is between an XBAR series element immediately adjacent to a first port and an XBAR series element farther from the first port; and wherein the last capacitor parallel element is between an XBAR series element immediately adjacent to a second port and an XBAR series element further from the second port;

for the second sub-filter, the first capacitor parallel element is between an XBAR series element immediately adjacent to the first port and an XBAR series element farther from the first port; and wherein the last capacitor parallel element is between an XBAR series element immediately adjacent to a third port and an XBAR series element further from the third port; and

for the third sub-filter, the first capacitor parallel element is between an XBAR series element immediately adjacent to the first port and an XBAR series element farther from the first port; and wherein the last capacitor parallel element is between an XBAR series element immediately adjacent to the fourth port and an XBAR series element further from the fourth port.

24. The filter of claim 20, wherein the first, second, and third sub-filters have a continuous passband.