The present invention relates to a reconfigurable broadband active power splitter, a ditto power combiner and a ditto bidirectional power splitter/power combiner and circuits based on these. Use of such circuits in an analog distribution network enables good insulation between different channels in combination with amplification in the distribution network. The latter reduces the requirements for subsequent amplification, which facilitates dynamics and linearity properties. It is important that the invention is ultra broadband, and that, by using MMIC technology, the circuit surface size can become small.
The invention is manifested in its basic form in a reconfigurable broadband active power splitter, a corresponding power combiner and a corresponding bidirectional power splitter/power combiner. These basic circuits can be combined in gradually more complex circuits, with several inputs and outputs.
The starting point of the invention is a reconfigurable beam-forming network in a military application. Examples of this will be presented below. The basic circuits included, as well as the gradually more complex distribution networks that will be presented in this patent specification, can, however, be used in many other contexts where there is a need for power splitter and combiner. This is obvious to a person skilled in the art. It is therefore the applicant's explicit intention that the patent protection for the stated circuits having the stated properties should relate to their construction independently of where they are being used. In the now presented most complicated composition of subcircuits, also a completely reconfigurable beam-forming network is included, which will be presented in a claim.
Future military platforms will be provided with a combined system for radar, communication and electronic warfare (EW). It will be of interest to provide the system with a common set of beam-forming network, transceiver module and antenna (Single RF System). An antenna which is convenient in the context is a broadband multifunctional phased array antenna. Known analog beam-forming networks cannot be reconfigured in a simple manner so that the desired multifunctionality is achieved. It is also desirable, with an analog beam-forming network, to have a possibility of performing also a digital beam-forming.
In the invention, this is achieved by reconfigurably beam-forming networks having the design that is apparent from one of the independent claims. At the same the network is composed of gradually less complex distribution networks, down to a reconfigurable broadband active power splitter, a ditto power combiner and a ditto bidirectional power splitter/power combiner. These distribution networks solve, at their different levels, the current problems by having the design that is evident from the respective appended claims.
The invention will be described in more detail below with reference to the accompanying drawings, in which
a shows the inner capacitances between the connections of the transistors in
b shows an artificial transmission line for the output of the transistors in
c shows an artificial transmission line for the input of the transistors in
a shows the inner capacitances between the connections of the transistors in
b shows an artificial transmission line for the input of the transistors in
c shows an artificial line transmission line for the output of the transistors in
a shows how the power splitter in
b shows how output port 1 in the power splitter in
c shows how output port 2 in the power splitter in
d shows how both output port 1 and output port 2 are shunted in the power splitter in
First the basic circuits will be discussed. They are constructed using distributed technique. The great advantage of this is that it gives the possibility of obtaining broadband function.
The transmission lines consist of series inductances Ld and Lg as well as parallel capacitances in the form of the inner capacitances Cds and Cgs of the transistors, see
For the signal from an amplifying step to be added constructively to the next, the phase response between the steps must be the same. Thus the broadband behaviour is achieved by the group delay for the gate line and the two drain lines being made identical. The bandwidth is limited upwards by the cut-off frequency of the lines and by the possibility of terminating the lines efficiently at all frequencies. The latter is due to the fact that the transmission lines have a frequency-dependent characteristic impedance that is to be terminated. The bandwidth is limited downwards by difficulties in adjusting the high resistances at the gate to a reasonable consumption of surface.
The number of amplifying steps further determines the total amplification that can be obtained from the input to the respective output. The number of steps is in practice limited since parasitic resistances of the transistors contribute to the attenuation in the lines, which will finally exceed the increase in amplification that is obtained by means of an extra step. Usually, the maximum is obtained at 6-7 steps.
The cut-off frequency fc is given by
for gate line, see
for drain line, fee
The gate line constitutes an artificial LC connection. In its simplest form, it is a low pass filter with a constant parameter
Similarly, the drain line constitutes an artificial LC connection. In its simplest form, it is a low pass filter with a constant parameter
To achieve broadband properties, the gate line and the drain line must have identical characteristic impedances and identical phase velocities, which gives the following equations
Kg=Kd (1)
and
LgCgs=LdCds (2).
In practice, Cgs is greater than Cds, and two methods can be used to make the phase velocity identical in gate line and drain line:
1. When it is desirable for Lg and Ld to be identical, an extra shunt capacitance Coff can be connected parallel with Cds so that the total capacitance equals Cgs and the equations (1) and (2) will be satisfied. This is shown in FIG. 3.
2. If it is not desirable to use a shunt capacitance, a series inductance αLd is connected to the drain of the transistor and Ld is selected so that the same phase velocity is obtained in the gate line and the drain line.
The new and actual basis for the present invention is that the shunt capacitance in method 1 is replaced with a switching element which is optimised, depending on a control voltage, to take two distinct states, a first corresponding to a capacitance Coff and a second state corresponding to a low impedance short circuit Ron, see FIG. 4. The switching elements can advantageously be field effect transistors.
When the control voltage to the switching elements for a distributed amplifier is high, the amplifier transmits power, but when the control voltage is low, it does not transmit power. As a result, a reconfigurable active power splitter is obtained, which is illustrated in FIG. 5 and is in this patent application called FC1A.
This circuit thus functions not only as a constant power splitter. It can connect the power to one or the other output port onto both, without deteriorating the matching conditions. The switching elements STi, i=1, . . . , N, are identical and controlled by a common control voltage VC, which is very favourable and desirable. The same applies to the lower switching elements ST′i, i=1, . . . , N, in the Figure and their control voltage VC′.
As stated above, a switch transistor corresponds in its high-impedance state, VC=VHigh, to a capacitance Coff. In its low-impedance state, VC=VLow, a switch transistor corresponds to a low resistance Ron. Depending on the switch state of the switch transistors STi nad ST′i, four states are thus possible for the power splitter. This is illustrated in
It should be noted that in FC1A the shunted switch transistors are part of the network itself. In known previous attempts to control the power, a separate switching network has been used in each output port, which is a more complicated and quite inferior method. FC1A is suitably made in MMIC technology, although this is not necessary.
A reconfigurable active power combiner FC1B, of a composition, which is closely, related to that of the above-described power splitter, is also part of the invention and is shown in FIG. 7. Based on the circuit solution in FIG. 7 and the discussion about the power splitter FC1A, it is obvious to those skilled in the art how the circuit is made up.
A third circuit according to the invention is a reconfigurable active bidirectional power splitter/power combiner FC1C. It is a combination of FC1A and FC1B in one and the same circuit and is shown in FIG. 8. The transistors Ti, Ti comp, Ti′, Ti comp′, are active switches. By an active switch is meant a component having two distinct states. A first corresponding to a capacitance network and a second state corresponding to an amplification. The amplification is obtained by a suitable voltage being applied between the drain and the source as well as between the gate and the source.
The transistors Ti, Ti comp are complementary, so that when one takes the first state, the other takes the second state and vice versa. Thus, when the transistors Ti comp and Ti comp′, i=1, . . . , N, are in a high-impedance state, corresponding to a capacitance network, Ti and Ti′, i=1, . . . , N, are in an amplifying state, and then the entire circuit corresponds to a power splitter. On the other hand, when the transistors Ti and Ti′, i=1, . . . , N, are in a high-impedance state and Ti comp and Ti comp′, i=1, . . . , N, are in an amplifying state, the entire circuit corresponds to a power combiner.
A number of power splitters FC1A can be interconnected in a power splitting network ASN in order to gradually achieve an increasingly split signal. This is illustrated in FIG. 9. In connection with
Similarly, a number of combiners FC1B can be interconnected in a power combining network ACN in order to gradually combine the signal to an output. This is illustrated in FIG. 10. Also in this case, the designations of the inputs indicate connection points for these beam formers.
Furthermore, special combinations of FC1A and FC1B can be formed to be useful in various contexts. A 6-port distribution network FC2A is shown in FIG. 12. It is made up of one FC1A and one oppositely directed FC1B. With four control voltages VC1A, VC1B, VC2A and VC2B, power can be connected in 16 different ways. Compared with FC1C, this solution gives a bidirectional function at the same time.
A 7-port distribution network FC3A is shown in FIG. 13. It is made up of one FC1A in one direction and one FC1B followed by one FC1A in the other direction. It uses six control voltages VC1A, VC1B, VC1C, VC2A, VC2B and VC2C to direct power to different ports. The solution indicates a further possibility of generally distributing, parallel in time, power between ports.
A 10-port distribution network MFAM is shown in FIG. 14. It is a more complicated variant of previous networks and is made up of three FC1A and three FC1B.
At the end of this patent specification, it is time to describe the application that constitutes the base for the present development, namely a multifunctional broadband electronically steered phased array system. Such a system with 8×8 antenna elements is shown in FIG. 15. In general the system can be assumed to have N×M antenna elements. It can, as stated by way of introduction, be connected to signal generation/reception equipment which is not shown in the Figure but is assumed to be located “to the left of” the same. The signal generation/reception equipment performs, simultaneously or in rapid succession, traditional radar functions, multilobe direction finding, exchange of data messages and other communication, transmission of jamming signals etc.
The system comprises M identical beam formers BF in a modular mode) to send broadband signals from the signal generation/reception equipment to the phased array antenna elements AE and vice versa. The beams are marked L in the Figure and their angles are marked θ1 and θ2. The beam formers consist, seen from the antenna elements, of one transceiver module T/R per antenna element. Two such T/R modules are connected to a circuit consisting of one MFAM and two up and down converters DUC. This circuit is shown in FIG. 16. AE1 and AE2 relate to antenna element 1 and 2. Signals to the antenna elements are called TX1 and TX2, and signals from the same are called RX1 and RX2. On the other side of MFAM there are ports on the one hand for digital signals TX1-D and RX1-D to and respectively from TX1 and RX1 and the equivalent for TX2 and RX2 and, on the other hand, ports for analog signals Tx-A and RX-A to and respectively from the antenna elements involved. The signal after the up and down converters DUC is called IF1 and IF2.
A number of different blocks for different functions, such as digital beam-forming and chanalisation, can be connected to the beam former BF which gives conventional analog beam forming. Since it is easy to direct signals to the two outputs in the basic circuits and, thus, also to and from the different inputs and outputs of the compound circuits which are made up of these basic circuits, it is possible to use the aperture of the entire antenna system for a function or, during certain periods of time, share the aperture among different functions.
The chanalisation network CN, which can be connected to the beam former, is made up of a combination of digital and analog circuits and is used, for instance, in the signal intelligence function to provide direction finding and position finding.
A power splitting network ASN and a power combining network ACN and/or a bidirectional network ASCN can also be connected to the beam-forming network. These networks are used to split and combine the signal from the entire antenna aperture or parts thereof.
Digital beam formers DBF can also be connected to the previously mentioned conventional (analog) beam-forming network. These digital beam formers are used, for instance, when side lobes need to be suppressed adaptively. The signal from each individual antenna element has its own channel and the digital beam former has the same number of inputs/outputs as the number of antenna elements.
In the conventional beam formers BF, a circuit of the type FC2A is connected to two circuits of the above-mentioned type consisting of one MFAM and two up and down converters. In the example involving 8×8 elements in the antenna matrix, a circuit FC3A follows the circuits FC2A, see FIG. 15. With a larger number of elements in the matrix, it is possible to use a network of circuits FC2A instead of a single circuit FC2A.
FC3A are used for analog beam forming, but they can be reconfigured to carry out switching functions for the chanalisation of the matrix and port termination. Also the MFAM circuit is used in the beam former BF for analog beam forming, but can be reconfigured to carry out switching functions for both analog and digital beam forming of the matrix.
The network ASN, which according to
With the aid of the chanalisation network CN, M channels are formed, each connected to a beam former BF. With the aid of MFAM, FC2A and FC3A in the beam former, the information from all elements connected to the beam former or only information from one or a small number of selected antenna elements can be selected.
As indicated, digital beam formers DBF can be connected to the analog beam formers BF. The digital beam-forming network consists of M×N channels (in this example 8×8=64 channels) and each channel consists of circuits handling both analog and digital signals. It converts down the RF signal to the base band, AD converts the signal and processes it. By means of MFAM in the analog beam former, the information from each individual antenna element can be directed to a separate channel for digital beam forming.
Number | Date | Country | Kind |
---|---|---|---|
0004242 | Nov 2000 | SE | national |
This is a nationalization of PCT/SE01/02562 filed Nov. 20, 2001 and published in English.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/SE01/02562 | 11/20/2001 | WO | 00 | 5/20/2003 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO02/41442 | 5/23/2002 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
4973918 | Schindler | Nov 1990 | A |
5027084 | Tsukii | Jun 1991 | A |
5361038 | Allen et al. | Nov 1994 | A |
5504745 | Petz et al. | Apr 1996 | A |
Number | Date | Country |
---|---|---|
2 759 508 | Aug 1998 | FR |
Number | Date | Country | |
---|---|---|---|
20040041632 A1 | Mar 2004 | US |