CA1271250A - Method for detecting the channel to which an electronic receiver system is tuned - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE
A cable meter for monitoring the channel selected by a converter of a television system which receives signals by means of cable. The cable is connected to the cable meter and the output of the cable meter is applied to the converter. The output of the converter is connected back to the cable meter which, in turn, provides an output to the television. During normal operation, signals received by the cable meter from the cable pass directly to the converter, and a selected channel from the converter passes through the cable meter to the television. To monitor the channel selected, an oscillator generates a substitution signal which is substituted for the television signal applied to the converter. The cable meter then monitors whether the sub-stitution signal passes through the converter, during which period, the cable meter prohibits this substitution signal from reaching the television. By varying the frequency of the substitution signal, a search is performed for the selected channel. Techniques are employed to enhance the accuracy of the monitoring. The present invention has applicability to any communications system in which radio frequency signals are transmitted over any medium, such as televisions with internal tuners which may not even be adapted to receive cable signals or radio receivers.

Description

5~
66082~173D

This application is a division of our Canadian Application Serial No. 451,975 filed April 13, 1984.
The present invention relates to the monitoring of communications receivers, and more particularly, to the monitoring of the channel to which a receiver is tuned.
In the entertainment field, the size of an audience enjoying an event or program is often monitored as an important indicator of popularity or success. This is particularly true with entertainment provided over electronic communications systems such as television and radio. The audience size is employed not only to determine the popularity of a particular program or showl but also to assist in making programming decisions. Furthermore, advertising rates are based upon audience size.
Determining the size of an electronic communications system audience is particularly difEicult due to the dispersed nature of the audience. Heretofore, telephonic surveys have been conducted to determine the number of individuals watching particular radio or television programs. However, such surveys are highly labor intensive. Furthermorel the necessity of calling thousands of households makes such surveys time consuming.
To overcome problems associated with telephonic surveys, electronic monitoring techniques have been developed.
Thus, United States Patents 4,058,829 to Thompson and 4,044,376 to Porter teach television monitoring devices. According to these patents, a signal is injected into the radio frequency input of the television at a frequency corresponding to the ~ ~ .,, ~ ~ f 1~ ~

carrier frequency of a particular channel. A probe attached to some point within the video circuits of the television determines whether the injected signal has passed through the tuner. I~
the injected signal has not passed through the tuner, then the frequency of the injected signal is changed to the carrier fre-quency of another channel and the determination is repeated.
This process continues until a frequency is selected which enables the injected signal to - la -- ~ ~7~5~) pass through the tuner. The channel to which the television is tuned is then known.
See also United States Patents 4,216,497 to Ishman et al and

2,630,367 to Rahmel which teach television monitoring systems.
Electronic channel detectors have also been developed which are particularly suited for cable television systems. Examples of such detectors are disclosed in United States Patents 4,048,562 to Haselwood et al, 3,769,579 to l-larney, 3,230,302 to Bruck et al and 3,987,397 to Belcher et al.
The present invention accurately detects the channel of a communications system medium which has been selected by a receiver.
The preferred embodiment of the present invention ~hereinafter referred to as a "cable meter") is employed in a cable television system.
In such a system, a cable carrying the television signals is connected directly to a multifrequency input of the cable meter. A multifrequency output of the cable meter is connected to a conventional cable converter. The output of the converter is connected to a signal frequency input of the cable meter. A
signal is provided from a single frequency output of the cable meter to a tele-vision. Thus, when used in a communication systems having a separate channel selector, the present invention may be connected to the system in a noninvasive manner.
During normal operation, cable signals pass through the multi-frequency terminals of the cable meter to the converter which selects the desired channel. The signals from the selected channel pass back tilrough the cable meter and are applied to the television. To determlne the channel selected, the cable meter generates a signal at a frequency related to the carrier frequency of one of the channels on the cable. This signal is sub-stituted at the converter input for the signals on the cable and the output of the converter is monitored by a single channel receiver to determine wllether ~27~

the substitution signal passes through the converter. If the substitution sig-nal does not pass through the converter, then the cable meter substitutes another signal related to the carrier frequency of a different channel, and the output of the converter is monitored. In the preferred embodimcnt, the fre-quency range over which searching occurs can be adjusted, so as to avoid search-ing unnecessary channels. Also in the preferred embodiment, searching begins with the high frequency and progresses to successively decreasing frequencies.
The search procedure continues until a substitution signal passes through the converter, indicating that the converter is set to select the channel having a carrier frequency related to the frequency of the sub-stitution signal. In this manner, the cable meter uses a signal substitution/
response measurement technique in some ways analogous to that employed by ~he Porter and Thompson patents, ~E~. However, since the output of the converter is applied to the cable meter, instead of being connected directly to the television, the cable meter is able to block the substitution signals from being applied to the television.
The power cord of the television may be plugged into the cable meter, so that the cable meter can monitor when the television is on. Data collected by the cable meter may then be sent to a household collector which receives data from other cable meters as well.
In the preferred embodiment of the present invention, the iden-tification of a selected channel during a first searching operation causes only a preliminary indication of the selected channel to be generated. The searching operation is performed again, and after two searching operations produce the same results, the indication of the selected channel is verified. To recluce the possibility of errors induced by the generation of sub-multiple fre~uency com-ponents Wit]l the substitution signals, the strength of the substitution signals applied to the converter may be reduced during the second search operation.

~t ~

In fact~ if the second search at the reduced level either fails to identify a selected channel or identifies a channel different from the channel identified during the first search, the searching operation is repeat-ed for a third time. In the former situation, the third search is conducted at a high level and if the same channel as in the first search is identified, the indication of the selected channel is verified. In the latter situation, the third search is conducted at a low level and if the same.channel is iden-tified in the second and third searches, the indication as to the channel identified during the second and third searches is veri-Eied. Once a channel indication has been verified, the program shifts into a shorter, more circum-scribed, operating sequence to monitor the verified channel, until the channel is changcd. l~hen the chalmel selected by the converter is changed, an indication of the change is generated only after the cable meter fails to con-firm that the selected channel remains the same in a predetermined plurality of consecutive attempts.
The intervals between transmissions to the household collector, in the preferred embodiment of the present invention, may be varied. In this manner, the probability of simultaneous transmissions from different monitors to the same household collector is reduced. Also, the timing of the sub-stitution signal with respect to the television signals on the selected channel is controlled so that the substitution signals are applied to the converter either during the blanking portion of the television signal or during the top few lines of the video portion of the signal. In this manner, interruption of the television picture is minimizcd. IJI fact, thc preferred embodiment enables the timing to be varied so as to avoid substitution during portions of thc television sigJlal which might be used locally for other purposes.
In addition tc receiving signals from cables, the prescnt inven-tion also includes au~iliary inputs wllicll may be selected by means of a switch.

~.~7~ V

Such inputs would be for video games, computers, video recorders or the like.
~hen one of tlle auxiliary inputs has been selected, the signal from the auxiliary source passes through the present invention and is applied to the television. During this period, the present invention generates a signal to the household collector indicating that an auxiliary input has been selected.
To maximize the efficiency of the present invention, the system must be tuned as well as possible to the selected channel. In the preferred embodiment, if the television signal is not being adequately received, the signal is attenuated to degrade the picture quality and force the viewer to attempt to better tune in the channel.
As a result of the present invention as described above, all connections to a receiver system employing the present invention may be made directly to the present invention. In receiver systems employing a separate channel selector, the present invention may be added with no connections inter-nal to any of the components. The inclusion of a single channel receiver with-in the meter avoids the necessity of making connections internal to the cable converter and television.
The present invention ensures accurate monitoring as a result of a number of features. The repetition of the search operation reduces the possibility of erroneously identifying a non-selected channel. Starting each search at a high substitution signal frequency reduces the possibility of error caused by substitution signal harmonics and repeating the search at a reduced substitution signal level reduces the possibility of error caused by sub-multiple components of the substitution signals. In fact, the particular pat-tern of high and low level substitution signals during consecutive searches is intended to maximize the probability of correctly idelltifying a selected channel. The possibility of erroneo-lsly reporting a change in channel selection 7~ ~t~
6~082-173D

is reduced in the present invention in that an indication that the selected channel has been changed is not generated until the present invention unsuccessfully monitors for the selected channel over a plurality of consecutive attempts.
The cable meter of the present invention is micro-processor based and employs a frequency synthesized oscillator which is under microprocessor control. As a result, whereas in the past a separate oscillator and discrete components were required for each channel to be searched, in the present inven-tion the desired substitution frequencies are generated by the frequency synthesized oscillator in response to control signals supplied by the microprocessor. This fea~ure greatly simplifies design and expense in construction as well as substantially expanding the capabilities of the meter.
The present inven~ion has application beyond cable television systems with detached cable converters. In fact, certain aspects of the present invention can be employed with any radio frequency communications receiver system which employs a channel selector, such as radio and television (including - 20 television with an internal tuner). Throughout this application, including the claims, the term "channel" will mean a signal carrying data, differentiable in some manner from other signals carrying data.
Thus, in accordance with a broad aspect of the inven-tion, there is provided a method of detecting which of a plurality of carriers has been selected for reception by a television system, said system i.ncluding a selector for selecting one of said carriers, said method comprising the steps of: counting the number oE horizontal sync pulses between consecutive vertical sync pulses in television signals from said selector; determining from said counting step when said television signals are accept-able; and only after said determination is positive, detecting which of said carriers has been selected by said selector.
The above and other objects and advantages of the present invention will become more apparent and more readily appreciated from the following detailed description of the pre-sently preferred exemplary embodiment of the present inventionr taken in conjunction with the accompanying drawings, of which:
Figure 1 is a schematic drawing of a conventional cable television system;
Figure 2 is a schematic diagram of the connection of the cable - 6a -meter of the presently preferre~ embodiment of the present invention to a television and cable converter;
Figure 3 is a block diagram of the cable meter;
Figure 4 is a block diagram of the frequency synthesized oscillator 154 of Figure 3;
Figure 5 is a block diagram of the control logic of the cable meter;
Figure 6 is a general flow chart of the channel detecting pro-gram of the present i`nvention;
Figures 7-13 represent a detailed flow chart of the monitoring program of the present invention;
J Figures 14 and 15 represent a detailed flow chart of the T-counter interrupt subroutine of the present invention in which data is trans-mitted from a cable meter to a central collector; and Figure 16 represents a flow chart of the input selection sub-routine of the present invention.
The presently preferred embodiment of the invention is described hereinafter for use with a cable television receiver employing a typical, separate cable converter. However, certain aspects of the present invention have applicability to any radio frequency communication system employing a channel selector, wllether the communication system be television, radio or the like.
Cable television systems are becoming more popular, and therefore more significant with respect to audience monitoring. Figure 1 illustrates tlle typical arrangement of a cable television system. In Figure 1, cables 100 and 102 are applied to cable conver-ter 104. Each of cables 100 and 102 carries ~i~7~

65 channels in tlle present embodiment. The output of cable converter 104 is applied to television 106. Cable conver-ter 104 may be in a separate housing wllich sits atop television 106. Cable converter lQ4 selects one of the 124 channels carried over cables 100 and 102 and adjusts the carrier frequency of the selected channel to a predetermined frequency, typically corresponding to the carrier frequency of channel 2, 3 or 4 on television. Cable converter 104 is, therefore, said to have a fixed, or single channel output. Tllus, television 106 remains set on channel 2, 3 or 4, as specified by the cable TV company, and channel selectlon is done at cable converter 104 by tuning to a particular carrier frequency on one of cables 100 and 102.
In Figure 2, the preferred embodiment of the present invention, referred to herein as "cable meter" 108 is connected to a medium s~ich as cables 100 and 102 through multifrequency or multicllannel inputs. ~lultifrequency, multichannel outputs of cable meter lOS are connected to a channel selector such as cable converter 104 via lines 110 and 112, respectively. The output of converter 104 is applied to a single frequency or channel input of cable meter 108 via line 114. A single frequency or channel output of cable me~er 108 is connected to a radio frequency (r.f.) communicati.ons system receiver SUC}l as television 106 via line 116.
Power cord 118 of television 106 is connected to cable meter 108 so that cable meter 108 can monitor when television 106 is energized.
Power is applied to cable meter lOS by means of po~er cord 120.
Data collected by cable meter 108 is outputted to a household collector over line 122.
Cable meter lOS also provides for the input of allxiliary video signals through its auxiliary 1 alld auxiliary 2 inputs. These inputs ellable the television receiver system to be utilized with a video cassette recorder, vidco disc, personal computer, video games, etc. Signals applied to auxiliary inputs 1 and 2, l~hen selected by cable meter 108, pass directly to television 106 over line 116.
Figure 3 provides additional details conceTning the components of cable meter 108. Signals on cables 100 and 102 are applied to switches 130 and 132. In the preferred embodiment, these switches are electronic and are actuated by control signals. As illustrated in Figure 3, switches 130 and 132 are norm-ally closed so that sïgnals on cables lQ0 and 102 pass over lines 110 and 112 to cable converter 104.
The signal from converter 104 at the fixed carrier frequency is applied to cable meter 108 where it passes through amplifier 134, bandpass filter 136 and splitter 138. Bandpass filter 136 narrows the frequency range of the output signal from cable converter 104 to prevent channel misidentification. A portion of the signal from splitter 138 passes through normally closed switch 140 and is applied to television 106. Switch 140 is, in the preferred embodiment, an electronic switch which responds to a control signal. Switch 140 is different from switches 13Q and 132 in that it ~las three positions. Switch 140 can either be open, be closed, or cause signals to be attenuated (i.e., reduced in strength) such as, for example, by passing them through attenuator 142 before applying them to television 106.
The other portion of the signal from splitter 138 is applied to single channel receiver 144 whic]~ generates a vertical oscillator signal, a horizontal oscillator signal and a sampling signal. These signals are applied to control logic 146 as illustrated. The vertical oscillator signal is generated from a local oscillator witllin single channel receiver 144 and is characterized by pulses synchronized with thc vertical sync pulses in the video signal. The horizontal oscillator signal is also generated from the local oscillator witllin _ ~ _ receivcr 14~ and is synchroni~ed with the horizontal sync pulscs in the vidco data. The sampling signal is digital having either a high "positive" value or a low "negative" value as will be described later.
The po~er being drawn by televisioll 106 is monitored througll transformer 148 and current sensing circuitry 150. The threshold at which a TV power on signal is sent to control logic 1~6 is determined by thresllold setting switches 152. The threshold switches are necessary to prevent a false indication of "power on" from so-called "instant on" televisions which always draw some current wllenever they are plugged in.
Freq-lcllcy synthesized oscillator 154 generates a frequency substitution signal related to a control signal provided by control logic 146. As shown in Figure 4, frequency syntllesized oscillator 15~ includes two voltages controlled oscillators 50 and 52 and a reference oscillator 54. The oscillators 50 and 52 are adjusted by control signals from compare circuitry 60 (later described) in oppositc directions and their respective outputs are input to mi~er 56 which generates a difference frequency. A difference frequency input is provided to frequency divider circuitry 58. Microprocessor 170 ~later described in more detail) loads a divisor into frequency divider 58 which is representative of the frequency ne.Yt to be substituted into cable converter 10'~. The difference fre-quency is divided do-Yn in frequency divider circuit 58 according to the divisor supplied by microprocessor 170 and then compared at compare circuit 60 with the reference oscillator frequency. The compare circuit 60 provides outputs to os-cillators 50 52 to continually adjust the oscillators 50, 52 until the divided do~Yn difference frequency is equal to the reference frequency. Once the compare circuit 60 inputs are equal compare circuit 60 indicates a "loc~" condition to microprocessor 170 and microprocessor 170 will then substitute the difference frequcncy from ]ni~er 56 into cable converter 104 at the appropriate time accord-ing to the microprocessor program sequence later described.

~;~ 7~lL~ rj~ `

The output of oscillator 154 is applied to attenuator 156 as shown in ~igure 3. Control logic 14G has an input to attenuator 156 to control wlletller or not the signal generated by oscillator 154 is attenuated.
The frequency substitution signal generated by oscillator 154 passes through attenuator 156 and splitter 162 to switches 15~ and 160.
These switches are similar to switclles 130 and 132, and are controlled by control logic 146. During normal operation, switches 158 and 160 are opened.
~lowever, during a channel detecting operation, switches 158 and 160 are closed, while switches 130, 132 and 14a are opened. As a result, tlle fre-quency substitution signal fronl oscillator 154 is applied to converter 104 over lines 110 and 112.
Switch 164 causes control logic 146 to select signals from cables 100 and 102, the auxiliary 1 input or the auxiliary 2 input. If the auxiliary 1 input is selected, control logic 146 closes switch 165 so that signals from the auxiliary 1 input terminal pass to television 106. At the same time, a signal is provided by control logic 146 ~hich causes switch 140 to open. Similarly, if the auxiliary 2 input is selected, control logic 146 closes switch 166 so that signals at the auxiliary 2 input pass to television 106. Installer s~itches 168 may be employed to set a number of parametersof cable meter 108, Control logic 146, together with associated components, is illustrated in Figure 5. The heart of control logic 146 is microcomputer 170, which, in the preferred embodiment, is a model 804911 microcomputer manufactured by Intel Corporation. Microcomputer 170 receives inputs from a number of sources. Thus, multiplexer 172 receives the vertical oscillatur and horizon-tal oscillator signals and the sampling si~nal from single channel receiver ~.2~5() 1~l4, and a frequency lock sigllal from frcquellcy synthesized oscillator 154.
~lultiple:~er 172 zl)plies ~hese sig1lclls to microcolllputer 170.
~ lultiplexer 174 receives signals from installcr switcllcs 168.
Installer switches 168 are used for several importa1lt pur1~oses. They are used to determine the code by wllicll cach particular cable meter 10~ identifies itself to a household collector. Also, installer switches 168 set the interval between data transmissions from cable meter 108 to the housellold collector.
The transmission intervals are set to vary abouta 2second transmission interval.
Each cable meter 108 connectecl to a common house]lold collector has a transmission interval of slightly different lengtll to minimize the number of times that data from different meters simultaneously arrives at the household collector. The simultaneous arrival of data from different meters would result in data des-truction. Also, installer switclles 16S de-termine the portion of the vertical blanking interval in a television signal where the frequency substitution signal from frequency synthesized oscillator 154 is substituted. Depending on the locality, certain portions of the blanking interval may be unavailable because they are reserved for television test signals, closed czption or teletext, for example. Consequelltly, the installer switches can be set to progressively move the position at which the frequency substitution signal is substituted within the vertical blanking interval or the top few lines of the television picture. Other installer switches 168 are used to limit the frequency range of the search to be made for a selected channel where certain cable channels are not to be logged. Installer sl,litches 168 also determine the highest cable channel frequency at which channel searching begins. In the preferred embodi-ment, cha1mels may be searched starting at 300 ~II-Iz, or starting at 450 ~-lz.
Incidentally, thresllold setting switches 152 in Figure 3 are also set by the installer.

The output of multiple~cr 174, TV current sensing circuitry 150 and R0~l 176 are all applied to data bus 178 wllicll is input to microcomputer 170.
ROM 176 is addrcssed by a signal from microcomputcr 170 and is customized for a particular locality to allo~ for special adjustments to frequency sclection and frequency decrementation of oscillator 154.
~ licrocomputer 170 also receives signals from switcll 164 to indicate whetller signals have been selected to be received from cables 100 and 102, auxiliary 1 input or auxiliary 2 input. The signalsfrom switch 164 also controi switch drivers 184 and 185 which operate switches 165 and 166.
Finally, microcomputer 170 receivcs a reset signal from reset circuit 192. When power to cable meter 108 is being received, cable meter lOS period-ically produces transmissions to a household data collector. Reset circuit lCJ2 monitors the transmission of data to the household collector (by monitoring transmit enable line 190). If transmissions should stop, nnicrocomputer 170 has become hung up in a loop of its program. If a transmission does not occur within a predetermined period of time, reset circuit 192 causes microcomputer 170 to be reset.
Microcomputer 170 controls a number of elements of cable meter 108.
Thus, microcomputer 170 sends a signal to frequency synthesizer oscillator 154 to select the frequency generated by oscillator 154 and a signal to attenuator 156 to control whether a full strength or reduceci strength signal will be substituted. Also, microcomputer 170 controls switch drivers 180 through 183 ~lich control switch 158, switch 160, switclles 130 and 132, and switch 140, respectively. Note that switches 130 and 132 are always in the same sta~e, and thus can be controlled by the same signal. As indicated above, switch 140 is a three position switch. Therefore, two separate signals must be applied to driver 183. One signal may be considered an on/off signal and the other signal may be considered a reduced level signal. The 3L;~ 7~L~. rj~3 reduced lcvel signal causes the signal to be attenuated such as by connccting switc]l 140 to the attcnuator 142 sllol~l in Figure 3.
In addition to controlling multiple~crs 172 and 174, microcomputcr 170 also gencrates signals wllicll are applied through current loop driver lS6 to a household collector. Instead of providing a ~ire between each cable meter 108 in a household and a common household collector, it is possible to emp]oy the AC power lines to transmit data signals betweell each cablc meter 108 and a common household collector. ~ccordingly, microcomputer 170 generates data for the household collector on line 188 and a transmit enable signal on line 190 which may be applied to an optional AC carrier current transmitter W]liC]
transmits the data on the AC power lines.
The channel monitoring operation of cable meter 108 as illustrated in Figures 3-5 will no~Y be described with respect to the simplified flow chart of the operation of microcomputer 170 in Figure 6. The channel monitoring program of micrOcomputer 170 begins when power is applied to microcomputer 170 or when microcomputer 170 is reset at step 194. The program then performs a number of initializations at step 196. Director 198 is the top of the main loop of the program, as will become apparent from the following discussion.
At step 200, microcomputer 170 performs various status testing steps to be certain that the hardware is performing properly. If any problems are detected, the transmission status is set to an appropriate code to identify the problem and a corresponding s-tatus is sent to the household collector. The program then returns to director 198. At step 201 the program goes through a procedure to ensure that the TV signal is being adequately received. If it is not, the program returns to director 198.
Assumillg that the TV signal is being received adequa-tcly the program moves to step 202 wllere the value of a variable stored in a register called "hit value" is considered. An e~planation of this variable will be provided - 14 _ ~7~

hereinafter. Since "hit value" has been reset to ZCTO at initialization step 196, the program passes to step 204 W]liC]l causes a pointer to indicatc an address at the start of a tablc. Thc table contains indicatiolls rclated to the particular frequencies of tlle substitutioll signals and, thcrefore, of the carriers of the channels to be monitored by cable meter 108. At step 206, microcomputer 170 causes an indication of the frequency of the ne~t substitution signal to bc generated by oscillator 154 to be retrieved from the -table. At step 208, it is determined whetller the table has been completely scanned.
During the first pass through the program, the pointer will not be at the end of the table so that the program passes to step 210. Up until step 20~, assuming that switch 164 in Figure 3 is set to cablc, signals from cables 101 and 102 have been passing through closed switches 130 and 132 to convcrter 104. The signal from converter 104 has been passing through amplifier 134, filtcr 136, splitter 138 and switcll 140 to television 106. At step 210 of the program a number of changes are made with respect to the switclles.
Switches 130, 132 and 140 are momentarily opened and switches 158 and 160 are momentarily closed. This causes the frequency substitution signal gener-ated by oscillator 154 under the control of microcomputer 170 to be applied to converter 104. The signal from converter 104 is monitored by receiver 144, and if converter 104 has been set to the channel having a carrier frequency very nearly the same as the frequency of the signal generated by oscillator 154, receiver 144 generates a smapling signal by whicll microcomputer 170 deter-mines that the channel selected by converter 104 has been determined, or, in other words, a "hit" has been made. If microcomputer 170 does not receive a sampling sigl~al, microcomputer 170 continues the search.
~he table of frcquellcies in microcomputer 170 is organized so that ~ rj~

indications of the higllest frequencies occur at the bcginning of the table and sequentially decrease througll the tablc. If oscillator 154 first generatcs a substitution signal havillg a frequency co~respondillg to thc hig]lest carricr frequency channel, any second harmonic COmpOJIent (twice the fundamental fre-quency) generated with the substitution signal will not cause channel misiclenti-fication. For example, if oscillator 15~ is generatillg a substitution fre-quency of 108 ~IHz and a 216 ~a~z second harmonic component, and if cable con-verter 10~ is set at 216 ~lz, a positive sampling signal may be generated, indicating incorrectly that tile channel selected by converter 10~ is lOS ~lz.
To avoid this problem, searching is begun at tlle highest frequency and pro-gressively stepped downwardly channel by channel. Since tlle 216 ~IIIz substitu-tion signal will be generated before the 108 ~lz substitution sigllal, no misidentificatioll can occur.
~licrocomputer 170 uses the vcrtical oscillator antlllorizontal os-cillator signals to determine tile precise portion of the televisioll signal for which the frequency substitution signal is to be substituted, i.e., the precise moment with respect to the television signal at wllicll switclles 15S and 160 are to be closed and switches 130, 132 and 1~10 are to be opened.
Assuming that a hit is not made at the first frequency, the program returns to step 206 to get the next frequency indication from the frequency table in microcomputer 170. This searching process continues until a hit is made. If the program goes through the entire frequency table without making a hit, step 208 will cause the program to progress to step 212 wllere the trans-mission status will be set to an appropriate code indicatillg that no channel has been identified and the program returns to director 198.
If a llit is detected at step 210, the next tas~ is to determinc ~7~

whether the selected channel is on cablc 100 or 102. Once this is resolved at step 214, the transmission status is updated at step 215 to prelilninarily indicate that the chanllel to whicll the converter is tuned has been idcntified.
At step 216 microcomputer 170 checks wllether the latest hit is the second consecutive hit at the same frequency. To reduce the possibility of erroneous reporting, cable meter 108 must again determine that converter 104 is se-t to the same channel in order to verify that the channel has been found. If the latest hit in step 216 is only the first hit, the program returns to director 198.
1~ If the latest hit was, in fact, the second consecutive hit at the same frequency, the hit value is set to four at step 218. The program then returns to director lC98.
After the second consecutive hit, the next pass through the main loop of the program will reach step 202. Since the hit value is not equal to zero, the program executes step 220 at which the hit value is decremented to three. At step 222, a substitution signal having a frequency the same as the frequency of the substitution signal whic}l caused the last hit is substi-tuted for the television signal received from cable 100 or 102. Step 222 determines whether converter 104 remains set to select the same channel. Thus, in response to the substitution signal, microcomputer 170 determines whether single channel receiver 144 generates a sampling signal. If microcomputer 170 does receive a sampling signal, the channel preliminary identified is verified at step 223 and now the computer will move through a shortened program loop as will be explained later on. The hit value is reset to four at step 224, and the program thell returns to director 198.
The program continues to cycle througll the shortened loop including steps 200, 201, 202, 220, 222, 223, 224 and 198 as long as converter 104 con-~ 7~

tinues to select the same chanllel as was identified during the last search.
Eventually, a different chamlel \~ill be selec-ted so that a sampling signal is not generated in response to the frcquency substitutioll signal translllittcd to converter 104 in step 222. As a result, the program progresses to step 226 at which it is determined whetller the hit valuc is zero. If the hit value is not zero, the program returns to director 198. Since stcp 220 decrements the hit value by one with each pass, the program ~ill cycle four t~nes througll steps 222 and 226 as long as a sampling signal is not generated. ~:Eter thc foùrth pass in which no sampling signal is generated, it is determined in step 226 that tlle hit value is ~ero. This causes microcomputer 170 to e~ecute step 212 where the transmission status will be challged to indicate tllat the channel has been lost. The program thell returns to director 198, and steps 204-216 perform another searching operation.
The requirement of four passes before the status is changed reduces the generation of erroneous data. The frequency substitution signal may occasionally be lost between cable meter 108 and converter 104. Before cable meter 108 does anything to change its status in response to such a loss, the frequency substitution signal must not be recovered on four consecutive attempts. It has been determined that i.f a sampling signal is not generated on any of four consecutive attempts, then it is assumed that tlle channel selected by cable converter 104 has been changed by the viel~er.
Figures 7 througll 1~ illustrate in more detail the channel detecting program executed by microcomputer 170, including important features of the present invention whicll are not included in the simplified flow chart illus-trated in Figure 6.
Turning now to Figure 7, tlle program starts witll a reset. This reset ~ 7~

can occur when power is turned on as indicated at step 230. Alternately, reset can occur as caused by eitller hardware~ as indicated in step 232, or software~ as indicated in step 231. Reset circuit 192 in Figure 4 causes the hardware reset while a variable stored in a register called "activity counter"
causes the software reset. The operation of the activity counter register will be described in greater detail with respect to the T-counter interrupt subroutine illustrated in Figures 14 and 15. Essentially, llowever, each time data is transmitted, the activity counter is incremented and each time the main loop of the program is executed, the activity counter is reset. If the activity counter reaches a predetermined level, it means that the main loop of the program is not being executed so that the interrupt subroutine issues a command to reset microcomputer 170.
After the reset, the program is initialized in steps 233 and 234.
Thus, in step 233, the external interrupt is disabled to prevent the program from being interrupted through the external interrupt pin of microcomputer 170.
In step 234, the R~ within microcomputer 170 is cieared to reset the sample count, consecutive good pass count, and good pass count (all of which will be later described) to zero. The substitution signal from frequency synthesized oscillator 154 generates a random frequency when the system first starts up.
The register "hit value"is set to zero. Also, microcomputer 170 deactivates attenuator 156 so that signals generated by oscillator 154 are not attenuated.
Thus, the signal applied to converter 104 from oscillator 154 will initially be at a high level. The meter address is read from lnstaller switches 168.
This is a code by whic]l the particular cable meter 108 will identify itself to a household collector. Finally, a counter called "overflow" is set to a predetermined number to fix the interval between data transmissions as will be described in more detail below Witll respect to ~igures 14 and 15. Tllis ~271;~

particular number is also set with installer switches 168. Attaining a certain value in tlle overflow counter register causes the e~ecution of the interrupt subroutine illustrated in Figures 14 and 15.
In step 235, microprocessor 170 calls the input selection subroutine which is illustrated in Figure 16 and ~hich will be described in more detail hereinafter. Generally, this subroutine determines whether cable meter 108 is set to receive cable signals or signals from the au~iliary 1 input or auxiliary 2 input.
Step 236 represents the top of the loop for all search cycles as will become apparent from the followillg description. This step is entitled "director".
The activity counter register is set to zero at step 237. As indi-cated briefly above, this counter is incremented whenever a transmission is made to a household collector. It is set to zero every tilDe a pass is made through the main loop of the program. If the activity counter counts up too high (to 32 in the preferred embodiment) before being reset, then the system is alerted to -the fact that the program is "hung up" on a particular routine, so that the software is reset from step 231. Additional details of this aspect of the invention are described wi.th respect to Figures 14 and 15 hereinbelow.
In step 238, the program inquires whetller the data in the transmit status register is the same as the last identified channel as stored in the register "new status". If it is not, then the transmit status register is set to the new channel status register data in step 239 and also the transmit word is set to the first word of the transmission.
In step 2~0, in Figure 8, the T-counter is enabled and incrementing of the counter is started. ~s will be e~plained below witil respect to Figures 1~ and 15, the T-counter, together with the overflol~ counter, -time the data transmission intervals. Note that once the T-counter is started on the first pass, it does not need to be restarted on each pass througll step 240. Instead, , . ,. .~

step 240 ensures that it continues to run while the program is running.
At the next step 241, microcomputer 170 determines whether television 106 is on. This is accomplished through the television current sensing cir-cuitry lS0 which generates a signal onto data bus 178. If television 106 is not on, the program moves to step 242 in which the transmission status is set to indicate that the television is off and the TV on and signal present LEDs are turned off. After step 242! the program returns to director 236 in Figure 7.
If microcomputer 170 determines that television 106 is on in step 241, microcomputer 170 moves to step 243 and turns on the TV on LED and clears the carry bit which will later be described witll respect to the input selection routine of Figure 16. In step 244, microprocessor 170 determines whether switch 164 in Figure 3 is set to select signals coming from cables 100 and 102 by accessing the input selection routine of Figure 16 (later described). If the cables are not selected, indicating that either the auxiliary 1 or auxiliary 2 input has been selected, the program returns to director 236. If it is determined in step 244 that cables 100 and 102 have been selected, microcomputer 170, in steps 245 and 247 determines whether the vertical and horizontal oscil-lator signals generated by receiver 144 are acceptable and related to a poss-ible television signal. These oscillator signals will be employed by micro-computer 170 to determine the proper substitution point for the frequency substi-tution signals. If either of these signals are not acceptable, step 246 of the program sets the transmission status to so indicate and the program returns to director 236.
If both of the signals are wor~ing properlyJ microcomputer 170 calls the TV signal good subroutine in step 24~. The l`V signal good subroutine is sl~own in Figure 9, and once intitiated in step 600, moves to step 604 where the _ 21 -horizontal line cOu]lt is cleared to zero, and microcomputer 170 finds tlle next high to low transitioll of thc vertical oscillator signal. l~'hen the transition is found, the program moves to step 606 wllere microcomputer 170 senses the first high to low transition of the horizontal oscillator signal.
This transition should represent the first horizontal line of the TV picture.
0llce this first transition is found, the horizontal line count is incremented to 1 in step 60S. In step 610, microcomputer 170 determines whether the next high to low transition of the vertical oscillator signal has arrived. There are approximately 262 horizontal lines in a TV picture between vertical oscil-lator higll to low transitions. Therefore, the program will return to step 606 from step 610 on this first pass. Steps 606-610 are repeated, incrementing the line count on each pass, until the next vertical oscillator transition is sensed at step 610. At step 612, if the horizontal line count is greater than 26S, the program moves to step 614 where a bad TV signal indicatioil is generated.
If the count is less than 26~, step 61~ determines whetller the line coun-t is less than 260. If yes, again a bad TV signal indication is generated at step 612. If no, then the line COUIIt is between 26S and 260 and this is considered to be a good TV signal. A good ~V signal indication is generated at step 620 and the program returns through step 616 to step 249 of the main program, Figure 10. Assuming, first, that a good TV signal indication is present at step 249, the consecutive good pass counter is incremented from 0 to 1 in step 250. At step 251, the inquiry of whether the signal present LED is on is an-swered no, and the good passs counter is incremented from 0 to 1 in step 252.
The sample count is incremented from 0 to 1 in step 253. Step 254 inquires whetller thc sample count is ~. The answer is no and step 255 inquires wlletller the count of the collsecutive good pass coun-tcr is five. The answer is no, so the program returns to director step 236. From step 236, the program again cycles through the steps leading up to step 249 and assuming a good signal indication, the program moves through the steps 250-255 again incrcmenting the three counters. This cycle repeats itself until the count of the consecutive good pass counter is 5. '~len step 256 decrements the count to 4, the inquiry of step 257 as to whether the signal present LED is on is answered no, and tlle program returns to directo 236. The cycle repeats itself until the sample COwlt is 8, at which time step 25S inquires whether the good pass count is less than 6. If we assume no ~i.e., that at least 6 of the 8 samples were good), the signal present LED is turned on in step 259 (the TV signal is set to full strength - no attenuation) and the good pass counter and sample counter are cleared to zero in step 260. At step 255 the inquiry is whether the consecu-tive good pass counter is 5. We will assume that it was incremented from 4 to 5 on the last pass through step 250 so that the answer is yes. At step 256, the counter is decremented back to 4. At step 257, the inquiry of whether the signal present LED is on is answered yes ~since it was turned on at step 259). I~ence, before passing this point in the program 6 of the last 8 samples, and t]-e last 5 consecutive passes must ]lave resulted in good signal indications from the TV signal good subroutine of Figure 7B.
If the answer at step 258 is yes (less than 6 of last 8 samples were good), the signal present LED is turned off and the microcomputer connects switch 140 to attenuator 142 to degrade the signal to cause the viewer to attempt to tune it in better. Step 263 inquires whether the good pass count is 0. If no, the program returns to tlle director 236 through steps 260 and 255. If yes, step 264 undates the transmission status to indicate that no signal is being received before returning to director 236 through steps 260 and 255.
If, at any time, a bad TV signal indication is detected at step 249, the consecutive good pass counter is set to ~ero at step 2~5 and a 1/2 second delay is introduced at step 267 before the program moves to step 253 The ~L~7~ 3 1/2 second delay allows the horizontal line counting circuitry time to self-correct after a bad signal indicatioll.
To summarize the foregoing, this portioll of the TV signal testing program, comprised of steps 2~S-267 sets the followiilg critcria:
~1) 6 of the last ~ samples and the last 5 consecutive samples must be good before the program can progress beyond this point to tlle channel searching portion of the program (later clescribecl);
(2) 6 of the last 8 samples must be good or the signal to the viewers TV set will be attenuated to attempt to force the viewcr to tune in the signal better; and

(3) at least 1 of the last 8 samples must be good or an indication that no signal is being received will be generated.
Once the program passes step 257, it moves to step 272, ~igure 11, in ~ ich tlle hit value register is examined. Since this is the first run through the program, the hit value will be zero, since it was set to zero at step 234. Accorclingly, microcomputer 170 will next execute step 274.
At step 274, the group count is set to one, in that the program initially asswnes that the first group of frequencies to be retrieved from the table of frequencies will include one frequency. In the presently preferred embodiment, the number of frequencies within each group can vary between one and thirty-one frequencies. Indications of frequencies stored in the table of frequencies are retrieved in groups to minimize the amount of memory re-quired to store the frequenc)~ table. For example, if we asswne that a group of frequcncies consist of ten frequencies, only the highest frequency in the group ancl the group count of ten must be s-tored. Once the highest frequency has been substitutecl, the frequency indication is then decremented by a fixed so value, for e~ample 6 k~lz, and the group count is reduced to nine. At each successive substitution the frequency is again decremented. IYllell the group count reaches zero, the program returns to the tab]e to get the ne~t higher frequency and the next group count. This procedure is e~plained in more detail below. Returning to step 274, a substitution flag is also set, indica-ting that a mode of operation is being entered in WlliCh signal substitutions can be made. ~inally, a pointer is set to indicate the start of the inter-preter table. At step 276, a frequency table interpreter looks to the pointer to determine which group of frequencies are to be selected ne~t. Since this is the first pass through the program, the first group of frequencies will be selected and we will assume that this group will consist of ten frequencies.
At step 278, microcomputer 170 monitors for the end of the frequency table.
Since this is the first pass, the answer will be negative so that the program advances to step 280.
At step 280, frequency synthesized oscillator 154 is commanded to generate a substitution signal having a frequency corresponding to the first indication in the group obtained from the table which, in this case, corres-ponds to the highest frequency in the table. klicrocomputer 170 then waits for oscillator 154 to lock on the selected frequency. Step 282 determines whether or not switch 164 is still set to select the cables. If the selection has changed, the program returns to the director step 236.
If, at step 282, it is determined that switch 164 is still set to its cable position, the substitution signal is provided to converter 104 on both channels at step 284. Thus, microcomputer 170 causes switches 130, 132 and 140 to momentarily opèn, while switches 158 and 160 momentarily close.
At step 286, it is determilled wlletller a "hit" has been made ~i.e.~ whether oscillator 154 has generated a frequency to whicll converter 104 has been set).
I~'hen a hit occurs, amplifier 134 receives a signal from cable converter 104, ~7~-~,S~

which signal passes through bandpass filter 136, splitter 138 and is applied to single channel receiver 144. Receiver 144 causes a sampling signal to be generated which is applied to microcomputer 170. If a hit haS not occurred, step 288 causes -the frequency to whicll frequency synthesized oscillator 154 is set to be decremented by a fixed frequency-(6 ~IIIz in the prcferred embodiment) to the next highest frequency in the first group. The group count which initially indicates tlle number of frequencies in each group retrieved from ~he frequency table, is decremented to indicate the number of frcquencies left in the group. At step 290, it is determined whether the group count is equal to zero. Since tlle first group taken from the table was assumed to COIlsist of ten frequencies, the group count will equal nine. Since the group count is not equal to zero, microcomputer 170 returns to step 280 where the decremented frequency is loaded into oscillator 154. Assuming no hit occurs, the program makes nine more passes through steps 280 through 290 until the group count is equal to zero. When the group count does equal zero, the computer moves to step 292 wllere the group count is reset to one before the program returns to the frequency table interpreter at step 276. The next group of frequencies is then taken from the table as deter~ined by the pointer which moves progressively along the table. The program again moves to step 278 to determine whetller the end of the table has been reached. Assuming that the end of the table still has not been reached, the program moves through steps 280, 282, 284 and 286 l~hen step 286 indicates a hit has occurred, it is ne~ct necessary to determine whether the hit has occurred on cable 100 or cable 102. This is accomplished at steps 294 througll 299. At step 294, the same signal whicl caused the hit is substituted only on cable 100. Thus~ microcomputer 170 causes switclles 130, 132 and 140 to momentarily open and only switcll 15S to ~ 7~

momentarily close. At step 296, it is determincd whetller a hit has occurred, i.e., whether receiver 144 generates a sampling signal. If a hit has occurred, then converter 104 has been set to receivc a channel from cable lO0. I~ a hit has not occurred, thcn the program progresses to step 29S WhiCIl causes the same substitution signal to be applied only to cable 102. Thlls, switches 130, 132 and 140 are momentarily opened and only switcll 160 is momentarily closed. At step 299, it is determined wlletller a hit llas occurred. IE a hit has occurred, then converter 104 is set to recieve a chanllel on cable 102. I~
a hit does not occur at either step 296 or 299, the program moves to step 288 and continues on as if a hit has not occurred.
Assuming that a hit occurs either at step 296 or 299, thc trans-mission status is updated at step 300 ~Figure 12) to preliminarily indicate that the chanllel to which tl-e converter is tuned has been ~ound.
It is necessary to test for hits occurring during two consecutive searches. Thus, at step 302, succeeding step 300, it is determined w]lether the frequency substitution signal generated by oscillator 154 has been at a high or a low level. Since attenuator 156 was set in step 234 ~Figure 7) to generate a high level, the determination at step 302 will initially be negative. This causes the program to progress to step 303, at which the level of the prior hit is examined. Since this hit is the first hit, there was no prior hit so the program progresses to step 304 which causes microcomputer 170 to actuate attenuator 156 to produce low level frequency substitution signals.
Also, a register entitled "high level prior hit" is set. Control then proceeds to step 306 wherein the challnel selected by converter 104 is compared to the channel indicated by the prior hit. Since no prior channel had been selected, this determinatioll is negative so that the program moves bac~ to director 236.

4 r i~

Assuming that converter 104 remains tuned to the same channel identified at tlle higll level, microcomputer 170 will execute the appropriate steps 236 (Figure 7) through 274 (Figure 11) wllere the pointcr will again bc set to the start of the frequency interpreter table. ~licrocomputer 170 again executes steps 276 through 292 (Figure 11) to search for a hit. Once a hit is found, the program progresses througll steps 294 througll 299 to determine whether converter 104 is set to a challnel on cable 100 or cable 102. Tllere-after, the program again updates the channel status at step 300 and moves to step 302.
Note that a search is made on this second pass witll the frequency substitution signal at a low level to eliminate the problem of identifying submultiple frequel1cies. Suppose oscillator 154 genbrates a fundamental frequency of 216 ~z and a submultiple frequency of 108 ~l~lz. The fundamental frequency will have a much stronger component than the submultiple frequency.
If converter 104 is set to receive signals at 108 ~1z, receiver 144 may vary well generate a sampli1lg signal based on tlle submultiple when frequency sub-stitution signals are at a higll level. ilowever, the changes are greatly improved that receiver 144 will not generate a sampling signal when the fre-quency substitution signals are set at a low level.
At step 302, it will be determined that the low level has been selected. Accordingly, control passes to step 307 at which the low level is selected, or reselected, and the higll level prior hit register is cleared.
Next, step 306 determines whether the channel just identified is the same as the channel identified in the first search. If the channels are the same, as they should be during normal operation, we have a high level hit followed b~ a lo~ level hit at the same frequellcy. As a result, microcomputcr ~`7~5~

170 ne~t e~ecutes step 30S at W]liC]I t]lC ]lit value is set to four. The program then returns to director 236 in Figure 7.
At this point, thc challnel to ~YhiCh convcrter 10~ is set has been prcliminarily identified at the higll level and confirmed at the low level, and the hit value has been set to four.
The program again executes the appropriate steps 236 ~Figure 7) tllrough 272 (Figure 11). At step 272, it is determined that the hit value is not equal to zero. Accordingly, computer 170 e~ecutes thc steps illustrated in Figure 13. At step 310, the hit value is decremented to three. At step 312, it is determined whether the last hit indicates that converter 104 is set to receive a channel on cable 102. If the last hit indicated that converter 104 was set to receive signals from cable 100, this determination is negative so th~t control passes to step 314. At step 314, cable 100 is selected, switches 130, 132 and 140 are momentarily opened and switcll 158 is momentarily closed to enable substitution of the frequency substitution signal at the same fre~uency as the last llit on cable 100. At step 318, it is determined whether a hit has occurred at the same frequency. If a hit has occurred, step 320 verifies the transmission status to indicate cable 100 and the channel of cable 100 which llas been selected by converter 104. As will become apparent below, now that the channel has been verified, the program will move into a shortened program cycle which avoids the searching steps 274-300, at least until the channel is lost.
After verifying the channel status at step 320, the frequency sub-stitution signal is disabled at step 32Z and the hit value is again set to four. Thereafter, the program returns to director 23~ ~Figure 7). 0f course, if the hit had occurred on cable 102 instead of cable 100, steps 324 througl - 29 _ 7~5V

330 in Figure 10 would h~ve e~ecuted corresponding operations.
In some cases, after getting a ]lit at the high level, it will not be possible to get a hit at tlle low lcvel. ~\s a result, on the second pass, the searching performed by steps 276 through 292 (~igure 11) will cause the entire frequency table to be accessed. After the last frequency has been accessed, the determination of whether the end of the table has been reached will be positive in step 278. ~icrocomputer 170 will next execute step 332 in which the level of the last scan is examined. Since the last scan was at a low level, processing will proceed to step 334 where it will be determined whether the last hit was at a high level. Since, in this situation, a higll level ]~it was followed by no hit at a low level, the determination at step 334 will be positive so that control will proceed to step 336 at wllicll a high level for the frequency substitution signal will be selected and the high level prior hit register will be set. At step 337, microcomputer 170 will wait for oscil-lator 154 to lock ~if it is not locked) before returning to director 236 (Figure 7).
The program then moves through steps 236 through 272 ~Figure 11). At step 272, since the hit value remains equal to ~ero as set by step 234 (step 308 ~Figure 12) has not yet been executed since the second, low level search produced negative results), step 274 is executed.
The program then searches by repeatedly executing steps 276 through 292 until a hit is made at the high level. When a hit occurs, the program executes steps 294 through 299 to determine which cable the hit is on. At step 300 the channel status is updated.
At step 302 ~Figure 12) a determination is made as to w}lether the hit occurred at a low level. Since the hit did not occur at a low level, the -- ~0 --~7~

program advances to step 303 to determine whetller the prior hit was at a high level. In this situation, the prior hit was at a high level so control ad-vances to step 306 to determine wllether tl~e channel identified on this pass is the same as the channel identified on the firs-t pass. If the challllel identified on this high level is the sc~me as was identified on the first high level pass, then step 308 is executed, setting the hit value to four.
Thereafter, the program returns to director 236 and proceeds through step 272 (Figure 11). Since the hit value is not equal to zero, step 310 (Figure 13) is executed next. Assuming the hit occurred on cable 100, the hit is verified at step 318 and the transmission status is verified at step 320.
Thus, wilere a channel is first identified at a high level, but cannot be identified at a low level, if a llit can again be made at a high level on the same channel as a result of a search (actually twice at the high level at steps 286 and 296 in Figure 11), and can then be confirmed at a high level at step 318 (Figure 13), the channel status will be verified at step 320.
A third possible mode of channel identification exists. In this mode, a different channel is identified at the low level than at the previous high level. In this mode of channel identification, it is assumed that a first hit occurs at a high level and a second hit occurs at a low level, but on a different channel. During the first pass, steps 302, 303, 304 and 306 (Figure 12) are executed before returning to director 236. On the second pass, the determination at step 302 is positive so that a step 307, the low level is selected and the high level prior hit register is cleared. At step 306, the determination is negative so that the progr~D returns to director 236.
If, during the next pass, a cha~mel is identified as a result of a searcll (steps 276-292 in Figure 11) w]licll is the same as the chanllel identified in the preceding lo~ level pass, then the de-termination at step 302 that a lt~3 low level was selected is positive and the determination at step 306 that the present chanllel is tlle same as the prior chanllel is also posjtive. There-fore, the program proceeds to step 30S wllere the hit value is set to four before returning to director 236. During the ne~t pass througll the program, at step 272 (Figure 11) the hit value does not equal zero so that the program proceeds to Figure 13, and assuming another hitat step 318 or 328, tlle cllallnel status is verified in step 320 or 330.
Accordingly, the three modes of channel identification can be sum-marized as follows:
(1) If consecutive high level and low level searches identify the same channel, the hit is valid. (high hit - low llit identifying mocle) (2~ If a hit is not found during a low level search aftcr a hit had been found during a high level search, a search is performed at a high level.
If a higll level hit indicates the same channel as tlle previous high level hit, the hit is valid. (high hit - iow miss - high hit iden-tifying mode) (3) If a hit is fowld during a low level search on a channel dif-ferent from that indicated during a previous high level search, a low level search is repeated. If a hit is found at the low level on the same channel as the previous low level hit, the hit is valid. (high hit channel x - low hit chamlel y - low hit channel y identifying mode) Once the channel has been verified in one of the three modes des-cribed, the progra3n cycles from director 236 tllrougll the appropriate steps to step 272 and then through the appropriate steps 310 througll 330 of Figure 13.
In this shortened cycle~ the program avoids the search sequence of steps 274-300. This cycli3lg continues ulltil the chan3lel selected by converter 104 is changecl. Ilence, wllile the program immediately updates the chan3lel status ~7~

at step 300 to preliminarily indicate that the channel has been found after the first hit at step 296 or 299, the channel must be confirmed in onc of the three modes described before the program moves into the shortenecl progr~n loop which avoids the scanning procedure of steps 276-292 Once the channel is changed, the inquiry at step 318 or 328 ~Figure 13) will be negative. As a result, step 338 or 340 will disable the substitu-tion of the frequency substitution signal and step 3~2 will-determine whether the hit value is zero. In step 310, the hit value had been decremented to three. Therefore, the program will return to director 236 following step 342.
On the next pass through the program, the hit value will be decre-mented to two in step 310, and if there is another miss at step 318 or step 328, the program will again return to director 236. If misses occur on the next two successive passes through step 318 or step 328, the hit value will equal zero so that the program will move from step 342 to step 332 (Figure 11).
At step 332, the level of the last scan is examined. If it is assumed that the channel was determined based on a higll hit-low hit identifica-tion mode, or the ]-igh hit chalmel x, low hit channel y, low hi-t channel y identification mode, the determination of step 332 will be positive so that the program will move to step 334. Here, the determination will be negative since the high level prior hit was cleared at step 307 ~Figure 9). Accordingly, the program will move to step 346 w}lere a high level for the frequency sub-stitution signal will be selected for the next pass and the high level prior hit register will be cleared. At step 348, tlle hit value is set to zero and the transmit status is to be changed to indicate that the channel has been lost. The progr.~m returns to director 23G through step 337.
Ilence, once the channel has been idcntified, four consecutive misses at step 318 or 328 (Figure 13) are required before the transmit status is 7~

changed to indicate that the channel has been lost.
If the channel was identified according to the second mode clcscri~ed (high llit, low miss, high hit), and four consecutive misses bring the program to step 332, step 332 will cletermine that thc scan was not for a low level, so tllat the program proceeds to step 346. Again, the transmission status will be updated in step 348 to indicate that the channel llas been lost before the program returns to director 236. ~lence, the program is designed so that aftcr identifying a channel, microcomputer 170 will not change the channel status until after four consecutive misses.
At step 240 in Figure 8, the T-counter was both enablcd and started.
After a predetermined time, (in the presently prefcrred embodiment 256 counts) the T-counter overflows. This initiates the interrupt subroutine illustrated in Figures 14 and 15. Thc purpose of this interrupt subroutine is to transmit the data collected by cable meter 108.
Thus, with the initiation of the T-countcr overflow interrupt sub-routine at step 400, the program proceeds to step 402. At step 402, it is determined whetller the interrupt routine was initiated during a time dependent routine such as tlle testing of the vertical and hori~ontal signals in steps 245 or 247 (Figure 8). If a time dependent routine has been interrupted, the return address for the interrupt routine is set to tlle beginning of the time dependent routine in step 404. After step 404, or if no time dependent routine was interrupted, the program advances to step 406 wllere it is determined whether flag 1 is set. In the first pass througll the interrupt subroutine, the flag is not set so that the program advances to step 40S which causes the overflow counter to be dccremented. It will be recalled tha-t the overflow countcr was initially sct to a predetermilled n~llber i31. step 234 ~igurc 7).

It is also important to note that the overflow counter is different from the T-counter.
After decrementation of the overflow countcr in step 40S, stcp 410 determines whether tlle overflow counter is equal to zcro. Durillg the firs-t pass -through the program, the overflow counter will not be equal to zero so that the program advances to step 412, in which the input section subroutine is called and the carry bit ~later described) is set. Step 414 re-enables the overflo~ interrupt (since it ~ecomes disabled as soon as the T-counter over-flows). Step 416 returns the program to the point of the main program from w}lere it was interrupted.
I~hen the T-counter again overflows the interrupt subroutine will be re-executed so that the program moves through steps 402, possibly, 404, 406, 408 to 410. During the second pass througll tlle interrupt subroutine, the overflow counter will still not be equal to zero so that the program continues through steps 412, 414 and 416. Eventually, enough passes througll the interrupt routine will have occurred so that the overflow counter will be decremented to zero.
During this pass through the interrupt subroutine, a de-termination will be made at step 410 that the overflow counter is equal to zero. At step 418, flag 1 is set and the T-counter is set to -118. The program then moves through steps 412, 414 and 416.
At the next interrupt caused by the overflow of the T-cowlter (the T-counter will llave counted to 256 plus 118), the program moves to step 406 at which it is determined that flag 1 has been set. Therefore, the program advances to step 420 at which flag 1 is cleared. At step 422, the value of the T-counter is examined. Note that upon overflo~ing~the T-coullter begins countillg again. Thc program cycles through steps 420 and 422 until the l`-counter equals 2 so that the program becomes svnchrollized with ~2 ~

the system clock. The program moves on to step 424 at which the transmittcr gate is turned on.
As illustrated in Figure 15, at the ne~t step 426, it is cleterminecl whether the first word needs to bc transmitted. In step 239 we set thc transmit word equal to the first word. Therefore, this determination is positivc.
Accordingly, the program advances to step 428 in which the first and second words are constructed from the transmission status and the address registers and the transmit word is set to equal the second word. Each channel on each cable is identified by an eight bit code. Two eigllt bit words, each carrying four bits of the code,are required to transmit the eight bit code to the household collector. Each of the eight bit words include three bits of meter identification information in addition to the four bits of channel code informa-tion. The remaining bit in each word indicates whetller it is the first or second wQrd of the sequence. The control logic transmits appro~imately one word every two seconds to the collector. Consequently, two separate transmissions, taking appro~imately four seconds, are re~uired to transmit one complete channel identification code to the collector.
Step 430 stores the first word in register A. Then, in step 432, the parity bit is calculated and the word is transmitted.
In step 434, the transmitter gate is turned off. In step 436, the identification of the cable meter 108 doing the transmitting is read again, the address register is set with this identification and the overflow counter is reset.
In step 438, the activity counter is increl-,iented, and in step 440, the activity counter value-is checked. Since this is the first pass through step 438 and 440, the activity counter will be equal to one, so that the program proceeds through steps 412, 414 and 416 ~Eigure 11) before returning to the main program.

~27~5V

Over the next T-counter overflo~ interrupts, the program repeatedly moves through steps 400 through 416 until the overflow counter again equals zero at step 410. This causes flag 1 to be set in step 418, and at the next overflow interrupt, the program branches at step 406 to steps 420 through 426.
At step 426, since the transmit word was set to -the second word in step 428, the answer to the inquiry is negative so that the program advances to step 442 at which step the second word is stored in register A. The word is then transmitted along with ~,he parity bit at step 432. The program then moves through steps 434 through 440, and assuming that the datermination step 440 is negative, micro computer 170 returns to the main program through steps 412 through 416.
Note that the initial value of the overflow count together with the -118 ~Iremainder~ determine the time interval between transmissions. The initial value of the overflow counter is set by installer switches 168.
The -118 remainder is added to the T-counter in the preferred embocliment so that the approximate transmission interval is around two seconds. This approach allows for very accurate setting of the transmission interval.
If a hung routine develops such that the program makes several transmissions without executing step 237 (Figure 7) in which the activity counter is reset to zero, eventually the determination at step 440 will be positive. The program will then advance to step 444 at which the interrupted routine (presumably in which the hand has occurred) will be identified. At step 446, it is determined whether this hung routine is the TV signal good routine of Figure 9. If the hung routine is tllis routine, step 448 causes the transmission status to indicate this and control passes to step 452. If this routine was not causing the hang-up, the determination at step 446 is negative so that the hung routine must be either tl-e horizontal test of step 247 or the 1~7~5(~

vertical test of step 245. In either of these cases, step 452 determines whethcr thc activity counter is greater than or equal to 32. If this determina-tion is negative, the program returns to the main program througll steps 412 through 416. If, however, the determination at step 452 is positive, the program moves to step 454 and then to step 232 to restartthe main program.
Thus, the activity counter is incremented each time a transmission occurs and is reset to zero each time computer 170 executes the main program.
If, at some point, a problem devélops such that one of the testing routines are not properly e~ecuted, the activity counter will continue to be incremented witllout being reset. I~hen the activity counter reaches a prcdetermined level, the software is reset.
Both the main program, at steps 235 and 244, and the T-counter overflow interrupt subroutine, at step 412 call the input section subroutine illustrated in Figure 16. Aftcr entry at step 500, step 502 determines whether the cable input has been selected by switch 164 ~Figure 3). If it has been selected, step 504 enables the cable, causing s-~itciles 130 and 132 to be closed, and enables the converter output, causing switch 140 to be closed, and the converter flag is set. ~licrocomputer 170 then returns to the main program or the interrupt subroutine throug]l exit 506.
If, at step 502, it is determined that the cable has not be selected, microcomputer 170, in response to step 50B, enables the cable, causing switches 130 and 132 to be closed (some cable companies require the cable signal to constantly be applied to the converter), but disables the converter output, causing switch 140 to open. The program then moves to step 509 where the inquiry is ~hether the carry bit has been set. The carry bit can only be set at step 412 of the transmission routine of Figure 14. It is clearcd by step 243 1~ 7~ O

of the main program. ~here the carry bit is set, the program moves directly to exit step 506 without changing the transmission status. ~here the carry bit has been cleared, the program moves to step 510 to determine whether Aux 1 or Aux 2 has been selected. If Aux 1 has been selected, the status is updated to so indicate at step 512 and the converter flag is cleared. If Aux 2 is selected, the status is updated at step 514 and the converter flag is cleared.
Once the program has moved past step 235, the input selection routine is entered either from step 244 of'the main program or step 412 of the data trans-mission routine. Note that just before the input selection routine is entered from step 412 of the data transmission routi-nc, the carry bit is set in step 412, ensuring that the transmission status will not be changed in the middle of a transmission. On the other hand, just before the input selection routine is entered from step 244 of the main program, the carry bit is cleared in step 243 to allow the transmission status to be updated to indicate whether auxi-liary 1 or auxiliary 2 has been selected. ~ote, however, that even though the transmission status is not changed when the carry bit is set, the converter output is disabled in step 50~ in response to an indication from step 502 that the cable input is no longer selected.
Although only a single exemplary embodiment of this invention has been described in detail above, those skilled in the art will readily appre-ciate that many modifications are possible in the exemplary embodiment without materially departing from the novel teachings and advantages of this invention.
~or example, the presently preferred ~mbodiment is employed with a television system which receives signals by cable through an external cable converter.
Although certain advantages are inherent in this particular arrangcment with respect to connecting the present invention to the converter and the television, - ~73.. ~
, .

those of ordinary skill in the art will realize the advantages of employing certain aspects of the present inventioll even with a television having an integral tuner which may or may not be adapted to receive cable channels. In such a case, the output of the tuner would be directed to the present invention and the output of the present invention would be connected to the remainder of the television circuitry. Furthermore, those of ordinary skill in the art would readily appreciate that certain aspects of the present invention are equally suitable for monitoring a radio receiver or any other communications receiver, whether the~communications are transmitted over a cable, by electro-magnetic signals througll the air or over any other media.
Accordingly, all such modifications are intended to be includedwithin the scope of this invention as defined in the following claims.

Claims (3)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A method of detecting which of a plurality of carriers has been selected for reception by a television system, said system including a selector for selecting one of said carriers, said method comprising the steps of:
counting the number of horizontal sync pulses between consecutive vertical sync pulses in television signals from said selector;
determining from said counting step when said tele-vision signals are acceptable; and only after said determination is positive, detecting which of said carriers has been selected by said selector.

2. A method as in claim 1 wherein said determining step produces a positive determination only when at least a first predetermined number of the just previous N counts by said count-ing step are acceptable and at least the just previous M conse-cutive ones of those N counts are also acceptable.

3. A method as in claim 1 further comprising the step of attenuating said television signals when at least a first pre-determined number of the just previous results of N of said counting steps are not acceptable.