GB2177863A - Impedance preserving amplifier - Google Patents

Abstract

An amplifier has the characteristics that the impedance presented by the amplifier at its input is proportional to the impedance seen by its output and that the impedance presented by the amplifier at its output is proportional in inverse proportion to the impedance seen at its input. Thus from an impedance point of view the amplifier is transparent when inserted into a circuit such as a filter circuit as used in radio receivers. The amplifier can be implemented by two cascaded stages, each stage comprising a transconductor 32 connected in a closed loop with a second similar transconductor. <IMAGE>

Description

SPECIFICATION Impedance Preserving Amplifier The present invention relates to an impedance preserving amplifier and to a ladder filter and a direct conversion receiver including such an amplifier.

In many filter circuits it is desired to provide some amplification. However if amplification precedes the filtering then one obtains an inferior intermoduiation performance. While if amplification follows the filtering then the noise of the filter degrades the overall noise performance. Ideally the required gain must be carefully distributed so that the front end noise dominates that of the later stages, while large unwanted signals must remain undistorted until they can be removed by a filter.

However if one attempted to implement the gain stage within the filter using an operational amplifier, the characteristics of infinite input impedance and low output impedance would upset the filter.

It is an object of the present invention to provide an amplification stage which can be inserted into a filter circuit without upsetting the filter performance.

According to one aspect of the present invention there is provided an amplifier in which the impedance presented by the amplifier at its input is proportional to the impedance seen by its output and in which the impedance presented by the amplifier at its output is proportional in inverse proportion to the impedance seen atits input.

By "proportional in inverse proportion" is meant that the constant of proportionality is the inverse of that in the forward direction. However in an embodiment in which scaling of the impedance is not required then the constant of proportionality can be set to unity.

Consequently from an impedance point of view such an amplifier is transparent and hence it can be inserted into a filter circuit without upsetting its performance. In fact it will enhance the filter performance because it can provide distributed gain in the filter which is regarded as desirable.

The amplifier may be implemented using transconductors, more particularly two pairs of transconductors, the transconductors of each pair being connected back-to-back.

The present invention also relates to a filter having a plurality of reactive stages connected in series and at least one amplifier made in accordance with the present invention connected between two successive stages.

If the input impedance is proportional to the output impedance and the output impedance is proportional to the input impedance with the inverse constant of proportionality, then the impedance level of one part of the filter can be made different from another part; this flexibility can be exploited to allow the use of preferred component values, or in an integrated circuit implementation to minimise the silicon area.

The present invention further relates to a receiver including at least one filter made in accordance with the invention.

The present invention will now be described, by way of example, with reference to the accompanying drawings, wherein: Figure 1 is a block schematic diagram of a basic, direct conversion receiver including impedance preserving amplifiers made in accordance with the present invention, various features having been omitted in the interests of clarity, Figure 2 is a diagram representative of an impedance preserving amplifier, Figure 3 illustrates an implementation of one pair of back-to-back connected transconductors.

Figure 4 is a diagram of an equivalent circuit of two pairs of back-to-back connected transconductors.

In the drawings corresponding reference numerais have been used to indicate the same features.

The direct conversion receiver comprises an antenna 10 which is connected to respective inputs of first and second mixers 12,14. Signals from the antenna 10 are mixed down to base band by a local oscillator 16 having the same frequency as the carrier frequency of the applied signal. The local oscillator signal applied to the mixer 14 is shifted in phase by tut/2 in a phase shifter 18. The quadrature related outputs of the mixers 12, 14 are applied to respective filter-amplifiers 20, 22. Outputs from the filter amplifiers 20, 22 are applied as respective inputs to a demodulator 24 which produces the modulating signal as an output. The operation of the direct conversion receiver is known, see for example British Patent Specification 2094079A (PHB32760), and accordingly a detailed description will not be given.

The filter-amplifiers 20, 22 are substantially identical. Each comprises an L.C. ladder filter in which the ladder is interrupted by impedance preserving amplifiers 28,30. Each amplifier 28, 30 is transparent from an impedance point of view because the impedance presented by the amplifier at its input is equal (or proportional if impedance scaling is desired) to the impedance seen at its output, and vice versa. Therefore it is able to amplify the signal without upsetting the operation of the filter. Consequently signals passing through the amplifier in the forward direction are amplified, and because of the bidirectional nature of the amplifier, signals can also pass in the reverse direction, attenuated by the gain of the amplifier.

Figure 2 is a diagram of an impedance preserving amplifier made in accordance with the present invention. Expressed mathematically the amplifier obeys the condition (Vout/lout)=N2 (Vin/lin) with Vout=Vin K N.

and lout=lin K N.

where N is the scaling of the impedance level between output and input (if required) and K is the gain of the amplifier. The power gain of the amplifier is K2, both the voltage and the current being amplified. An amplifier obeying these conditions can be constructed by rearranging the equations to give Vout=Vin K N.

lin=lout N/K.

In otherwordsthe inputvoltage is measured, multiplied by "K N." and developed atthe output, while the current lout which is drawn at the output is measured, then a fraction N/K is taken to give the current drawn at the input.

The amplifier circuit can be implemented using two pairs of back-to-back connected transconductors of which Figure 3 shows one pair. If one considers a transconductcr, such as that shown in the block 32 in Figure 3 it comprises a long tailed pair of bipolartransistors 34,36. The commonly connected emitters of these transistors are connected to a constant current source 38 having a current output. Constant current sources 40, 42 are connected in the collector circuits of the transistors 34,36. The current output of each of the sources 40,42 is J1/2.In operation the sum of the currents through the transistors 34,36 is constant and therefore an increase in voltage across the inputs, that is the base electrodes of the transistors 34,36 causes current to be drawn in through one output, for example the collector circuit of the transistor 36, and expelled through the other one, for example the collector circuit of the transistor 34.

A decrease in the voltage across the inputs produces a reciprocal effect and also a voltage inversion across the inputs inverts the sign of the current flow out of the transconductor. The operation of the transconductor in the block 32 can be summarised as producing a differential current output proportional to the differential input voltage: 11=V1 'gi.

In connecting two transconductors back-to-back the collector (or output) circuits of one pair of transistors are connected respectively to the base (or input) circuits of the other pair of transistors and vice versa. In operation the current drawn at the input is related to the output voltage l2=V2 g2 where g1, and 92 are the respective transconductances of the back-to-back connected transconductors. This arrangement (with the currents J1 =J2) is used to form a gyrator commonly used in filter circuits. Note however, that in this case the currents are deliberately different so as to produce differing forward and reverse transconductances gel, 92.

Figure 4 illustrates diagrammatically the operation of an impedance preserving amplifier formed by two pairs 44, 46 of back-to-back connected transconductors each represented by a voltmeter controlling a current source. In the operation of the amplifier if the input voltage V1 increases then the corresponding increase in the current 11 increases the voltage across the input of the pair 46 that is V2=V3. This in turn causes 13 to increase thus developing a greater voltage V4 across the load impedance (not shown). The voltage V4 developed by 13 across the load impedance causes a corresponding increase in the current 14 flowing through its respective transconductor. This process continues until the current 14 balances the current 11. This in turn determines the voltage V2 which sets the current in 12.

Thus the increase in the output voltage V4 due to the current flow in the load impedance has been reflected as an increase in the current drawn at the input 12. This gives the required property that an increase in the applied voltage (V1) caused the current drawn (12) to increase, in accordance with load impedance at the amplifier output port.

The effect of the connection of the two pairs of back-to-back connected transconductors can be illustrated mathematically as follows: From the pair 44 one can express 11=V1 g1 V2=11/g2 from the pair 46 one can express V4=11/g4 (11=14 at equilibrium) 13=V2 g3 (V2=V3) by substitution it can be shown that V4=V1 (91/94) 13=12 (g3/g2) that is, the output voltage is proportional to the input voltage, and the current drawn at the input is proportional to that at the output, as required.

The power gain K2 of the amplifier is given by

Vout lout K2= Vin x iin =(g1/92) (g3/g4) and the impedance scaling N2 is given by

Vout Vin N2 t =(ski g2)/(g3 g4).

I out lin If no impedance scaling is required, then the two pairs of transconductors will be identical so that g1 =93 and g2=94 which means that K=g1/g2 and N=1.

When impedance preserving amplifiers made in accordance with the present invention are used within filters, they make only the noise contributions of the first stages of the filter significant, the signal being amplified to make the contribution of the later stages (such as the termination resistor) negligible.

In a radio receiver such as is shown in Figure 1 the provision of these amplifiers 28,30 results in an improvement of the overall noise figure, and thus a better receiver sensitivity. Furthermore, the noise introduced by the amplifiers themselves is filtered by the following sections of filtering, preventing the degradation of receiver sensitivity by the presence of wide-band amplifier noise.

If the input impedance is proportional to the output impedance and the output impedance is proportional to the input impedance with the inverse constant of proportionality, then the impedance level of one part of the filter can be made different from another part; this flexibility can be exploited to allow the use of preferred component values, or in an integrated circuit implementation of the filter or receiver to minimise the silicon area.

Claims (9)

1.An amplifier in which the impedance presented by the amplifier at its input is proportional to the impedance seen by its output and in which the impedance presented by the amplifier at its output is proportional in inverse proportion to the impedance seen at its input.

2. An amplifier as claimed in Claim 1, in which the constants of proportionality in the two directions are reciprocal of each other.

3. An amplifier as claimed in Claim 1, in which the constants of proportionality are set to unity.

4. An amplifier as claimed in any one of Claims 1 to 3, comprising two pairs of back-to-back connected transconductors.

5. An amplifier constructed and arranged to operate substantially as hereinbefore described with reference to and as shown in the accompanying drawings.

6. An amplifier as claimed in any one of Claims 1 to 5, arranged to attenuate the signal.

7. A filter comprising a plurality of reactive stages coupled in series and at least one amplifier as claimed in any one of Claims 1 to 6 connected between two successive stages.

8. A filter as claimed in Claim 7, in which there is scaling ofthe impedance level between the input and output ports of the amplifier or at least one of the amplifiers.

9. A receiver comprising at least one amplifier as claimed in any one of Claims 1 to 6.