Microprocessor Controlled Self-Tuning Resonant Cavity and Method

The invention relates to methods and apparatus for tuning resonant cavities, and particularly to systems in which reflected power and/or standing wave ratios are sensed by circuitry, especially microprocessor controlled circuitry, to effectuate generating of motor control signals that adjust an internal tuning element in the cavity to minimize power reflected from the cavity.

Transmitter combiners are multiple tuned cavity devices which allow simultaneous transmission of signals from a plurality of transmitters at different but closely spaced frequencies by means of a single antenna. Transmitter combiners include one tuned cavity for each frequency, typically. Each of the tuned cavities is coupled by a coaxial cable to a separate transmitter and is also coupled to a common coaxial connector to which a single antenna cable is connected leading to the antenna. Recently, the mobile telephone communication technology has been rapidly growing. Numerous holders of FCC licenses have established "cells" or regions that are typically several miles in extent, each having low power transmitters generally centered in a particular cell. A major problem that has always existed for transmitter combiners, and for tuned cavities in general, is the fact that numerous external influences cause the tuned cavities to become detuned, substantially increasing their insertion losses and decreasing the amount of transmitter power that reaches the transmitter antenna. Variations in critical dimensions of tuned cavities due to normal changes in the temperature of the cavities has always been a major problem. To avoid detuning of cavities due to temperature, tuned cavities, especially square prism filters, commonly have been made of Invar, which is an expensive metal that must be coated with copper to provide the high surface conductivity required in a tuned cavity that is useful in the high frequency bands assigned for mobile communications. Variations in humidity in many parts of the country are great enough to severely detune the cavities of transmitter combiners, and variations in atmospheric pressure also can significantly detune such cavities. Up to now, there has been no practical means of compensating for such detuning, and users of mobile communications equipment have had to live with the fact that often an inadequate amount of the transmitter power actually reaches the transmitting antenna. Mobile communications systems commonly utilize one of the transmitter channels as a "signaling channel" which controls transmission of messages between the cell site and the mobile units to effectuate switching operations to determine on which channel a particular cell and mobile unit shall operate. If the signaling channel becomes inoperative for any reason, the entire cell becomes inoperative until repairs can be made. Although it is possible to reassign channel frequencies remotely, allowing one of the voice channels to operate as a signaling channel, up to now it has been impossible to remotely make corresponding changes in the resonant frequencies of the various cavities of a transmitter combiner. Therefore, whenever it has been advantageous to reassign channel frequencies, it has been necessary for a technician to travel to the antenna site and manually tune the cavities to the newly assigned transmitter frequencies. This is an expensive and time consuming procedure which it would be very desirable to eliminate.

It is clear that there is a unmet need for an economical, automatic, self-tuning tuned cavity which continuously and automatically self-tunes its resonant frequency to the frequency of the incoming transmitter signal, regardless of the influence of external factors such as temperature, humidity, atmospheric pressure, etc.

Accordingly, it is an object of the invention to provide an economical tuned cavity that automatically self-tunes its resonant frequency to the frequency of an exciting signal fed into the cavity.

It is another object of the invention to provide a transmitter combiner with the capability of having its transmitter signal frequencies remotely changed without the necessity of manually retuning the resonant frequencies of the corresponding cavities of the transmitter combiner.

It is another object of the invention to provide a transmitter combiner in which all of the tuned cavities are automatically repetitively self-tuned to their respective transmitter frequencies and which occupies a minimum amount of space.

It is another object of the invention to provide a tuned cavity that has a resonant frequency which is stable with respect to an incoming exciting signal regardless of the material of which the cavity is composed and regardless of the influence of external factors such as temperature, humidity, atmospheric pressure, etc.

Briefly described, and in accordance with one embodiment thereof, the invention provides a self-tuning tuned cavity which senses power reflected from the cavity and power incident to the cavity and, in response to the sensed reflected power and incident power, determines whether the cavity needs tuning and generates motor control signals to control the rotation of a motor that displaces a tuning element located inside of the tuned cavity in such a way as to reduce the reflected power, thereby tuning the resonant frequency of the cavity to the frequency of the incident power signal, regardless of the influence of temperature, air humidity, air pressure, etc.

In the described embodiment of the invention, detector diodes in a cable connector receiving incident power from a transmitter cable produce signals indicative of the incident power to the cavity and the power reflected from the cavity. The signals produced by the diodes are converted to digital numbers by an analog-to-digital converter controlled by a microprocessor. The microprocessor, in accordance with a stored tuning program, reads and stores the digital numbers representing the incident power and the reflected power. If the ratio of the reflected power and incident power exceeds a particular threshold, the microprocessor generates motor control pulses applied to a stepper motor that is mechanically coupled through the wall of a tuned cavity to a tuning element therein, modifying the orientation of the tuning element and slightly modifying the resonant frequency of the cavity. The microprocessor repetitively samples the reflected and incident power and generates control pulses to the stepper motor in such a fashion as to minimize the reflected power. Since the reflected power of a tuned cavity is minimum when the resonant frequency is tuned precisely to the frequency of the incident power, the tuned cavity remains tuned precisely to the frequency of the resonant power despite changes in temperature, humidity, air pressure, and other variables which would otherwise result in detuning of the cavity relative to the transmitter frequency. In the described embodiment of the invention, a single microprocessor control system is "time shared" between the stepper motors of a large number of tuned cavities each having a respective stepper motor controlling an internal tuning element. The self-tuning system makes possible maximum coupling of transmitter power through the cavities of a transmitter combiner and conduction thereof to the transmitting antenna, and also makes it possible to conveniently reallocate transmitter channel frequencies without the need to make manual adjustments to the corresponding tuned cavities.

In one described embodiment of the invention, the microprocessor simply computes the standing wave ratio and advances the stepper motor, always in the same direction, until the standing wave ratio exceeds a threshold value, and then simply advances the stepper motor a predetermined number of steps to approximately "center" the resonant frequency of the cavity about the frequency of the incident power.

FIG. 1 is a block diagram of a simplified self-tuning transmitter combiner of the present invention.

FIG. 2 is a graph of reflected power of a tuned cavity versus the frequency of the signal exciting the cavity.

FIG. 2A is a graph of the ratio of incident power to reflected power of a tuned cavity versus the frequency of the signal exciting the cavity.

FIG. 3 is a partial cutaway view of an assembly of square prism filters which are automatically self-tuned in response to the microprocessor control system shown in FIG. 1.

FIGS. 4A through 4C constitute a program flow chart of the self-tuning program executed by the microprocessor of FIG. 1.

FIGS. 5A and 5B constitute a detailed logic diagram of the microprocessor system of FIG. 1.

FIG. 6 constitutes a program flow chart of an improved self-tuning program executed by the microprocessor of FIG. 1 to, in effect, time-share the time of the microprocessor among the plurality of tuned cavities of the transmitter combiner of FIG. 1.

FIG. 7 is a graph useful in explaining a simplified embodiment of the invention.

Referring to the drawings, particularly FIGS. 1-3, self-tuning transmitter combiner system 1 includes sixteen tuned cavities collectively referred to by reference numeral 15, and individually referred to by reference numerals 15-1, 15-2. . .etc The tuned cavities can be of any variety that is externally tunable, for example, the square prism filter cavities described in commonly assigned Pat. No. 4,644,303 entitled "MULTIPLE CAVITY SQUARE PRISM FILTER TRANSMITTER COMBINER WITH SHARED SQUARE WALLS AND TUNING CONTROLS MOUNTED ON RECTANGULAR END WALLS", filed Mar. 13, 1984 by Dickson et al., and incorporated by reference herein, wherein external manipulation of the internal tuning elements is accomplished by control elements located on the non-square faces of the square prism cavities. This structure allows the cavities to be constructed with common or abutting square walls, thereby saving a great deal of space. Alternately, the tuned cavities can be conventional cylindrical cavities with conventional axial tuning stubs. Or, cantilevered tuning elements such as those illustrated in FIG. 3 can be supported by the cylindrical walls of a conventional cylindrical cavity, if desired.

Each of the tuned cavities 15-1, 15-2, etc. in FIG. 1 has an internal tuning element such as 10 coupled by a rod 8 to a stepper motor and reduction drive unit such as 11-1 in FIG. 1.

Each of the cavities 15 has a coaxial input coupled by means of a coaxial cable such as 3-1 to a separate transmitter such as transmitter 2-1. Each of the coaxial cables, which are collectively referred to as cables 3, have a pair of diode detectors such as 5-1 built into it for the purpose of detecting the amount of incident power (P.sub.I) indicated by arrow 16 and the amount of reflected power (P.sub.R) indicated by arrow 4 flowing in the cable 3. The diode detectors are collectively referred to by reference numeral 5. The two detector diodes in each of the diode detector assemblies 5 produce two voltage signals V.sub.I and V.sub.R, respectively, indicative of the amplitude of P.sub.I and P.sub.R, respectively.

Each of the pairs of diode detector outputs such as 6-1 and 7-1 are connected to a respective pair of inputs of an analog multiplexer circuit 9.

A selected one of the diode detector output signals is multiplexed to the analog input of an analog-to-digital converter 19 by means of a conductor 17. The resulting digital number produced by analog-to-digital converter 19 is produced on a plurality of conductors 18 that are connected to a microprocessor 14, which is understood to include an appropriate amount of memory, including program storage memory such as electrically programmable read-only memory, (PROM) and data memory, such as high speed random access memory (RAM). Certain outputs 12 of microprocessor 14 are fed into a decoder 22 to produce sixteen selection signals 23 that are connected to "enable" or "select" sixteen separate data registers 21. Some of the conductors 12 are fed back to multiplexer 9, as indicated by reference numeral 12, to allow microprocessor 14 to select which of the 32 (two for each of the sixteen cavities) diode detector signals 6-1, 7-1, etc., are to be coupled to the input of analog-to-digital converter 19. Other outputs 20 of microprocessor 14 are connected to load data into sixteen registers 21, the outputs 24 of which are connected to sixteen motor driver circuits 25. Four of the outputs 24 are connected to each of the stepper motor drivers. The sixteen outputs of motor driver circuits 25 are designated by reference numerals 26-1 through 26-16, and are connected to the control inputs of stepper motors 11-1 through 11-16, respectively.

As will be described in more detail subsequently, microprocessor 14 executes a subroutine that, for one cavity at a time, generates signals on bus 20 and loads them into registers 21 and causes the motor drivers 25 to adjust the orientation of the internal tuning rods 10 so that the tuned cavities are essentially continually kept precisely tuned to predetermined frequencies.

At this point, it will be helpful to refer to FIG. 2, which shows the reflected power (P.sub.R) characteristic for a typical tuned cavity. When the cavity is perfectly tuned to the frequency of an exciting signal, the value of the reflected power (P.sub.R) is at a minimum, as indicated by point 34 on tuning curve 31. If the cavity is badly out of tune, the value of P.sub.R is high, at points such as 32 and 35. Point 33 indicates an intermediate detuned condition of the cavity.

Those skilled in the art will appreciate that for a particular group of, for example, sixteen tuned cavities, each will have a somewhat different characteristic curve such as the one shown in FIG. 2, and that some cavities in a cluster will "tune up" more precisely than others. As previously mentioned, variations in temperatures of the cavity and in the humidity of the ambient air will cause detuning of a cavity that has been initially tuned under different conditions.

Referring to FIGS. 2 and 2A, the basic operation of the self-tuning transmitter combiner system 1 of FIG. 1 is that the microprocessor system 14 continually samples the diode detector output voltages of each of the sixteen transmitter cables 3 and, if these voltages indicate that there is incident power and that the cavity is out of tune (a cavity is deemed "out of tune" if V.sub.I /V.sub.r <(V.sub.I /V.sub.R) THRESH, where (V.sub.I /V.sub.B) THRESH is an arbitrary but suitable threshold, designated by reference numeral 35A in FIG. 2A), causes the stepper motor to adjust the tuning element of that cavity in such a manner as to minimize the reflected power from that cavity, provided that the transmitter connected to the subject cavity is transmitting a minimum amount of power to it, and provided the standing wave ratio, i.e., the ratio of the incident power to the reflected power exceeds a certain minimum level 35A. If there is no incident power being transmitted to a particular cavity, of course there is no need to tune that cavity. Once the preset threshold value of the standing wave ratio is exceeded, then a subsequently described routine is executed by microprocessor 14 to adjust the position of tuning element 10 until the minimum point 34 (FIG. 2) is obtained for the reflected power sensed by the reflected power diode detector 5 built into the transmitter cable connector.

The foregoing transmitter combiner structure and operation has numerous major advantages over any prior transmitter combiner system. Those skilled in the art will appreciate that maximum transmission of power through a cavity of a transmitter combiner occurs when the cavity is precisely tuned to the transmitter frequency. If the cavity is not tuned to the frequency of the transmitter, then the reflected power is large, and can be equal to the incident power if the cavity is completed detuned because under these conditions, the cavity looks like an open circuit, resulting in all of the incident power being reflected back down the cable to the transmitter. The construction of most square prism filter cavities of expensive, temperature-insensitive, copper clad Invar material, and the use of various bimetal automatic temperature compensating internal tuning elements can be avoided by using the above-described self-tuning system. The self-tuning system allows tuned cavities to be made of very inexpensive material, such as pure copper, without concern for temperature dependence of the dimensions of the cavity.

The above-described self-tuning cavity system avoids the need for factory tuning of the cavity, and avoids the need for a technician to travel to the antenna installation site to manually tune the various cavities of a transmitter combiner both at the intial installation time and at any subsequent time when it is necessary to change transmitter frequencies. The invention avoids the inherent detuning of cavities in climates in which there are great ranges in the ambient temperature, humidity, and/or air pressure which are known to greatly detune transmitter combiners. Another advantage of the self-tuning transmitter combiner of the present invention is that it allows transmitter frequencies to be reassigned by remote control almost instantly, as the self-tuning system of FIG. 1 will automatically cause each of the sixteen cavities to be retuned immediately to the newly-assigned transmitter frequencies. This will enable a communications company to temporarily reassign frequencies that allow a much larger number of subscribers than normal to operate mobile telephones in a particular cell area without overloading the system.

Another advantage of the self-tuning system is that the entire communication system does not automatically fail if the signaling channel (a channel that is used to identify a particular vehicle and instruct its transmitter what frequency it needs in order to make voice contact with the mobile telephone cell) fails, because one of the voice channels can be remotely temporarily reassigned to the frequency of the signaling channel. FIGS. 5A and 5B show a detailed implementation of the system shown in FIG. 1. Where applicable the same reference numerals have been used in FIGS. 5A and 5B as FIG. 1. Microprocessor 14 can be Motorola MC146805E2CP CMOS microcomputer. A clock synthesis circuit 40, which can be a Motorola MM74HC74N integrated circuit, derives the necessary clock signals for the analog-to-digital converter 19 from the microcomputer clock signal on pin 38 of the MC146805E2CP. The crystal, resistor, and two capacitors on pins 38 and 39 of the microcomputer set the microcomputer clock frequency. Analog-to-digital converter 19 can be a TLC541N eight bit analog-to-digital converter manufactured by Texas Instruments. The analog muliplexer 9 is implemented by means of three eight-channel CD4051BCN analog multiplexers manufactured by National Semiconductor (RCA also makes the CD4051) and designated by reference numerals 9A, 9B, and 9C in FIG. 5A. The detector diodes which measure the reflected power produce voltage signals on lines designated RFL0, RFL1 . . . RFL15, and the lines on which the detector diode voltage signals representing the incident power are produced are designated by INC0, INC1, . . . INC15. The various resistors shown in FIG. 5A are used to "low pass filter" the detector diode output signals. The conversion clock for analog-to-digital converter 19 is generated by the clock 9enerator circuit 40. (Note that the potentiometers shown in FIG. 5A are a part of the detectors, not part of the printed circuit board, and they are used to adjust the full scale output voltage range of the detectors.)

Decoder 22 is implemented by means of an MM74HC154N four-to-sixteen line integrated circuit decoder. The program executed by microprocessor 14 is stored in a 2,048.times.8 bit CMOS electrically programmable read-only memory designated by reference numeral 14A, which can be an NMC27C16Q-45 integrated circuit EPROM manufactured by National Semiconductor. The registers 21 in FIG. 1 are implemented by means of the sixteen MM74HC175 integrated circuit quad D flip flops generally designated by reference numeral 21 in FIG. 5B. The motor drivers 25 can be implemented by means of the descrete power transistor/diode circuits shown within the dotted line 25, or can be implemented by means of ULN2803 driver circuits manufactured by Sprague. The MM74C373 CMOS octal latch in FIG. 5B performs the function of demultiplexing the address and data lines coming from the microcomputer.

An algorithm that can be executed by microprocessor system 14 to effectuate the above-described operation for a particular tuned cavity is shown in the flow chart of FIGS. 4A and 4B.

Next, two programs executed by microprocessor 14 will be described in detail. The first program described with respect to the flow chart of FIGS. 4A-4C, is an initial program that was written to tune a single cavity. The second program, described with reference to FIG. 6, that was written to tune sixteen cavities by means of a single microprocessor 14. The latter program is more suitable for transmitter combiners, while the former is suitable for applications in which there is only a single cavity.

Referring now to FIGS. 4A-4C, the automatic tuning algorithm is entered at label 100 and immediately enters block 101 and clears the registers 21 and also some light emitting diodes (not shown) that indicate the present system condition, and also stops the stepper motor of the cavity to be tuned. The program then goes to block 102 and obtains the present values of the P.sub.R and P.sub.I for the cavity by executing a subroutine called ADC, which causes multiplexer 9 of FIG. 1 to couple the detector diode outputs to the input of analog-to-digital converter 19 and causes the analog-to-digital converter 19 to produce corresponding digital numbers representing the present values of P.sub.R and P.sub.I. Those skilled in the art can readily provide various subroutines that can generate the necessary control signals to multiplexer 9 and analog-to-digital converter 19 to effectuate this foregoing operation.

The same subroutine also causes microprocessor system 14 to read the digital values of P.sub.R and P.sub.I and store them in software registers in the microprocessor memory labelled INCIDENT and REFLECTED, respectively.

The program then enters decision block 103 and determines if the value of the variable INCIDENT, i.e., P.sub.I, is less than six. If this determination is affirmative, the program loops back to block 102, but if the decision is negative, the program goes to block 104.

In block 104 the program updates two software registers (stored in the microprocessor memory) named OLD INCIDENT and OLD REFLECTED and assigns to them the values of INCIDENT and REFLECTED, respectively, stored in block 102. Thus, the prior values of P.sub.I and P.sub.R are stored and the microprocessor 14 can now take new readings of P.sub.I and P.sub.R without erasing the old values.

The program then goes from block 104 to 105 and jumps to a subroutine that advances the stepper motor 11 by two steps in a particular direction. The program then causes the microprocessor 14 to jump to a subroutine 105B to create a twenty-four millisecond time delay. Note that the reason for this millisecond time delay is to allow the system enough time for accurate, stable analog to-digital converter readings of P.sub.I and P.sub.R.

The program then goes to block 106 and again jumps to the ADC subroutine, reading new values of P.sub.I and P.sub.R converting them to digital members, and storing new values of INCIDENT and REFLECTED. Once the new values of INCIDENT and REFLECTED have been obtained, the program goes to decision block 107 and determines if the new value of INCIDENT is less than six.

If the incident power is determined in block 107 to be greater than six, the program goes to block 108, wherein the microprocessor jumps to a subroutine that calculates the standing wave ratio, i.e., the ratio of the latest value of P.sub.I divided by P.sub.R and sets a variable called RATIO to that value. The program then goes from block 108 to block 109 and determines whether or not the standing wave ratio is greater than or equal to six. If it is not, the program goes to block 112 and causes the system to wait for a twelve millisecond time delay to elapse. The program then goes to block 113 and causes microprocessor 14 to again jump to the ADC subroutine, obtaining new values for P.sub.I, P.sub.R, INCIDENT, and REFLECTED.

The program then goes to block 136 and, as in block 104 above, updates OLD INCIDENT and OLD REFLECTED with the new values of INCIDENT and REFLECTED, respectively. The program then goes to block 114 and executes a subroutine which causes the stepper motor 11 to advance by one step.

The program then goes to block 115, wherein the microprocessor 14 again executes the ADC subroutine and obtains new values of P.sub.I, INCIDENT, and REFLECTED. The program then goes to block 128 and executes the previously mentioned RATIO subroutine which computes a value of RATIO wave ratio based on the latest values of INCIDENT and REFLECTED.

The program then goes to decision block 129 and determines if the new value of RATIO is greater than or equal to six. If this determination is affirmative, the program goes from block 129 back to block 104 and repeats the foregoing loop. If the determination of decision block 129 is negative, i.e., if the standing wave RATIO is less than six, the program goes back to block 114, executes the ADC subroutine, causes the stepper motor to advance one more step, and continues through blocks 115 and 128 and back to decision block 129 as before.

The program contains a software counter which counts the number of times the loop including blocks 114, 115, 128, and 129 is passed through. If this loop is completed 256 times without the standing wave ratio attaining a value of greater than six, as determined by decision block 129A, the program goes to decision block 130. If the standing wave ratio is determined, in decision block 129, to be greater than six, it means that it has exceeded a predetermined threshold, such as the value of point 33 in FIG. 2, and that initial tuning by the stepper motor in response to microprocessor 14 is in the correct direction, and it is time for the program to go into a sequence of instructions that causes the microprocessor to determine tuning rod orientation at which the minimum value of P.sub.R occurs. In other words, from this point on the microprocessor 14 will be looking for a P.sub.R minimum, i.e., a valid "negative peak" such as 34 in FIG. 2.

After decision block 129A has determined that the loop including blocks 114, 115, 128, and 129 has been passed through 256 times and decision block 130 has been entered, the microprocessor determines if the present value of the variable REFLECTED is greater than the earlier stored value of the variable OLD REFLECTED. If this determination is negative, it means that the program has been causing the stepper motor to step in the correct direction, and the program then goes to block 132, causes the stepper motor to advance one step and then proceeds to block 133. In block 133, the program causes the microprocessor 14 to jump to the ADC subroutine and obtain new values of the variable INCIDENT and REFLECTED. The program then goes to block 134 and determines if the new value of INCIDENT is less than six. If this determination is affirmative, the program returns to block 101 and clears the various registers, turns off the light emitting diodes, and stops the stepper motor because the value of INCIDENT is so small that it indicates that there is no significant power being received by the cavity being tuned from the transmitter.

If the determination of decision block 134 is negative, the program goes to block 135 and jumps to the standing wave ratio computing subroutine and computes an updated value of the variable RATIO. The program then goes to decision block 136 and determines if the standing wave ratio is greater than or equal to six. If it is not, the program returns to block 132, and causes the stepper motor to advance one more step. If the determination of decision block 136 is affirmative, the program goes to block 104.

On this next pass through blocks 104 through 109, the determination of decision block 109 will be affirmative. The program will move into a new loop which looks for a valid minimum point such as 34 of the reflected power curve of FIG. 2.

If the above determination of decision block 130 is that the value of REFLECTED is greater than the prior value OLD REFLECTED, it means that the stepper motor is tuning the resonant frequency of the cavity in the wrong direction, moving away from a reflected power minimum point such as 34 FIG. 2. In this event, the program goes to block 131 and changes the direction in which the stepper motor advances and then goes to block 132 and executes the following blocks in the manner previously described.

If the value of the standing wave ratio variable RATIO ever exceeds six, as determined in decision block 129 or 136, an affirmative decision is subsequently made in decision block 109, causing the program to enter decision block 116.

The program then determines if the present value of REFLECTED is greater than OLD REFLECTED, and if this determination is negative, the program goes to decision block 110 and determines if REFLECTED is equal to zero. If this determination is affirmative, it means that the program has found a point very close to the minimum on the reflected power curve of FIG. 2. The program goes to block 111 and causes the stepper motor to advance one more step in the hope of centering the resonant frequency of the cavity in the center of the minimum or negative peak 34 of the reflected power curve of FIG. 2. The program then goes to block 121 and executes an idling routine in which the stepper motor is turned off and a light emitting diode (not shown) indicating that the cavity is precisely tuned is turned on.

If the determination of decision block 116 is that the present value of REFLECTED exceeds the present value of OLD REFLECTED, it means that the reflected power is increasing as a result of the most recent tuning changes. The program then goes to decision block 117 and determines if this is the first time that the reflected power has begun to increase. If this determination is affirmative, the program goes to block 119 and changes the direction of the stepper motor. The program then returns to block 104 and repeats the loop. The next pulse applied to the stepper motor, which will occur in block 105A, will be in the opposite direction toward, but hopefully not past, the center of the negative minimum point of the power reflected curve of FIG. 2.

The determination of decision block 109 will be affirmative on the next pass through the loop, so decision block 116 will be entered again. At some point, the determination of decision block 116 should indicate that REFLECTED is greater than OLD REFLECTED, indicating that the reflected power minimum has been found. The determination of decision block 117 then will be negative, and the program will go to block 119 and jump to a subroutine that changes the direction of the stepper motor. The program then goes to block 120 and causes the stepper motor to advance in the proper direction by two steps. The intent is to step twice at a time while far from the peak ratio to speed up the tuning process and then step one step at a time once the threshold has been crossed and the peak is near. The program then goes from block 120 to 121, turns on the light emitting diode indicating that a valid minimum has been found, and turns the stepper motor off.

The program then goes to block 123 and determines new values of INCIDENT and REFLECTED, and then goes to decision block 124 and determines if the value of INCIDENT is still greater than or equal to six. If this determination is negative, the program goes back to block 101 and waits until the incident power exceeds six. Otherwise, the program goes from decision block 124 to block 126 and computes a new value of RATIO, and then, in decision block 127, tests RATIO to determine if it is greater than or equal to six. If this determination is negative, the program goes to block 101, but if this determination is affirmative, the program returns to block 123 and repeats the above loop.

If the determination of decision block 124 is that RATIO is greater than or equal to six, the program assumes that the cavity is still tuned to the minimum point of the reflected power curve, since the stepper motor has not been advanced in the present loop. The program keeps repeating the loop including blocks 123, 124, 126, and 127 until either the incident power falls below five or the standing wave ratio falls below six.

Referring now to FIG. 6, an "improved" flow chart, which represents the program listed in the printout of Appendix A attached hereto, is shown in which the microprocessor 14 of FIG. 1 is "time shared" among the sixteen cavities that need to be self-tuned. The program of FIG. 6 is entered via label 140, which initializes the system in a conventional manner, setting initial values of various software variables and registers, and turning all stepper motors off. The program then goes to block 141 and sets a software pointer to point to the first channel or cavity. The program then goes to block 142 and initializes a three part tuning sequence effectuated by the program by setting a software variable equal to a value indicating that part 1 of the three part sequence is to be performed next.

Part 1 of the three part sequence involves making a determination of whether the cavity needs to be tuned. If it is already adequately tuned to the transmitter frequency (i.e., if the value of the standing wave ratio exceeds a predetermined threshold), there is no need to turn on the stepper motor for that cavity. Part 2 of the three part tuning sequence involves generating appropriate stepper motor control pulses to orient the tuning element 10 in such a manner as to tune the resonant frequency of the cavity precisely to the transmitter frequency. Part 3 of three part tuning sequence is verifying that the present channel has been properly tuned.

After block 142 (the tuning sequence initialization) has been executed, the program goes to block 143 and clears a flag called the "R-DECREASING" flag. This flag indicates whether the reflected power of the cavity presently being tuned has been observed to decrease as a result of rotation of its stepper motor during the present tuning sequence.

The program then goes to block 144 and executes a subroutine similar to the above-mentioned ADC subroutine and obtains the present values of incident power P.sub.I and reflected power P.sub.R. The program then goes to decision block 145 and determines if the present reading of incident power is great enough that it exceeds a maximum noise level expected of the output of the incident power detector.

If the determination of block 145 is negative, the program simply goes to block 146 and turns off the stepper motor of the present cavity. The program then goes to decision block 147 and determines whether the current channel or cavity number is N, where N is the total number of cavities to be self-tuned by the microprocessor system of FIG. 1, in this case 16. If this determination is negative, the program goes to block 148 and points to the next channel or cavity to be self-tuned and then re-enters the loop at block 142. If the determination of decision block 147 is affirmative, the program re-enters the loop at block 141, and points back to the first channel and begins automatic self-tuning of the entire transmitter combiner again.

If the determination of decision block 145 is affirmative, and the determination for the present channel or cavity being pointed to by the program is that there is transmitter power being received, the program then goes to decision block 149 and determines if it is in part 1 of three part tuning sequence. If this determination is affirmative, the program goes to block 162 and determines if a "peak indicator", indicating that reflected power is at a minimum, has been set. If this determination is negative, it means that the cavity is not yet adequately tuned. The program then goes to block 165 and turns on the stepper motor of the present cavity and advances it a predetermined number "X" (typically 1) of steps, as indicated in block 166.

Next the program goes to block 167 and advances the software variable indicating which part of the three part tuning sequence of block 142 is being performed. In this case, it advances it from part 1, which was just completed, to part 2.

If the determination of decision block 162 was that the peak indicator flag is already set, the program goes to decision block 163 and determines if the standing wave ratio P.sub.I /P.sub.R is less than a predetermined threshold, which is typically 4. If this determination is negative, it means that the cavity is already adequately tuned, and the routine goes to block 146, turns off the stepper motor of the present cavity, and advances to the next cavity to be tuned.

It should be appreciated that it is very important to not attempt to adjust the tuning of a cavity that already is quite adequately tuned. It should also be appreciated that it is necessary to use the ratio of P.sub.I and P.sub.R to make the determination as to whether the cavity is reasonably adequately tuned in order to make this determination independent of the transmitter power level, which may vary.

Returning to block 167, after this step has been completed, the program goes to block 168 and initializes a software timer that runs to limit the amount of time (i.e., the number of passes through block 166) spent for part 2 of the tuning sequence. This amount of "time" has been set to be approximately the number of steps needed to be taken by the stepping motor to tune the combiner all the way across its frequency range and back to where it started at the beginning of the 3-part tuning sequence. The timer in block 168 may be set, as a typical value, at approximately 8960 single steps of the motor. Part 2 of the tuning sequence is the part that actually performs the self-tuning operation, generating a sufficient number of stepper motor pulses to orient the tuning element 10 so as to essentially minimize the reflected power.

The program then goes to block 144 and executes the ADC subroutine to obtain new values of P.sub.I and P.sub.R.

Once the new values of P.sub.I and P.sub.R are obtained in block 144, the program passes through blocks 145 and 149, as before. Since the program is no longer in part 1 of the three part tuning sequence, the determination of block 149 will be negative, and the program will go to block 150 and test the present value of P.sub.R to determine if it is increasing. Decision block 150 really asks "Is the tuning in the wrong direction?" If the answer in decision block 150 is negative, it means that the tuning may be progressing in the desired direction, and the program goes to decision block 151, which asks the opposite question, i.e., "Is tuning in the correct direction?" If this determination is negative, it corresponds to a "can't determine" condition, and the program goes to block 156 and advances the software "timer" that was initialized in block 168. A "can't determine" condition occurs when no change in P.sub.R is detected or when the accumulated change in P.sub.R since either the last "valid" change or the first measurement is less than some maximum noise level expected from the detector and analog-to-digital conversion circuitry. A "valid" change is any accumulated change greater than the maximum noise level expected (represented by the constant "NOISE" in the program).

The program then goes to decision block 157 and determines if enough "time" or motor steps have been taken in this part of the three part tuning sequence. If this determination is negative, the program goes to block 158, advances the stepper motor, gets back into the loop at block 144, and repeats. If the allocated number of motor steps have already been performed in part 2 of the three part tuning sequence, the program goes to block 146, turns off the stepper motor of the cavity being tuned, and moves on to the next cavity, as previously described.

If the determination of decision block 151 is affirmative, this means that present tuning of the cavity is in the correct direction, and the program goes to block 152 and sets the "reflected power decreasing" flag that was cleared in block 143. The program then goes to decision block 153 and determines if the present stage of the three part tuning sequence is part 2. The first time through decision block 153, the determination will be affirmative, and the program will go through block 156 and decision block 157 as described above. If the determination of block 153 is negative, it means that the reflected power has been found to be at a minimum. The program then goes to decision block 154 and determines if the standing wave ratio is less than the threshold (in this case, defined as 4).

In block 154, if the standing wave ratio P.sub.I /P.sub.R is greater than or equal to the predetermined threshold, the determination is negative. This means that the standing wave ratio is at a high level while, in addition, the reflected power is at a minimum. These are the desired conditions indicating that the present cavity is quite adequately and precisely tuned to the transmitter frequency, so the peak indicator flag is set in block 155, the stepper motor of the present cavity is turned off, and the program goes to the next cavity. If the determination of decision block 154 is affirmative, the stepper motor then is turned off, because the program has tuned up the present cavity as well as it can at the moment and therefore will not spend any more microprocessor time on this cavity at this time (i.e., the cavity is "precisely", but not "adequately" tuned).

If the determination of decision block 154 is negative, it means that the cavity is both adequately and precisely tuned. The program then goes to block 155 and sets the peak indicator flag that was tested in block 162 and possibly cleared in block 164, and then goes to block 146 and turns off the stepper motor of the cavity presently being tuned and moves on to the next cavity.

If the determination in decision block 150 is affirmative, it means that tuning is in the wrong direction on the curve of FIG. 7, and that the reflected power minimum 34 in FIG. 2 has been passed. The program therefore goes to block 159 and changes the stepper motor direction. The program then goes to decision block 160 and determines if the program is in part 2 of the three part tuning sequence. If this determination is negative, the program goes to block 146, turns off the stepper motor, and moves on to the next cavity to be tuned. If the determination of decision block 160 is affirmative, the program goes to block 161 and checks the reflected power decreasing flag to see if it is set. On the first pass through decision block 161, this flag is probably set in accordance with prior execution of block 152. If the determination of decision block 161 is affirmative, it means that tuning is in the correct direction, and the program goes to block 166, increments the stepper motor and continues. If this is done, the program then goes to block 167 and advances to the next part of the three part tuning sequence. The next pass then would result in a negative decision in either decision block 160 or 153.

If the determination of decision block 161 is negative, the program goes to block 156, advances the software "timer" for the present part of the three part tuning sequence, and advances the stepper motor and gets back in the loop at block 144 or turns off the stepper motor and advances to the next cavity to be tuned, depending upon whether or not the allocated "time" for this cavity has been spent.

Although the above-described algorithms can result in extremely precise tuning of each of the cavities in a transmitter combiner, as a practical matter it has been found that much less precision in tuning of the cavities is needed than is produced by means of the programs of FIGS. 4A-4C, and 6. In fact, adequate tuning can be achieved with a much simpler tuning technique that always rotates the tuning element 10 in the same direction and does not reverse directions. A minimum in the reflected power, i.e., optimum tuning of the cavity, occurs at successive positions of the tuning element 10 that are 90.degree. apart. That is, once a reflected power minimum is obtained, another identical reflected power minimum will occur if the tuning element 10 is rotated another 90.degree.. Therefore, instead of using the above-described, somewhat complex routines (which cause the stepper motor to rotate past the reflected power minimum such as 34 in FIG. 2 and then "back up" to center the tuning element in the reflected power minimum), adequate and precise tuning can be achieved by causing the stepper motor to continue rotating in the original direction by slightly less than 90.degree..

As a practical matter, it has been found that the voltage V.sub.R, i.e., the voltage produced by the reflected power detector diode, has a characteristic similar to the one shown in FIG. 7. With the range setting resistors of the reflected power diodes adjusted so that when the cavity is badly out of tune, V.sub.R (the voltage produced by the reflected power diode) is about two volts, then, when the tuning element 10 is positioned so that the resonant frequency of the cavity is equal to the incoming transmitter or excitation signal, V.sub.R falls to a level of approximately 30 to 50 millivolts, at a point such as B in FIG. 7. Points A and C designate the values of V.sub.R at which the standing wave ratio P.sub.I /P.sub.R is equal to 4 (regardless of the transmitter power level). The frequency difference between points A and C is only approximately 100 kilohertz. For the particular stepper motors being used, each step produces a change of approximately 10 kilohertz per step of the stepper motor 11.

In the simplified method of operation, microprocessor 14 simply alternately steps the stepper motor ahead and then compares the standing wave ratio to a fixed threshold value of, for example, 4. At points A and C on the reflected power voltage curve of FIG. 7, the cavity is already fairly closely tuned to the transmitter frequency exciting that cavity, as the standing wave ratio is about 4 at those points. Microprocessor 14 then simply advances the stepper motor of the cavity being tuned a fixed number of steps in the same direction. It has been found that only one or two additional steps of the stepper motor comes very close to centering the curve of FIG. 7 so that point B is equal to the transmitter frequency.

The above-described embodiments of the invention have all been constructed and tested in prototype experiments. The last described embodiment of the invention avoids the problems of rather high amounts of system noise generated as a result of the experimental transmitter being used in the experiments, and has been found to result in reasonably fast, reasonably precise, and reasonably continuous tuning of all cavities receiving incident power from their respective transmitters. The self-tuning time of each cavity is very short.

While the invention has been described with reference to a particular embodiment thereof, those skilled in the art will be able to make various modifications to the described embodiments of the invention without departing from the true spirit and scope thereof. It is intended that all self-tuning apparatus and methods which are equivalent to those described in that they perform substantially the same function in substantially the same way to accomplish substantially the same result be encompassed within the invention. Although a digital approach to self-tuning of a cavity has been described herein, it would be possible, and in some instances entirely practical, to use primarily analog circuitry to generate feedback signals which control a servo motor to orient the tuning element 10 so that it minimizes the reflected power. ##SPC1##