US 4,149,122 AGrant
Wideband Tracking Generator for Harmonic Mixing Receivers
Issue Date:1979-04-10
•3 Claims
•6 Drawing Sheets
Abstract
A tracking generator for a wideband superheterodyne receiver in which a fixed-frequency signal source supplies an IF signal to a mixer for conversion to an RF signal at the frequency to which the receiver is tuned. A second signal is supplied to the mixer from the local oscillator within the receiver. The fixed frequency IF signal may be converted to an RF signal at every frequency that the receiver can accommodate merely by sweeping the local oscillator throughout its frequency range.
Metadata
Assignee
- Cutler-Hammer, Inc.
Inventor
- Vito J. Parato
Application Information
Application Number:US 8117098
Filing Date:1977-06-30
Priority Date:1977-06-30
Art Unit:233
Classifications
IPC:
H04B 1700
Field of Search:
32532467;363;134;455;430-432;435;438130
Patent Drawings (6 sheets)
Description
This application is related to U.S. application Ser. No. 802,173.
Background
1. Field
This invention pertains to the generation of test signals for receiving systems and, in particular, to the generation of test signals for wideband systems employing harmonic mixing.
2. Prior Art
Frequency scanning superheterodyne receivers are well known. With the development of modern microwave devices, the frequency range of such receivers has been extended to frequencies well beyond 20 GHz. The ability of these receivers to continuously sweep over wide bandwidths has also been extended, as illustrated by the multioctave tuning range of the receiving system described in U.S. Pat. No. 3,701,949.
In some cases, harmonic mixing is used to achieve a wide receiver bandwidth. In this type of mixing, the harmonics of a relatively narrow bandwidth local oscillator are applied to the receiver mixer to cover a wider RF tuning range. For example, a local oscillator with a range of 1.8 to 4.0 GHz may be swept over its range repetitively, while the receiver preselector sweeps continuously over a much wider frequency range of 2 to 12 GHz. For the receiver tuning range of 2 to 4 GHz, the local oscillator sweeps from 1.8 to 3.8 GHz to produce an IF frequency of 200 megacycles. For the receiver RF tuning range of 4 to 8 GHz, the oscillator sweeps over the range of 1.9 to 3.9 GHz, to produce a second harmonic signal which sweeps over the range of 3.8 to 7.8 GHz and again produces an IF frequency of 200 MHz. For the 8 to 12 GHz RF tuning range of the receiver, the third harmonic of the local oscillator produces a 200 MHz IF signal in a similar manner.
Tracking generators suitable for operating with such wideband receivers are generally of two types. In the first of these types, the generator accepts a portion of the sweep control voltage from the receiver, such as the signal applied to drive the preselector. This voltage is applied within the tracking generator to drive an oscillator which covers the full range of the receiver. Unfortunately, oscillators with the extremely wide bandwidths are often not commercially available. An additional problem is the sweep control voltage for the preselector may not be a continuous. A receiver using a harmonically generated local oscillator signal may have a preselector control voltage which stops and retraces a portion of the sweep cycle while the local oscillator is retraced through its complete range. In this type of system, the sweep control voltage may include a retrace portion which can result in drift that prevents the tracking oscillator from properly tracking the receiver.
In the second type of tracking oscillator system, a portion of the local oscillator signal from the receiver is supplied to the tracking generator. An offset oscillator within the generator produces a signal at the frequency to which the receiver is tuned. This type of tracking generator cannot function properly where the receiver local oscillator does not directly produce the local oscillator signal. For example, in a receiver employing harmonic mixing, the harmonics may be produced in the mixer itself rather than in the local oscillator. In this case, it would be difficult to couple a portion of the harmonic signal from the mixer for use in the tracking generator. In addition, the local oscillator harmonic signal produced in the mixer is usually weak and if it is to be applied to the tracking generator it must first be extracted from a large number of signals present in the mixer. The technical difficulties in accomplishing this generally makes the offset oscillator system unsuitable as a tracking generator for wideband receivers employing harmonic mixing.
Summary
It is an object of the present invention to reduce the cost of providing a wideband tracking oscillator. It is another object of this invention to provide a tracking oscillator capable of functioning with wideband receivers using harmonic mixing.
According to the present invention, an IF signal and a portion of the local oscillator signal from the receiver are applied to a mixer to produce a signal at the RF frequency to which the receiver is tuned. Where harmonic mixing is used within the receiver, the desired harmonic is selected from the mixer by passing the RF signal through a separate preselector similar to that contained within the receiver.
Brief Description of the Drawings
FIG. 1 is a block diagram of a portion of a basic superheterodyne receiver incorporating an IF signal source for calibration purposes.
FIG. 2 is a block diagram of a complex superheterodyne receiver incorporating an IF signal source for calibration purposes.
FIG. 3 is a block diagram of a superheterodyne receiver incorporating an IF signal source and a reflective termination at the RF port of the receiver for calibration purposes.
FIG. 4 is a block diagram of a superheterodyne receiver incorporating an IF signal source to generate a calibration and test signal at the RF port of the receiver.
FIG. 5 is a block diagram of a two channel superheterodyne receiver with a common local oscillator.
FIG. 6 is a block diagram of a portion of a two channel superheterodyne receiver with a common local oscillator and a local oscillator drive circuit.
FIG. 7 is a graph of a ramp voltage waveform for driving the local oscillator.
FIG. 8 is a graph of various voltage waveform for driving the local oscillator.
FIG. 9 is a block diagram of a time shared, self-calibrating receiving system.
FIG. 10 is a block diagram of a continuous, self-calibrating receiving system.
FIG. 11 is a block diagram of an absolute, self-calibrating receiving system.
FIG. 12 is a block diagram of a tracking generator for a wideband harmonic mixing receiver.
Description
The superheterodyne receiver of FIG. 1 comprises a mixer 12, a local oscillator 13, a detector 14, a calibration signal source 15, an RF input port 11 and an IF output port 16.
This receiver is designed to operate in either a reception or calibration mode. In the reception mode, an RF signal supplied to the RF input port 11 is transmitted to the mixer 12 where it is converted to an IF signal which appears at the IF output port 16. As there is no preselection, the tuning of this basic superheterodyne receiver is accomplished by merely changing the frequency of the local oscillator. In a more practical system, a preselector and an IF filter are usually necessary to eliminate spurious radiation and image responses; however, these considerations are not taken into account in the simple receiver of FIG. 1 to permit the presentation of only those components necessary to describe the present invention.
In the calibration mode, a fixed frequency IF signal from source 15 is supplied to the IF port of mixer 12 where it is converted to a signal at an RF frequency. This RF signal is always at the frequency to which the receiver is tuned by virtue of the fact that the mixer is reciprocal in frequency. For example, when an RF signal at the frequency to which the receiver is tuned is applied to the mixer, it will produce a signal at the IF frequency of the receiver and when a signal at the IF frequency of the receiver is applied to the mixer, it will produce an RF signal at the frequency to which the receiver is tuned. The mixer reciprocal properties remain true for up or down conversion and therefore the present invention is applicable for receivers using either type of conversion.
The IF calibration signal from source 15 is maintained at a constant frequency and amplitude. A wideband detector 14 is connected to the RF port of the receiver to measure the RF level. The ratio of the IF input signal at port 16 to the RF level as indicated by the output of the detector 14 is a measure of the RF gain of this receiver.
The calibration system described with the aid of the basic receiver of FIG. 1 may be applied to a more complex receiver, such as the receiver shown in FIG. 2. This receiver includes the calibration system of FIG. 1 as well as a microprocessor and a digital memory for automatically correcting the gain of the receiver in accordance with stored calibration data.
The receiver of FIG. 2 comprises an RF switch 202, a first RF attenuator 203, a preselector 204, a second RF attenuator 205, a mixer 206, an IF band-pass filter 207, an IF switch 208, an IF amplifier 209, an IF attenuator 226, an IF detector 227, an RF input port 201, a detector output port 228, an IF calibration signal source 225, a local oscillator 221, an RF detector 210, a video amplifier 211, a preset reference input port 217, a comparator 212, an analog to digital converter 213, a microprocessor and digital memory 214, a frequency tuning input port 224, a frequency tuning signal divider 223, and digital to analog converters 215, 219, and 220.
This receiver is designed to operate in either a reception or calibration mode. The mode of operation is determined by the position of the RF and IF switches. With the switches in the position shown in FIG. 2, the receiver is set to operate in the reception mode. In this mode, an RF signal at port 201 is passed through the switch 202, the RF attenuator 203, the preselector 204, the attenuator 205 to the mixer 206 where it is converted to a signal at the IF frequency of the receiver. The IF signal is passed through the bandpass filter 207, the switch 208, the IF amplifier 209, the IF attenuator 226 to the IF detector 227. The detected output is supplied to port 228.
Frequency tuning information is fed to the frequency tuning divider 223 where it is divided three ways and supplied to the digital to analog converters 219 and 220 and to the memory 214. The digital to analog converter 219 produces an analog voltage to tune preselector 204, while converter 220 produces a voltage to tune the local oscillator 221.
In the calibration mode, the switches 202 and 208 are switched. The signal from the IF calibration source 225 is fed through the switch 208, the band-pass filter 207 to the mixer 206 where it is converted to a signal at the RF frequency to which the receiver is tuned. The RF signal is passed through the attenuator 205, the preselector 204, the attenuator 203, the switch 202 to the detector 210 which produces a video signal that contains information indicating the gain of all the components through which the calibration signal has passed. The output of the detector is amplified in video amplifier 211 and then compared in the comparator 212 with a preset reference which corresponds to the level of the IF calibration signal to provide an output from the comparator 212 which represents a measure of the gain of the components through which the calibration signal has passed. The output of the comparator is passed through analog to digital converter 213 before it is stored in the memory 214.
The frequency tuning information is supplied directly to the memory 214 from divider 223 so that at the time the amplitude calibration data is stored in the memory, the frequency at which the receiver is tuned is also stored. In this way, each entry of the calibration data is identified by the corresponding frequency of the receiver at which the data was measured.
As the receiver is tuned across its RF band in the receiving mode, the calibration data, which is amplitude correction data, is retrieved from the microprocessor and memory 214 and supplied to the digital analog converter 215. The output of converter 215 controls the attenuator 226 which adjusts the receiver gain in accordance with the calibration data to provide a corrected receiver output at port 228.
The output of the converter 215 may be used to drive a gain adjusting device located in the RF, IF or video portions of the receiver. Where a log video amplifier is used to amplify the output of the detector, a subtraction circuit may be conveniently used in the video circuit to correct the gain as a logarithmic amplifier mathematically converts the gain adjustment function from multiplication to subtraction.
The microprocessor and memory unit 214 may be preprogrammed to compensate for nonlinearity or other errors in the detector and other calibration components over their respective frequency and dynamic ranges.
FIG. 3 illustrates a calibration system which embodies the principal of the invention illustrated in FIGS. 1 and 2, but eliminates the wideband detector shown in these figures. This can be a significant advantage as there can be difficulty in providing a wideband detector with a flat response over the entire RF band of the receiver. The system shown in FIG. 3 comprises reflective termination 31, a mixer 32, a local oscillator 33, an IF directional coupler 34, and IF calibration signal source 36, an IF detector 35 and an IF output port 37.
In calibrating this system, the IF calibration signal source supplies a first IF signal to the mixer for conversion to an RF signal. The RF signal produced in this manner is transmitted from the RF port of the mixer to the reflective termination 31 where it is reflected and returned to the mixer. The reflected signal is converted to a second IF signal which is detected by the IF detector 35. The flow of the signals from the IF source 36 to the detector 35 is shown by the arrows 38 above the drawing. As can be seen from these arrows, the calibration signal flows through the RF portion of the receiver twice, and therefore the level detected by detector 35 is an indication of twice the gain of the RF portion of the receiver.
The system of FIG. 3 offers a number of cost advantages. The IF detector 35 operates at a single frequency and therefore is lower in cost than the wideband detectors required for the systems shown in FIGS. 1 and 2. In most receivers there is usually an IF amplifier which can be used to amplify the second IF signal before detection, thereby reducing the sensitivity requirements of the IF detector and, in some cases, it may be possible to use the IF detector which is a standard part of the receiver for both calibration and reception.
There are a number of precautions which must be observed if useful results are to be obtained with the system of FIG. 3. All of the components must exhibit low VSWR and the directivity of the coupler must be high to keep the two IF signals separate. In addition, a constant, high VSWR is required for the reflective termination across the entire RF band of the receiver to provide sufficient reflected signal for detection. The VSWR and directivity requirements may restrict the usable frequency range over which this system may be used.
The present invention may also be used to produce a test signal which is always at the RF frequency to which a receiver is tuned. The production of this type of test signal usually requires a separate instrument, generally referred to as a tracking oscillator. The system shown in FIG. 4 illustrates one method of producing the RF test signal without such an oscillator. This system also embodies a number of methods for differentiating the test signal before and after it has passed through the device under test. These methods aid in reducing the directivity requirements of the directional coupler in this circuit.
The system of FIG. 4 comprises an amplifier 402, a modulator 403, a delay line 404, a first switch 405, a first directional coupler 406, a mixer 407, a local oscillator 409, a second directional coupler 413, an IF signal source 411, a second switch 410, a switch control input port 414, an IF output port 412 and a device under test 401.
In operating this system for the purpose of measuring the gain of a device under test, an IF signal supplied by the source 411 is passed through the switch 410 and the auxiliary port of the directional coupler 413 to the mixer 407 to produce an RF signal at the RF port of the mixer. The RF signal is passed through the directional coupler 406, the switch 405, the delay line 404, the modulator 403, the amplifier 402 to the device under test 401. The signal is then returned from the device under test to the auxiliary port of the directional coupler 406 where it is directed back through the mixer 407, the directional coupler 413 and finally to output port 412.
A means for measuring 415, such as an IF voltmeter or detector is connected to the IF output port 412 and to the IF signal source 411. By measuring the signal transmitted from the IF generator and that received at port 412, with, and then without, the device under test in the circuit, the gain of the device under test may be ascertained.
The directional couplers 406 and 413 separate the signal transmitted by IF source 411, from the signal returned through the receiver after passing through the device under test. If the directional couplers and the components exhibit ideal directivity and VSWR, only the IF signal, produced from the RF test signal after it has passed through the device under test, is received at the output port 412. However, the directional couplers and components in practical systems will exhibit less than ideal directivity and VSWR and a portion of the signal transmitted from the IF signal source 411 will leak or be reflected to the output port 412 without ever having passed through the device under test. The effect of the leaked or reflected IF signal is to produce erroneous output levels.
The leaked IF signal may be separated from the desired IF signal by several methods. In the first of these methods, the signals are separated by means of the delay line 404 and the switches 405 and 410. The IF signal is pulse modulated by the switch 410. Switch 405 is driven in synchronism with switch 410 by a common control signal supplied at the port 414. The synchronism of these switches will pass the pulsed signal to the front end. The pulse is delayed in the line 404 before it appears at the device under test 401. The switches 405 and 410 are then shut off to terminate the pulse. The pulse produced in this manner is passed through the device under test and then is returned through the receiving system for detection. While the switches are in the off position, the pulse returning through directional coupler 406 is prevented from entering the device under test by switch 405 and from entering the IF signal source 411 by switch 410.
A second method of separating the IF signals employs tone modulation of the test signal which is produced by the modulator 403. The tone modulated signal is detected at the IF output. The transmitted and return signals are separated by the fact that only the tone modulated signal will have passed through the device under test.
An alternate approach to producing a tracking test signal is illustrated by the dual channel receiving system shown in FIG. 5. The system of FIG. 5 comprises a first mixer 502, a local oscillator 503, a first switch 504, a first detector 505, a device under test 506, a second switch 507, a second mixer 508, and a third switch 509, a second detector 510, an IF signal source 501 and an output port 511.
In this system, an IF signal supplied by the IF source 501, is converted in mixer 502 to an RF signal at the frequency to which the receiver channel, comprising the mixer 508 and the local oscillator 503 is tuned. This frequency is produced in mixer 502 by virtue of the fact that both the mixers 502 and 508 are supplied by the common local oscillator 503. The RF signal produced at the RF port of mixer 502 may be used to test a device under test such as 506 or may be used for calibration by bypassing the device under test and feeding the signal directly to mixer 508. The bypassing is accomplished by switches 504 and 507.
The signal level produced by mixer 502 is monitored by the detector 505 which is connected to the RF output port of this mixer. The IF output of the receiver is measured by detector 510 which is connected to the output by means of switch 511. The difference in amplitude levels at the detectors supplies the information required to determine the receiver gain.
FIG. 12 is an embodiment of the present invention comprising an IF signal source 1201, a mixer 1202, a preselector 1203, a device under test 1204, a second preselector 1205, a second mixer 1206, a sweep control unit 1207, and a local oscillator 1208. The control unit transmits signals to various components of the system over control lines 1209, 1210, 1211, 1213 and 1214. The IF output from the receiver mixer is designated by drawing numeral 1215.
In the operation of the system of FIG. 12, an IF signal produced in the IF source 1201 is transmitted to the mixer 1202. The local oscillator 1203 supplies a portion of the receiver local oscillator signal to the mixer 1202. The IF signal is converted in the mixer 1206 to a signal at the RF frequency to which the receiver is tuned. The RF signal is passed through the preselector 1203, the device under test 1204, and the receiver preselector 1205 to the receiver mixer 1206, where it is converted to a signal at the IF frequency of the receiver.
The system of FIG. 12 is similar to the system of FIG. 5 with the exception that two preselectors and a preselector control unit have been added. The preselector 1203 aids in making the system of FIG. 12 particularly useful for receivers employing harmonic mixing. In such systems, the local oscillator is repetitively swept across its relatively narrow band while the preselector is swept across the complete RF band. The preselector 1203 selects the appropriate RF signal produced by mixing the IF signal with one of the harmonics of the local oscillator. Although the preselector 1205 could be applied alone to select the proper RF signal, the preselector 1203 accomplishes this prior to subjecting the device under test to an undesired number of signals present within mixer 1202. In addition, the use of the preselector 1203 enables this system to function as a sweep generator for general application, not directly connected with the receiver.
The sweep control unit 1207 supplies a number of signals including the same sweep voltage to the preselectors 1203 and 1205 to insure that the signal selected by the preselector 1203 is at the RF frequency of the receiver. The sweep control unit 1207 also supplies bias voltage to the two mixers. This bias voltage is adjusted to enhance the output of the desired harmonic.
Amplification of the local oscillator signal enhances harmonic production and thus harmonic mixing by producing higher level harmonics. This occurs for two reasons. The harmonic content present in the local oscillator signal are amplified and high level drive of the mixer diodes causes clipping and thus richer harmonic production.
A sweep voltage such as that which may be used in sweep control unit 1207, may be produced by a number of well known methods. One of the simplest is to apply the output of a sine wave oscillator to a clipping circuit to produce a square wave and then apply the square wave to an integrator to produce a sawtooth wave.
To cause the two identical preselectors to be at the same frequency at the same time, the same sweep voltage is applied both preselectors.
Where there is a difference in the preselector's control characteristic, a shaping network may be interposed to compensate for this difference. If, for example, the required sweep voltage for one preselector is a linear sawtooth rising from zero volts to one volt in one second, while the second preselector requires a sawtooth which rises to only one-half volt in one second, a shaping network may be interposed for the second preselector which is simply a voltage divider to cause the voltage delivered to one preselector to be half that of the other.
In more complicated cases where a nonlinear relationship is required, biased diodes may be employed in the shaping network to switch different resistors into the dividing network as different voltages are applied to the network.
A swept, narrow band, linear control voltage may be produced from a wideband control voltage by attenuating the wideband sweep signal and then adding an offset bias voltage by way of an operational amplifier. If, for example, a one volt sawtooth were attenuated to 0.25 volts and then a bias voltage of 0.5 volts were added, the sweep control voltage would start at 0.5 volts and increase to 0.75 volts, causing a linear device, requiring the 1 volt for a full sweep, to sweep only over only one-fourth of its range, the particular portion of the range depending on the value of the bias voltage.
In the example given above, the 0.5 volt bias would cause the device to sweep over the third quarter of its range.
The sweep control unit 1207 also supplies bias voltages to the mixers. It is well known that fundamental and harmonic mixing is performed with greater efficiency when the mixer diodes are biased for a particular operating range. For each frequency range, a different bias may be required. This may be supplied by a variable power supply within unit 1207. In some cases, it might be desired to have the mixer bias voltage swept as the input frequency is swept for maximum efficiency. The swept bias voltage for the mixer in this case may be produced from the same master square wave or sawtooth as is used for the preselector sweep by the same general type of shaping networks described in connection with producing the preselector sweep control voltage.
In a superheterodyne receiver, the local oscillator must track or follow the preselector frequency in some known manner. The known manner is usually a constant offset frequency equal to the center IF frequency. Tracking as used herein is defined on page 604 of IEEE Standard 100 (1972), as follows:
Tracking: the maintenance of proper frequency relationship in circuits designed to be simultaneously varied by ganged operations.
A shaping network may be again employed with the same "master" square wave or sawtooth to produce the desired control signal for the local oscillator.
An alternate method of producing the control voltages is by digital means. A clock may be used to drive a counter. The counter feeds a look-up table for each device to be driven, such as the preselector or local oscillator. The look-up tables provide digital outputs representative of the required voltage at a particular time. The output of each look-up table is connected to supply a digital to analog converter, which generates the desired analog control voltage for each device.
The preselectors and local oscillator tuning elements in many current receivers are yig filters. Most filters of this type exhibit a delay between the applied control signal and the tuning of the filter which prevents their following a fast retrace control signal. Generally a retrace signal is applied to the preselectors after the local oscillator has been swept over its range. The yig preselectors are unable to stop sweeping instantaneously which causes them to overshoot the frequency indicated by the control voltage when it stops sweeping. To overcome this problem, the preselectors are retraced over a portion of their range.
Similar components are used for both preselectors so that the preselectors will track each other despite the irregularities in the control caused by retracing. The use of similar preselectors and mixers and the application of the same control voltages and local oscillator signals overcome the tracking problems which make presently available tracking systems unsuitable for wideband systems and especially unsuitable for receiving systems employing harmonic mixing.
It should be noted that terms such as "a portion of the receiver local oscillator signal" or "a portion of the sweep control voltage" are intended to include the many obvious equivalent methods for producing replicas of the receiver local oscillator signal, sweep control voltage and mixer bias voltage.
In the case of the local oscillator a number of well known equivalent methods may be applied to produce a replica of the local oscillator signal. For example, a separate but similar oscillator supplied with the same control signal may be employed to produce the local oscillator signal for the mixer 1202. A second alternate is to employ a separate oscillator, phase locked to the receiver oscillator. Finally, a third alternate is to extract a portion of the receiver local oscillator signal and simply filter and amplify this signal to produce the desired replica.
A swept receiver generally incorporates a local oscillator drive circuit, such as shown in FIG. 6. The system of FIG. 6 is a dual channel receiver which comprises a first mixer 63, a common local oscillator 62, a second mixer 64 and a sweep control voltage to the local oscillator 62 which is usually a sawtooth waveform such as that shown in FIG. 7. In FIG. 7, ordinate 72 represents the voltage axis, the abscissa 71 represents the time axis, and the waveform 73 represents the ramp portion of a sawtooth waveform.
The systems employed to calibrate a receiver may be divided into two general classes. In the first class, the system relies on stored data, such as the system shown in FIG. 2. In the second, there is no reliance on stored data. Instead the receiver is either time shared to perform the calibration and measurement functions or these functions are carried out simultaneously. In stored data systems, the number of calibration points is limited by the capacity of the memory unit, whereas the continuous class is theoretically unlimited in this respect. However, the time shared class does require a longer measurement period than the stored data system because it successively measures and then calibrates. Despite the longer measurement period, the time shared class does provide two important advantages. All points may be calibrated and no memory unit is required.
The time shared class may be divided into two subclasses. In the first of these, the receiver is calibrated at one frequency and a short time later, before the frequency has changed appreciably, a receiver measurement is made. The calibration and measurement points are nearly at the same frequency so that there is little chance the calibration data has changed appreciably. For example, in FIG. 7 at time "a," the receiver may be calibrated and at time "b" a measurement may be made.
In the second subclass no calibration error occurs exept for drift. The reason for this is illustrated by the oscillator drive voltage shown in FIG. 8A. In this Figure, the voltage is represented by the ordinate 82 and time is represented by the abscissa 81. A staircase waveform 83 represents the sweep drive voltage. Waveform 84 shown in dashed lines represents a second staircase waveform. Two instants in time at the same frequency are represented by letters c and d.
In the operation of a system with this type of oscillator drive voltage, the receiver is calibrated at time c and a receiver measurement is taken at time d. During the time period from time c to time d, the oscillator control voltage remains constant and thereby maintains the local oscillator at the same frequency except for any tuning drift which might occur over this short interval.
The number of calibration points can be easily increased in a system using the type of staircase sweep drive voltage shown in FIG. 8A by varying the starting voltage of the staircase waveform on successive sweeps. Consider, for example, waveforms 83 and 84 in FIG. 8A. These waveforms are identical except for their starting voltages, which are represented by the letters e and f. If the sweep drive voltage 83 is first applied to the oscillator, calibration at time c will occur when the oscillator is at a frequency corresponding to the control voltage h, whereas when drive voltage 84 is applied, the oscillator frequency at time c will correspond to the voltage g. A different set of calibration and measurement points may be covered on successive sweeps by simply placing the staircase on a dc pedestal and shifting the pedestal to shift the starting voltage of staircase waveform. If this is done in a number of equal increments, the frequencies at which accurate measurements can be made is multiplied by the number of increments required to equal a single step.
A random or jittered increment in the dc starting voltage may also be used. If the value of the increment is conpletely random, over a period of time, virtually all frequencies will be calibrated. Although the time for producing virtually all frequencies may be quite long in a practical measurement operation, a jittered system has an advantage in that there is a chance that any selected frequency will eventually be produced, whereas a sweep drive voltage which is controlled by a uniform staircase waveform is guaranteed to never produce certain frequencies. This may be a serious drawback as important information may be present at the frequencies that are missed.
The staircase waveform may be modified as shown in FIG. 8B to produce all frequencies. In this Figure, the waveform 85 is formed of a series of waveform sub-sets comprising a ramp and a flat such as i and j. On the flat, calibration and measurement may be made while during the period of the ramp an output is provided without calibration. In a spectrum analyzer the output would be displayed over the complete frequency range and calibration would occur at points corresponding to the flats.
FIG. 8C shows a waveform 86 which appears to be a simple ramp with a number of calibration points indicated along the ramp, however, it is actually waveform similar to waveform 85 in FIG. 8B with the periods of the flats shortened to be insignificant in comparison to the period of the ramp. Viewed in another wave, the oscillator sweep is a ramp interrupted momentarily at a number of points for calibration. In a practical application, such as in a spectrum analyzer, this waveform would sweep the local oscillator and preselector across the operating frequency band in virtually a linear manner with short interruptions in the sweep for calibration and gain adjustment.
A system for applying any of the waveforms of FIG. 8 is shown in FIG. 9. This system comprises a switch 902, a mixer 903, a second switch 907, an attenuator 908, a detector 909, a local oscillator 904, a sweep drive circuit 905, an IF signal source 906, a sweep generator 910, a device under test 901, and an IF output port 911.
In the receive mode of operation of this system, a swept signal from the sweep generator 910 is passed through the device under test 901 and switch 902 to the mixer 903 which converts the RF signal to an IF signal. The IF signal is passed through the attenuator 908 to the IF output port 911. The sweep generator 910 is designed to track the local oscillator frequency to maintain the generator signal at the frequency to which the receiver is tuned. This is accomplished by way of a connection between the local oscillator and the sweep generator indicated by line 912.
In the calibration mode, a signal from IF generator 906 is passed through the switch 907, to the mixer 903 where it is converted to an RF signal. The RF signal is passed through the switch 902 to the detector 909. The output of detector 909 is fed back to attenuator 908 to automatically adjust the level of the attenuation for a corrected output as described previously in the discussion concerning the system of FIG. 2.
The sweep drive generator 905 produces a staircase waveform such as that shown in FIG. 8A. The switches 902 and 907 are switched to permit calibration at a time c. The attenuator is properly set by the calibration system for the frequency to which the receiver is tuned. At a time d, while the frequency of the receiver remains the same, the switches are returned to their former positions placing the receiver in the reception mode. A correct measurement can then be made. The output of the receiver is automatically corrected by the calibration correction signals transmitted to the attenuator 908 from the detector 909. The correction signal is stored for the short period from time c to time d. The storage device for this short time span may be a sample-and-hold circuit shown housed with detector 909 in FIG. 9. A large digital storage unit such as the memory 214 in FIG. 2 is not required.
Calibration and measurement may also be carried out simultaneously to remove the limitations on the number of calibration points and delays in measurement occurring in time sharing system. A system illustrating the way in which this may be accomplished is shown in FIG. 10.
This system comprises a directional coupler 1001, a mixer 1002, a first divider 1003, a first IF amplifier 1007, an attenuator 1008, a second IF amplifier 1004, an IF detector 1005, a tone filter 1002, a local oscillator 1010, a mixer 1011, a tone modulated IF source 1015, a second divider 1012, an RF detector 1014, an attenuator control and delay circuit 1016, an RF input port 1013 and an IF output port 1009.
When this system is operated in the reception mode, an RF signal applied to the input port 1013 is passed through the directional coupler 1001 to the mixer 1002 where it is converted to a signal at the first IF frequency. This signal is passed through the divider 1003, the first amplifier and the attenuator 1008 to the IF output port 1009.
Calibration is carried out by generating an RF signal from the IF signal source 1015. This source produces an IF signal that is at the center frequency of the second IF amplifier. The passbands of the first and second IF amplifier are sufficiently separated so that a signal at the center frequency of one will be rejected by the other. The IF signal from the signal source is supplied to the mixer 1011 where it is converted to a signal at an RF frequency. This signal is passed through the divider 1012, the auxiliary port of the directional coupler 1001 to the mixer 1002 where it is converted to the same IF frequency as the IF source by virtue of the fact that both mixers receive a signal from a common local oscillator. This IF signal is passed through the divider 1003, and the second IF amplifier 1004 to the IF detector 1005. The output of the IF detector is passed through the tone filter 1002 to the attenuator control circuit 1016. A portion of the RF signal from the mixer 1011 is supplied to the RF detector 1014 by way of the divider 1012. The output of the RF detector is also supplied to the attenuator control circuit. The output of the attenuator control circuit controls attenuator 1008 after a delay to account for the difference in time between calibration and measurement at a particular frequency.
The RF signal produced in mixer 1011 passes through all of the RF components of the receiver. Its level before and after passing through these components is detected by the RF detecjtor and the IF detector respectively. The ratio of the outputs of these detectors indicate the gain of the RF components of the receiver. The gain information is determined in the attenuator control circuit to produce a control signal for attenuator 1008. The output at port 1009 is thereby adjusted in accordance with the gain variations determined by the detector outputs.
The IF signal produced in the source 1015 may be tone modulated and the tone may be filtered after detection in filter 1002 to aid in separating the first and second IF signals. This technique can also be applied to eliminate one of the IF amplifiers by using a single IF frequency.
In the operation of the system shown in FIG. 10, a sweeping tracking oscillator linked to the local oscillator may be used to provide a swept input signal. Such sweeps are usually from a low frequency to a high frequency with the output being blanked on retrace. With this arrangement, the second IF frequency may be set to be at a higher frequency than the first IF so that the calibration signal frequency leads the test signal. In this way, there is a time lag in which to take the calibration measurement and adjust the attenuator 1008 to set it at the proper value for the test signal at the time the test signal is received.
A self-calibrating receiving system such as shown in FIG. 2 may be absolutely calibrated over its entire frequency range by correcting the internal IF calibration system at one frequency with a standard. Absolute calibration, as used herein, refers to calibration with a standard such as a signal source with a known output. When absolute calibration has been completed, the output of a receiver such as the display of a spectrum analyzer may be read in terms of absolute values such as decibels above or below zero dbm. The difference in the gain between that measured with an absolute standard and that measured with an internal IF calibration system constitutes a correction factor for the internal calibration system. The internal IF calibration system shown in FIG. 2 can be adjusted with this correction factor for all RF frequencies.
FIG. 11 illustrates a system designed to provide absolute calibration of a receiver over its entire frequency range with the aid of a single frequency, absolute signal source. Such a source provides a known power output at one frequency. The system of FIG. 11 comprises a first switch 1101, a second switch 1102, a single frequency absolute source 1103, a receiver 1104, an IF source 1105, a detector 1106, a calibration system 1107, a microprocessor and memory unit 1108, a receiver output port 1109, and a analog to digital converter 1110.
The receiver 1104 shown in FIG 11 comprises essentially all of the components in the receiver shown in the upper portion of FIG. 2 with the exception of a calibrated IF signal source, which is shown separately in FIG. 11, as source 1105. Similarly, the calibration system 1107 shown in FIG. 11 comprises essentially all of the components in the calibration circuitry shown in the lower portion of FIG. 2 with the exception of a microprocessor which is shown separately in FIG. 11 as microprocessor 1108.
In the absolute calibration of the receiver of FIG. 11, a representation of the power output provided by the absolute source is entered into the memory 1108 and switches 1101 and 1102 are set to the positions to the right, as shown. The absolute source 1103 is connected to the input of the receiver 1104 by way of the switch 1101, while the output of the receiver is fed through the switch 1102 to the analog to digital converter 1110 where this signal is converted to digital form and supplied to the microprocessor 1108. The receiver output signal is compared with the representations of the power delivered by the absolute source 1103 in the microprocessor 1108 to determine the absolute gain of the receiver. The absolute gain is then compared with the gain determined by the internal IF gain calibration system at the RF frequency of the absolute source 1103 to produce a correction factor which is then applied to the gain determined by the internal calibration system at all RF frequencies of the receiver.
Absolute gain calibration may also be carried out with an absolute source at IF. This merely requires substituting an absolute IF source for the calibration source 1105. The absolute RF source, such as 1103, can be eliminated in a system employing this type of IF source. As an IF source such as 1105 is generally an internal source that may be subject to drift with age, the output level over a period of time may be monitored and corrected with the aid of a separate detector, such as the detector 106, and a correction program contained in the microprocessor 1108.
In the calibrating system of FIG. 11 with the internal IF oscillator, the switchs 1101 and 1102 are set to the positions to the left. The output from the detector 1106 is supplied to the microprocessor 1108 through the switch 1102 and the analog to digital converter 1110. This signal is compared with the RF signal produced within the receiver by the IF source after it has passed through the RF portion of the receiver 1104. This RF signal is supplied to the microprocessor through the switch 1101 and the calibration system 1107. The comparison produces a continuous correction at all RF frequencies to which the receiver is tuned.
The absolute IF calibration system may be simplified by making the comparison in the microprocessor with a representation of the known output power of the absolute IF calibration source that has been entered and stored in the microprocessor, in a manner similar to that carried out for the absolute RF source 1103. This simplified system eliminates the need for the monitoring components such as the detector 1106, the switch 1102, and the analog to digital converter 1108.
Although the calibration method illustrated here apply to gain, it is obvious that phase or other parameter may be measured in a similar manner with the principal change being merely the substitution of a phase detector or other appropriate detectors for the amplitude detector shown.
Claims
I claim:
1. Apparatus for measuring the gain of a device under test, which includes a superheterodyne receiver comprising: (a) means for applying a first IF signal to the mixer of the receiver, (b) means for extracting from the receiver an RF signal produced from said IF signal by the mixer, (c) means for applying said RF signal to a device under test, (d) means for applying said RF signal after passing through the device under test to the RF port of the receiver for conversion in said mixer to a second IF signal, (e) means for separating said second IF signal from said first RF signal, and (f) means for measuring the relative signal strength of said first and second IF signals after separation to determine the gain of said device under test.
2. Apparatus for generating an RF signal which is at the frequency to which a superheterodyne receiver is tuned for application to a device under test where the RF signal after passing through the device under test is applied to said superheterodyne receiver for detection and measurement purposes, said receiver having a preselection device, the tuned frequency of which is controlled by a preselection control voltage, comprising: (a) a mixing means separate from the mixing means within the superheterodyne receiver, (b) means for generating a first IF signal at the IF frequency of said receiver, (c) means for applying said first IF signal to said separate mixing means, (d) means for extracting a portion of the local oscillator signal from said receiver and applying it to said separate mixing means, (e) means for extracting an RF signal from said separate mixing means, said RF signal forming a signal at the RF frequency to which said receiver is tuned, (f) a preselection means separate from the preselection means within the receiver, (g) means for connecting said separate preselection means between said separate mixing means and said R.F. extracting means, (h) means for applying a portion of the receiver preselection control voltage to said separate preselection means to cause said separate preselection means to be tuned to the same frequency as the preselection means within said receiver, and (i) means for connecting said RF signal at the RF frequency to which said receiver is tuned after it has passed through said separate preselection means to said device under test.
3. Apparatus for generating an RF signal which is at the frequency to which a superheterodyne receiver is tuned for application to a device under test where the RF signal, after passing through the device under test is applied to said superheterodyne receiver for detection and measurement purposes, said apparatus being especially adapted for application where a harmonic mixing type of receiver is employed and said harmonic mixing receiver including a bias voltage applied to its mixer to enhance harmonic mixing, comprising: (a) a mixing means separate from the mixing means within the superheterodyne receiver, (b) means for generating a first IF signal at the IF frequency of said receiver, (c) means for applying said first IF signal to said separate mixing means, (d) means for extracting a portion of the local oscillator signal from said receiver and applying it to said separate mixing means, (e) means for amplifying said extracted portion of the local oscillator signal prior to application to said separate mixer to enhance the generation of harmonics within said mixer for harmonic mixing purposes, (f) means for extracting a portion of the bias for the mixing means within said receiver, (g) means for applying said portion of the bias to said separate mixing means to enhance the generation of a specific local oscillator harmonic, and (h) means connected between said separate mixer and said device under test to extract said R.F. signal from said separate mixer for application to said device under test.