Background Art A switchable power combiner, typically being used in an amplification stage which controls the output power of a base station in a communication system, combines a plurality of signals provided by respective power amplifiers. In other words, the switchable power combiner carries out, along with a counterpart power divider, the functions dividing and combining of signals while providing a redundant signal path against an abnormality of any power amplifier.

In a Wilkinson type divider/combiner having fixed signal paths, the optimum characteristic impedance of impedance matching lines are determined based on the number of signal paths to be divided or combined. For example, the characteristic impedances of the impedance matching lines are determined to be {2Zo in case of 2-way divider/combiner and {3Z0 in case of 3-way divider/combiner, where 7 is the common characteristic impedance, 50Q. On the other hand, in the case that the number of combined signal paths is variable rather than being fixed, the average of the optimum values for different operation modes have been chosen as the characteristic impedances of the impedance

matching lines. Alternatively, a combination of a single path transmission line and the average-matched transmission lines have been used as well.

FIG. 1 illustrates an example of the average-matched 3-way switchable power combiner, which has a shape that switches are added to a typical 3-way combiner having fixed paths so as to perform the switching function. One or more desired signals of the input signals received through input ports PIl, PI2, and PI3 are selected by switches Sl through S6 to be combined, and a combined signal is output through an output port PO.

The number of combined paths is determined by the number of power amplifiers being operated in the system. Switching control signals for controlling the switching operation of the switches S 1 through S6 are provided by a controller not shown in the Figure. For example, when the signals received through the input ports PIl and PI2 are to be combined, the switches S1, S2, S4, and S5 are turned on while the switches S3 and S6 are turned off and the combining signal is output through the output port PO. Meanwhile, isolation resistors RI, R2, and R3 and terminators 14,16, and 18 are provided after the input ports PIl, PI2, and PI3. Additionally, each of the switch pairs S 1 and S4, S2 and S5, and S3 and S6 operate simultaneously responding to a respectively common switching control signal.

As mentioned above, however, the characteristic impedance Zm of each of the impedance matching lines 4,6, or 8 shown in FIG. 1 is generally determined to be the average, Zo/2 (#2+#3Z0) (=78.7Q), of {24 (=70.7Q) and 13. g (=86.6Q) which are optimal impedances for 2-way and 3-way combining operation modes, respectively. While the average matching may provide satisfactory reflection loss requirements in 2-way or 3- way operation mode, it may be difficult to use the combiner for 1-way operation mode

since the reflection loss is increased. To avoid such a problem, some devices employ a combination of a single path transmission line and the average-matched transmission lines alternatively as mentioned above, which, however, cannot accomplish the perfect matching condition for all the operation modes either.

To overcome such a problem, the applicants of the present invention described, in an International Patent Application No. PCT/KR/01631 filed on September 27,2001 and entitled"SWITCHABLE POWER COMBINER,"a switchable power combiner having plural matching lines each of which is selectively activated for respective operation mode. FIG. 2 illustrates an example of the switchable power combiner described in the specification and drawings of the International Application. Switches SW11, SW12, and SW13 select desired signals of the input signals received through the input ports PIl, PI2, and PI3, so that the selected signals are combined at a combining node 30.

The combined signal is provided to an impedance matching unit through a connection line 32 having a characteristic impedance of Zo and an electrical length of A/4.

The impedance matching unit includes three matching lines 34, 36, and 38, each of which corresponds to respective operation mode, i. e., the number of combined paths. Switches S21 through S33 carry out switching functions so that just a single one of the three matching lines 34,36, and 38 connect the connection line 32 to an output port PO. A first matching line 34, effectively operable in the 1-way operation, has a characteristic impedance ofZo. A second and a third matching lines 36 and 38, respectively operable in the 1-way and 2-way operations, has characteristic impedances expressed by equations 2 and 3, respectively. All the matching lines 34,36, and 38 have electrical length of A/4.

We review, in more detail, the 2-way operation with reference to FIG. 3A. The impedance viewed from the combining node 30 toward the side of input ports is 25 Q (=50/2 Q) since two resistive transmission lines of 50 Q are connected in parallel. Such impedance of 25 Q corresponds to point PI in the Smith chart of FIG. 3A. The point PI is transformed to a point P2 by the connection line 32 having the characteristic impedance of 50 Q and the length of A/4. Further, the point P2 is transformed to a point P3 by the matching line 36 having the characteristic impedance of 70.7 Q and the length of A/4, so that the internal characteristic impedance of the combiner is exactly 50 Q and the impedance matching is fulfilled.

Similarly, we review the 2-way operation with reference to FIG. 3B, in more detail.

The impedance viewed from the combining node 30 toward the side of input ports is 16.6 Q (=50/3 Q) since three resistive transmission lines of 50 Q are connected in parallel. Such impedance of 16.6 Q corresponds to point PI in the Smith chart of FIG. 3B. The point PI

is transformed to a point P2 by the connection line 32 having the characteristic impedance of 50 Q and the length of A/4. Further, the point P2 is transformed to a point P3 by the matching line 38 having the characteristic impedance of 86.6 Q and the length of A/4, so that the internal characteristic impedance of the combiner is exactly 50 Q and the impedance matching is fulfilled.

The switchable power combiner shows optimum impedance matching characteristics for all the operation modes. However, the switchable power combiner may have a drawback that the insertion loss is large because of lots of transmission lines and connectors. Furthermore, the assembling process may be complex or require much labors because separate transmission line having characteristic impedance different from one another is to be provided for each operation mode.

Disclosure of the Invention To solve the above problems, the object of the present invention is to provide a switchable power combiner in which the impedance matching is efficiently fulfilled and insertion loss is minimized.

The switchable power combiner to achieve the above object employs open circuit terminated stub lines. Also, all the internal transmission lines are preferably implemented using lines of 50 Q. The switchable power combiner receives a plurality of input signals through a plurality of input ports and combining at least some of the input signals to output a combined signal to an output port. A first through A third transmission lines selectively couples a first through a third input ports, respectively, to a combining node. A fourth

transmission line couples the combining node to a first intermediate node, and a fifth transmission line couples the first intermediate node to the output port. A first stub line with open circuit termination is selectively coupled to the output port, and a second stub line with open circuit termination is selectively coupled to the first intermediate node. The first stub line is coupled to the output port only when two of the input ports are coupled to the combining node. The second stub line is coupled to the first intermediate node only when all the input ports are coupled to the combining node.

Preferably, the first stub line includes a sixth transmission line for selectively coupling the output port to a second intermediate node, and a seventh transmission line having an end coupled to the second intermediate node and another end open-circuited.

Also, it is preferable that the second stub line includes an eighth transmission line for selectively coupling the first intermediate node to a third intermediate node, and a ninth transmission line having an end coupled to the third intermediate node and another end open-circuited.

Preferably, a length of the fourth transmission line (D3) and a combined length of the fourth and the fifth transmission line (D2) are determined by an equation: where M= 2, 3... (3) (where RL and XL respectively denote resistive and reactive components of a combined characteristic impedance of the first through the third transmission lines). Also, lengths of the first and the second stub lines (L2, L3) are determined by another equation:

RL t <BR> <BR> <BR> LM (deg) = tan- (-), BS = B, B =... (4)<BR> <BR> Y0 Z0[R2L+(XL+Z0t)2] (where Bs denotes stub susceptance).

In a more generalized aspect of the switchable power combiner of the present invention, a first through an N-th transmission lines selectively couples a first through an N-th input ports, respectively, to a combining node. An (N+1)-th through a (2N-1)-th transmission lines are connected in series between the combining node and an output port via an (N-2)-th through a first intermediate nodes arranged sequentially. A first stub line with open circuit termination is selectively coupled to the output port, and a second through an (N-1)-th stub lines with open circuit termination is selectively coupled to the first through the (N-2)-th intermediate nodes, respectively. Particularly, the first stub line is coupled to the output port only when two of the first through the N-th input ports are coupled to the combining node. Also, the j-th stub line (j=2,..., N-1) is coupled to the (j- 1)-th intermediate node only when j+1 ports of the first through the N-th input ports are coupled to the combining node.

Brief Description of the Drawings The above objectives and advantages of the present invention will become more apparent by describing in detail preferred embodiments thereof with reference to the attached drawings in which: FIG. 1 illustrates an example of a conventional 3-way switchable power combiner;

FIG. 2 illustrates another example of 3-way switchable power combiner disclosed in a pending International Application of the present applicants; FIGS. 3A and 3B are Smith charts for explaining the operation of the switchable power combiner of FIG. 2; FIG. 4 is a block diagram of an embodiment of a 3-way switchable power combiner of the present invention ; FIG. 5 illustrates an implementation of the switchable power combiner of FIG. 4 ; FIG. 6 illustrates an implementation of one of bars shown in FIG. 5; FIG. 7 is a perspective view of the switchable power combiner of FIG. 4; FIG. 8A is a Smith chart for explaining the operation of the switchable power combiner of FIG. 4 in 2-way combining mode; FIG. 8B is a Smith chart for explaining the operation of the switchable power combiner of FIG. 4 in 3-way combining mode; FIG. 9 illustrates a general open circuit terminated stub lines; and FIG. 10 illustrates an N-way switchable power combiner of the present invention.

Embodiments A 3-way switchable power combiner according to the embodiment of FIG. 4 includes three input ports PI,, PI2, and PI3, a first through a third switching unit 50,54, and 58 for selectively coupling the input ports PIl, PI2, and PI3, respectively, to a combining node 62, a connection line 64 connecting the combining node 62 to an output port PO, a first open circuit terminated stub (hereinbelow, referred to as"open stub") 66 selectively

coupled to the output port PO through a fourth switch S22, and a second open stub 68 selectively coupled to a position on the connection line 64 through a fifth switch S23.

The first switching unit 50, having a characteristic impedance of Zo, includes a switch S 11 and selectively provides the RF signal received from a power amplifier (not shown) through the input port PIl to the combining node 62. The second switching unit 54, having a characteristic impedance of Z0, includes a switch S12 and selectively provides the RF signal received from another power amplifier through the input port PI2 to the combining node 62. The third switching unit 58, having a characteristic impedance of Z0, includes a switch S13 and selectively provides the RF signal received from the other power amplifier through the input port PI3 to the combining node 62. Some or all of the Input signals supplied through the input ports PIl, PI2, and PI3 are combined on the combining node 62 depending on the switching positions of the switches S 11, S 12, and S 13, and the combined signal is provided to the output port PO through the connection line 64.

FIG. 5 illustrates an implementation of the switchable power combiner of FIG. 4.

In the implementation of FIG. 5, the first switching unit 50 is implemented by a switching bar B 1 selectively connecting the input port PIl to the combining node 62. The second switching unit 54 is implemented by a switching bar B2 selectively connecting the input port PI2 to the combining node 62. The third switching unit 50 is implemented by a switching bar B3 selectively connecting the input port PI3 to the combining node 62.

The connection line 64 may be implemented using a switching bar or a fixed bar.

Further, the connection lien 64 is preferably divided into more than one bar so as to provide a connection point of the second open stub 68. In the example of FIG. 5, the

connection line 64 is divided into two bars: a fourth bar B4 connecting a combining node port 62 to an intermediate node port 65 and a fifth bar B5 connecting the intermediate node port 65 to the output port PO.

In FIG. 5, the first open stub 66 is comprised of a sixth bar B6 connecting the output port PO to a second intermediate node port 67, and a seventh bar B7 having an end connected to the second intermediate node port 67 and another end open terminated. The sixth bar B6 connects the output port PO to the second intermediate node port 67 only when two of the input ports PI,, PI2, and PI3 are coupled to the combining node 62. The second open stub 68 is comprised of an eighth bar B8 connecting the first intermediate node port 65 to a third intermediate node port 69, and a ninth bar B9 having an end connected to the third intermediate node port 69 and another end open terminated. The eighth bar B8 connects the first intermediate node port 65 to the third intermediate node port 69 only when all the input ports PIl, PI2, and PI3 are coupled to the combining node 62.

FIG. 6 illustrates an implementation of one of the bars B1 through B9 shown in FIG. 5. In a preferred embodiment, all the bars B1 through B9 shown in FIG. 5 have the configuration similar to one another. Specifically, each of the bars B1 through B9 is implemented by a connecting segment 128 in the switch of FIG. 6. That is, the preferred embodiment of the switchable power combiner of FIG. 4 includes nine electromechanical switches. The connecting segment 128 of each switch has the characteristic impedance of 50 ohm. Alternatively, however, some of the bars (e. g., B4, B5, B8, and/or B9) may be implemented by fixed bars rather than switching bars.

A first and second connector 100 and 102, which correspond to input and output terminals of each switch, respectively, are fixed at a bottom plate of a combiner housing 104 and selectively connected with each other by a driving mechanism 110 in response to control signals applied through control lines 106 and 108. The signals applied through control lines 106 and 108 are DC signals having the same magnitude but opposite polarity.

The driving mechanism 110 includes a first and second electromagnet 116 and 118, a rotating segment 120, and the connecting segment 128. The first and second electromagnet 116 and 118 are fixed under the bottom surface of an upper plate 112 and arranged symmetrically with respect to the horizontal position where the first connector 100 is located. On a lower panel 114 is installed a supporting member 113. The rotating segment 120 is made of magnetic material and connected to the upper portion of the supporting member 113 by a pin 126 so as to be rotated within a certain angle range. The rotating segment 120 has a shape extending from its horizontal center connected to the supporting member 113 to the position under the first and second electromagnet 116 and 118, and has a plate spring 123 thereunder. The connecting segment 128 includes a vertical portion 129 and a connecting portion 134 extending laterally from the bottom end of the vertical portion 129 to the location of the first and second connectors 100 and 102.

The connecting segment 128 is installed by inserting the vertical portion 129 through a not- shown hole formed in the lower plate 114, mounting a spring 130 to the upper side of the vertical portion 129 and putting a cap 132 on the vertical portion 129.

The switch of FIG. 6 operates as follows. When non-zero control signals are applied through the first and second control lines 106 and 108, currents having opposite

directions flows through the first and second electromagnets 116 and 118. At this time, the first electromagnet 116 draws a first end 121 of the rotating segment 120 and the second electromagnet 118 retracts a second end 122 of the rotating segment 120.

Accordingly, the plate spring 123 of the rotating segment 120 presses the connecting segment 128 downward, and the first and second connectors 100 and 102 are electrically connected with each other by the connecting portion 134 of the connecting segment 129.

On the other hand, when control signals of zero magnitudes are applied through the first and second control lines 106 and 108, the connecting portion 134 moves upward by the restoring force of the spring 130 and the first and second connectors 100 and 102 are electrically disconnected. Since the switch is configured electromehanically as above, the switchable power combiner is prevented from being damaged in an application handling large current.

Referring back to FIGS. 4 and 5, the length (D2) of the connection line 64, that is the sum of lengths of bars B4 and B5, is determined by the equation: where M=2, 3... (3)

, where RL and XL respectively denote resistive and reactive components of a combined characteristic impedance of the first through the third switching bars. Also, the length (D3) of the fourth bar B4 is determined by the equation 3. On the other hand, the length (L2) of the first open stub 66, that is the sum of lengths of bars B6 and B7, is determined by the equation:

BS RL2t(Z0-XLt)(XL+Z0t)<BR> <BR> <BR> <BR> LM (deg) = tan-1( ), BS = B, B= ...(4)<BR> <BR> <BR> <BR> Y0 Z0[RL2=(XL=Z0t)2] , where B denotes a susceptance and Bs denotes stub susceptance. Also, the length (L3) of the second open stub 68, that is the sum of lengths of bars B8 and B9, is determined by the equation 4.

FIG. 7 is a perspective view of the switchable power combiner of FIG. 4. As shown in the figure, the switchable power combiner roughly has a hexahedral shape. On the exterior surface of the housing are installed coaxial connectors corresponding to the input and output ports. The switching bars having characteristic impedances of 50 ohms inside the housing are shown by dotted lines. A connector 200 for receiving switching control signals from an external controller is installed beneath the lower exterior surface of the housing. Direct current (DC) control signals are provided through the connector 200 and the switches are turned on or off in response to the control signals. In an alternative embodiment, however, the controller may be embedded into the switchable power combiner.

The operation of the switchable power combiner of FIGS. 4 through 7 will now be described in detail.

Each of the input ports PII, PI2, and PI3 receives respective input signals to provide such signal to the switching units 50,54, and 58. Depending on the condition of the first through the third switches S 11, S 12, and S 13, some or all of the input signals received through the input ports PI1, PI2, and PI3 are combined at the combining node 62. The

combined signal is provided to the output port PO through the connection lien 64.

Meanwhile, the fourth and fifth switches S22 and S23 are turned on or off depending on the number of signals to be combined.

First, we consider 1-way operation mode. Considering, for example, the case that only the signal received through the input port PI3 is to be output through the output port PO, the first and the second switches S 11 and S 12 are turned off but the third switch S 13 is turned on. At this time, the fourth and the fifth switches S22 and S23 are turned off, so that the first and the second switches are isolated from the connection line 64. Thus, the signal received through the input port PI3 is provided to the output port PO through the combining node 62 and the connection line 64, that is, through the bars B4 and B5. Since all the transmission lines B3, B4, and B5 along the signal transmission path have characteristic impedances of 50 ohms, the impedances are matched during the signal transmission process.

As an example of a 2-way combining mode, we consider the case of combining the signals received through the input ports PIl and PI2. The first and the second switches S 11 and S 12 are turned on but the third switch S 13 is turned off. Thus, the signals passing through the switching bars B1 and B2 are combined at the combining node 62, and the combined signal is provided to the output port PO through the bars B4 and B5.

Meanwhile, the fourth switch S22 is turned on while the fifth switch S23 is turned off, so that the first open stub 66 is connected to the output port PO.

At this time, the length (D) of the connection line 64 and the length (L2) of the first open stub 66, which are determined by the equations 3 and 4, respectively, play important

roles in the impedance matching. Inserting Zo = 50 ohm, XL = 0, and RL = 25 (50/2) ohm in the equation 3, we get D2= 34.89 degrees. The actual physical length is determined by an equation 5 and dependent on the dielectric coefficient of the dielectric material involved with the transmission line. When air is used for the dielectric material, the actual length is approximately 32.96 millimeters (mm) in the frequency band of 800 Megahertz (MHz).

Meanwhile, inserting Zo = 50 ohm, XL = 0, and RL = 25 ohm in the equation 4, we get L2 = 34.89 degrees. If the relative dielectric coefficient is 1, the actual length is approximately 32. 96 mm from the equation 5.

Referring to FIG. 8A, the impedance viewed from the combining node 62 toward the side of input ports, 25 ohm, corresponds to point PI in the Smith chart of FIG. 8A.

The point PI is transformed to a point P2 by the connection line 32 having the characteristic impedance of Zo and the length of D2. Further, the point P2 is transformed to a point P3 by the first open stub 66 having the characteristic impedance of Zo and the length of L2, so that the internal characteristic impedance of the combiner is exactly 50 Q and the impedance matching is fulfilled.

In a 3-way combining mode, all of the first through the third switches S 11, S 12, S 13 are turned on, so that all the input signals received through the input ports PIl, PI2, and PI3 are combined at the combining node 62 and provided through the bars B4 and B5.

The fourth switch S22 is turned off while the fifth switch S23 is turned on, so that the second open stub 68 is connected to the first intermediate node port 65.

At this time, the length (D3) of the bar B4 and the length (L3) of the second open stub 68, which are determined by the equations 3 and 4, respectively, play important roles in the impedance matching. Inserting Zo = 50 ohm, XL = 0, and RL = 16.7 (50/3) ohm in the equation 3, we get D3 = 29.84 degrees. The actual physical length is determined by the equation 5 and dependent on the dielectric coefficient of the dielectric material involved with the transmission line. When air is used for the dielectric material, the actual length is approximately 27.7 mm in the frequency band of 800 MHz. Meanwhile, inserting Zo = 50 ohm, XL = 0, and RL = 16.7 ohm in the equation 4, we get L3 = 46.94 degrees. If the relative dielectric coefficient is 1, the actual length is approximately 44.35 mm from the equation 5.

Referring to FIG. 8B, the impedance viewed from the combining node 62 toward the side of input ports, 16.7 ohm, corresponds to point PI in the Smith chart of FIG. 8B.

The point PI is transformed to a point P2 by the bar B4 having the characteristic impedance of Zo and the length of D3. Further, the point P2 is transformed to a point P3 by the second open stub 68 having the characteristic impedance of Zo and the length of L3, so that the internal characteristic impedance of the combiner is exactly 50 Q and the impedance matching is fulfilled.

As described above, according to the present invention, the impedance matching is performed using open stubs. FIG. 9 illustrates a general open stub. As can be seen in FIG. 9, an open stub can be inserted into a transmission line having a characteristic

impedance of Zo, or 50 ohm, and having an end connected to a load of ZL. The present invention carries out the impedance matching by adaptively changing the distance (DN) from the combining node to the point at which the open stub is inserted along with the length (LN) of the open stub. Values of DN and LN can be obtained by inserting parameters (RI, XI, Zo, t, and B) in the equations 3 and 4. The appropriate values for the parameters may be summarized as follows.

<Table 1> Parameters according to the number of the combined paths

Parameters 2-way 3-way 4-way N-way RL 25. 000 16. 667 12. 500 Zo/N XL 0. 000 0. 000 0. 000 0.000 Zo 50.000 50. 000 50. 000 50.000 t 0707 0577 0500 ' RL-ZO RL t (ZOXLt) (XL +Zot) B 0.014 0.023 0.030------------ ZO [RL + (XL +Zot)] Values of DN and LN calculated using the above values of the parameters are as follows: <Table 2> Values of DN and the LN (deg.) 2-way 3-way 4-way N-way DN 34.885 29.836 26.477 tan-lt B LN 34.885 46.943 51.861 tan-- Po

When we apply the values of DN and LN expressed in the unit of degrees to center frequency of a cellular phone system, 881. 51 MHz, a center frequency of a personal communications system (PCS), 1855 MHz, and a center frequency of an IMT-2000 system, 2140MHz, we obtain the actual lengths, IDN and IN, as follows: <Table 3> Actual lengths of transmission line and open stub

2-way 3-way 4-way N-way 32.979 28.206 25.031 0. 945DN 885.5 MHz 32. 979 44. 378 49. 027 0.945LN 15.672 13. 403 11.985 0.449DN 1855 MHz 15.672 21.088 23.298 0.449LN 13.585 11.618 10.310 0.389DN 2140 MHz 13.585 18.280 20.195 0.389LN

Even though the present invention has been described in the viewpoint of 3-way switchable power combiner above, the present invention may be generalized to an N-way switchable power combiner. FIG. 10 illustrates an N-way switchable power combiner of the present invention generalized from the 3-way power combiner. Some or all of the input signals received through input ports PI1 through PIN are combined at a combining node G1 and provided to an output port PO. Depending on the number of signals to be combined, one of the open stubs may be coupled to an connection line. In case of a 3-way combining mode, for example, just a single open stub spaced by D3 from the combining node G1 and having a length of L3 is coupled to the connection line but the other open stubs are isolated from the connection line. The lengths of the transmission lines, D and L ;, are determined by the equations 3 and 4.

Although the present invention has been described in detail above, it should be understood that the foregoing description is illustrative and not restrictive. For example, short circuit terminated stub lines may be used as well instead of the open circuit terminated stub lines. Those of ordinary skill in the art will appreciate that many obvious modifications can be made to the invention without departing from its spirit or essential characteristics. Thus, it should be apparent that the invention can be modified in arrangement and detail without departing from such principles. We claim all modifications and variation coming within the spirit and scope of the following claims.

Industrial Applicability

As described above, the switchable power combiner according to the present invention is implemented using stub lines while all the transmission lines have characteristic impedances of 50 ohms. Thus, the number of components are reduced and the insertion loss of the combiner is decreased. Also, the combiner may be easily implemented and shows superior impedance matching performance.