Method, and application thereof, for transmitting information over a common signal path

Claim 3

A method as claimed in claim 1, in which the units of information in the individual signal channels are supplied by associated sources of information in the form of immediately consecutive words, each of which comprises a pre-determined number p - the same for all words supplied on the same signal channel - of binary units of information, the words that contain the information preferably being repeated, when the units of information of the information sources remain unchanged, until the units of information change.

Claim 4

A method as claimed in claim 2, in which, on each of the n signal channels, each of the words supplied comprises a number p - the same number for all words - of binary units of information.

Claim 5

A method as claimed in claim 2, in which, to prevent such errors in transmission of the information as might be caused by faults on the common signal path, the individual words are each repeated many times by the information sources upplying them, a number of successive words being evaluated for determination of the invormation content of those words, for which purpose the duration of the words supplied by the information sources is made many times shorter than the period during which a unit of information from those information sources remains constant or during which the information from those sources representing an analogue magnitude lasts as a minimum in the event of a maximum change, in the same direction, of that magnitude.

Claim 6

A method as claimed in claim 4, in which q immediately consecutive words, each comprising p units of information in p successive positions and of which the first word in a zth. evaluation cycle starts at an initial point in time tA to+koT+(z-1)pqT+ Delta tm, are evaluated for determining the information content, the sum

Claim 7

A method as claimed in claim 4, in which the units of information are used for transmitting the values of physical or chemical magnitudes subject to variation with time, for which purpose the maximum possible range of variation of those values, which are to be transmitted over an mth. signal channel, is divided into a number of adjacent fractional ranges, preferably to the number of 2w, to each of which fractional ranges a characteristic binary number comprising w binary units of information is allotted, that binary number being fed as a word or part of a word to the mth. signal channel when the value to be transmitted lies within that allotted fractional range; and in which the duration chosen for the individual words to be transmitted over the mth. signal channel is many times shorter than the minimum time required by the time-variabLe magnitude, the value of which is to be transmitted over the mth. signal channel, to pass through one fractional range.

Claim 8

A method as claimed in claim 5, in which the binary numbers allotted to the individual fractional ranges are such, or are so converted by Gray coding before transmission over the common signal path, that binary numbers denoting adjacent fractional ranges differ from each other only in their w positions; and in which, after the said summation, it is determined for each of the said p word positions whether the result of summation falls either into a pre-determined tolerance range, the width of which is governed by the permissible noise level, by a value equal in magnitude to zero and representing the binary decision zero, or into a similar pre-determined tolerance range by a value equal in magnitude to q times the said integration result obtained in the complete absence of noise for a single unit of information transmitted by the impression of the associated carrier signal on the common signal path, which value represents the binary decision ''''one,'''' or, again, into a range outside both of those tolerance ranges; and in which, the foregoing having been determined, a word is formed when in at least (p-1) of the p positions the summation result falls into one of the two tolerance ranges, so that well-defined binary decisions are produced for the positions in question; and in which that word, from which it can be deduced that the value of the physical or chemical magnitude to be transmitted lies within two particular adjacent fractions of the said range of variation, is passed forward as the transmission result, preferbly being converted from the Gray code to the natural binary code before being passed forward.

Claim 9

A method as claimed in claim 1, in which the orthonormal functions Fl(t) to Fn(t) chosen are Walsh functions, the cycle of which is divided into s equal time divisions, at the transitions from the first to the next of which may lie abrupt variations in the Walsh functions, s being a power of two which is at least equal to that power of two which is next higher in relation to n .

Claim 10

A method as claimed in claim 8, in which to determine the integral

Claim 11

A method as claimed in claim 9, in which the time-shifted summation signal S in each individual time division is multiplied by each of the n time function Fl to Fn and the n products S.Fl to S.Fn of those multiplications are added singly in sequence, in the final output-side stage of an n-stage shift register forming a closed ring by connection of its output to its input, to whatever value is held therein; and in which, after each such addition, the values held in the individual stages of the shift register are shifted one stage forward, so that the values held in the shift register carry out one complete circuit in the shift register during every time division and accordingly the products S.Fm associated with a given mth. signal and formed in successive time divisions, in each instance one time division apart, are added together iN the shift register, so that within one determination period the said sum of the products S.Fm is formed.

Claim 12

A method as claimed in claim 3, in which, over a complete evaluation cycle comprising p.q consecutive determination periods, the connection between the output and input of the n-stage shift register is broken between every two consecutive determination periods, following which the output of the n-stage shift register is connected to the input of a further shift register having (p-1) .n stages and the input of the n-stage shift register is connected to the output of that further register and then, in the ring thus formed from the n-stage shift register and the futher shift register, the values in the individual register stages are shifted forward n stages at a time, after which the further shft register is separated again from the n-stage shift register and connection is re-established between the output and input of the n-stage shift register and then, at the end of the evaluation cycle, the values in the n-stage shift register and in the further shift register, each of which constitutes the sum

Claim 13

A method as claimed in claim 11, in which, for withdrawing the values stored in the two shift registers, the connection is broken between the input and output of the n-stage shift register and its output is connected to the input of the further shift register and the values in both the registers are then, by further shifting by n.p stages, discharged from the two registers via the output of the further register, and after the first n-stage shift the n-stage shift register thereby cleared is again separated from the further shift register and connection is re-established between input and output of the n-stage shift register, whereupon the fresh evaluation cycle starts.

Claim 14

A method as claimed in claim 1, in which, for transmitting information from several differently located senders to a common reception point, the common signal path is routed from the reception point through each of the send locations, a separate sender being provided at each send location and a central receiver at the common reception point; and in which, for filtering out, in the central receiver, the information allotted to the individual channels from the summation signal arriving at the reception point, n period-synchronous time functions Fl to Fn are generatd and, for synchronisation of the carrier signals contained in the summation signal arriving at the reception point, synchronising signals are generated with those time functions and fed to the receiver and to the individual senders and with their aid in the receiver the strat of cycle of the time functions Fl to Fn, and in the individual senders the start of cycle of the carrier signals applied to the common signal path by the sender concerned, are so determined that the carrier signals, after transmission from the sender concerned to the central receiver, which requires a certain transmission time, are at least approximately in period synchrony at the reception point with the time functions Fl to Fn generated in the receiver.

Claim 15

A method as claimed in the claim 8, in which the synchronising signals passed to the receiver and the individual senders trigger the start of cycle of the time functions Fl to Fn generated in the receiver as they arrive in the receiver and trigger the start of cycle of the carrier signals applied to the common signal path by the sender concerned as they arrive in the individual senders; and in which the duration chosen for the time divisions is greater than the time summAtion derived from the transmission time of the synchronising signals from the receiver to the most distant sender on the common signal path and the transmission time of the carrier signals from that most distant sender to the receiver.

Claim 16

A method as claimed in claim 8, in which control signals derived from the synchronising signals fed to the receiver cause within the receiver a backward shift of the start of cycle of the time functions Fl to Fn generated therein, in relation to the synchronising signals generated in the receiver, by a time interval at least equal to the time summation derived from the transmission time of the synchronising signals from the receiver to the nearest sender on the common signal path and the transmission time of the carrier signals from that nearest sender to the receiver, and at most equal to the time summation derived from the transmission time of the synchronising signals from the receiver to the most distant sender on the common signal path and the transmission time of the carrier signals from that most distant sender to the receiver, and preferably equal to the mean value between those two time summations, those control signals preferably following the synchronising signals in the said time interval and, as they arise, triggering the start of cycle of the time functions Fl to Fn generated in the receiver; and in which the synchronising signals fed to the senders, as they appear in the individual senders, trigger the start of cycle of the carrier signals applied by the sender concerned to the common signal path; and in which the chosen duration of the time divisions is both greater than the difference between the duration of the said time interval required in all for transmission of the synchronising signals from the receiver to the nearest sender and for transmission of the carrier signals from the nearest sender to the receiver, and greater than the difference resulting between the time required in all for transmission of the sysnchronising signals from the receiver to the most distant sender and for transmission of the carrier signals from the most distant sender to the receiver and the length of the said time interval.

Claim 17

A method as claimed in claim 13, in which, in the individual senders, the start of cycle of the carrier signals applied by the sender concerned to the common signal path is shifted forward in time, in relation to the start of cycle of the time functions Fl to Fn generated in the receiver, by the transmission time from the sender concerned to the central receiver and this forward shift is maintained by the synchronising signals fed by the central receiver.

Claim 18

A method as claimed in claim 14, in which control signals derived from the synchronising signals fed to the receiver and the individual senders cause in the receiver a backward shift of the start of cycle of the time functions Fl to Fn generated therein and in the individual senders a backward shift of the start of cycle of the carrier signals generated therein, in each instance by a time interval, in relation to the synchronising signals supplied, which, at the receiver, equals the time summation derived from the transmission time of the synchronising signals from the receiver to the most distant sender on the common signal path and the transmission time of the carrier signals from that most distant sender to the receiver and an additional time shift lying optionally between zero and the time difference between two successive synchronising signals and, at the individual senders, equals the time summation derived from the transmission time of the synchronising signals from the sender concerned to the most distant sender and the transmission time of the carrier signals from the most distant sender to the sender concerned and the said additional time shift, the control signals derived from the synchronising signals preferably following the synchronising signals provided in the said tIme interval and, as they appear, triggering the start of cycle of the time functions Fl to Fn in the receiver and the start of cycle of the carrier signals in the individual senders.

Claim 19

A method as claimed in claim 17, in which the synchronising signals are generated periodically and the additional time shift is equal to the difference between the duration of a synchronising signal cycle and the total time required for transmission of the synchronising signal from the receiver to the most distant sender and for transmission of the carrier signals from the most distant sender to the receiver, so that the time interval by which the control signals in the receiver cause a backward shift in the start of cycle of the time functions Fl to Fn generated therein in relation to the synchronising signals fed to the receiver is equal to the duration of a synchronising signal cycle and hence the periodically generated synchronising signals themselves can be used as control signals in the receiver.

Claim 20

A method as claimed in claim 3, in which the synchronising signals are generated periodically and the chosen duration of a synchronising signal cucle is equal to the duration of a word, and in which, in the individual senders, the start of each of the words supplied by the associated information sources is set to the same moment as the start of cycle set by means of the synchronising signals for the carrier signals applied by the sender concerned to the common signal path, so that each of the words from all senders, after transmission from the sender concerned to the central receiver, starts at least approximately at the same moment at the reception point.

Claim 21

A method as claimed in claim 11, in which the synchronising signals are generated periodically and the chosen duration of a synchronising signal cycle is equal to the duration of an evaluation cycle comprising q consecutive words, and in which, in the individual senders, the start of the first words in each evaluation cycle supplied by the associated information sources is set to the same moment as the start of cycle set by means of the synchronising signals for the carrier signals applied by the sender concerned to the common signal path, so that each of the first words in each evaluation cycle from all senders, after transmission from the sender concerned to the central receiver, starts at least paproximately at the same moment at the reception point.

Claim 22

A method as claimed in claim 13, in which, as period-synchronous time functions Fl to Fn and as carrier signals, use is made of Walsh functions, the cycle of which is divided into s equal time divisions, s being a power of two, 2r , which is at least equal to the power of two next above n , and in the central receiver a timing oscillation in the form of a square pulse with a duration corresponding to twice the time division is generated and sent to the individual senders, the timing oscillation being used in the receiver for generating the Walsh functions constituting the time functions Fl to Fn and in the individual senders for generating the Walsh functions constituting the carrier signals applied by the sender concerned to the common signal path.

Claim 23

A method as claimed in claim 21, in which the Walsh functions in the receiver and in the individual senders are derived from Rademacher functions, which in turn are generated from the timing oscillation by means of an (r-1)-stage binary counter controlled by the timing oscillation and stepping one count forward for every cycle of the timing oscillation, and, for deriving a predetermined Walsh function from the Rademacher functions arising at the counter input and at the count-stage outputs of the (r-1)-stage binary counter, pre-determined Rademacher functions appropriate to the Walsh functions to be derived are selected and the binary values represented by these are added on a binary basis in the binary zero order position and, according to the binary decision arising as the result of that addition, one or other of the two functions values which a Walsh function can assume is set up, the function value thus set up being the function value of the derived Walsh function in the time division associated with the count of the (r-1)-stage binary counter and the function value of the timing oscillation.

Claim 24

A method as claimed in claim 10, in which, in the receiver, to generate the function values of all n Walsh functions constituting the time functions Fl to Fn, in every individual time division, each of the r Rademacher functions arising at the counter input and at the outputs of the various counting stages of the (r-1)-stage binary counter is applied to a line connected to the input and output concerned, which is routed through and can be completed by an electronic switch, each of those elecronic switches being operated by one of the r counting stages in a further r-stage binary counter, which passes through n pre-determined count positions in every time division, and is opened or closed according to the position of the counting stage by which it is operated; and in which, over the lines when thus completed, the binary-values represented by the Rademacher functions applied to those completed lines - and, over the lines when not so completed, the binary values zero - are fed into an adding devcie, in which the binary values fed in from all the lines are added binarily in the binary zero order position and by which, according to the binary decision of the number produced as the result of that addition, one or other of the functions values which a Walsh function can assume is set up, the function value so set up being the function value of the Walsh function associated with the instantaneous count of the said r-stage binary counter in the time division associated with the count of the (r-1)-stage binary counter and the function value of the timing oscillation.

Claim 25

A method as claimed in claim 14, in which the (r-1)-stage binary counters in the receiver and in the individual senders are set, by the synchronising signals fed to the receiver and the individual senders, to the start-of-cycle count level, preferably zero.

Claim 26

A method as claimed in claim 15, in which the (r-1)-stage binary counter in the receiver is set in each instance, by setting pulses constituting the said control signals, which occur at least approximately at the same moment as the synchronising signals that give rise to them, to a count level which is in advance of the start-of-cycle count, preferably zero, of that binary counter by the said time interval or by a number of count units corresponding to the quotient of time interval/time division; and in which the (r-1)-stage binary counters in the individual senders are set to the start-of-cycle count level, preferably zero, by the synchronising signals associated with the individual senders.

Claim 27

A method as claimed in claim 17, in which the (r-1)-stage binary counters in the receiver and in the individual senders are set in each instance, by setting pulses constituting the said control signals, which occur at least approximately at the same moment as the synchronising signals that give rise to them, to a count level which is in advance of the start-of-cycle count, preferably zero, of the binary counter to which the setting pulses are applied, by the said time interval associated with the receiver or sender in which that binary counter is incorporated or by a number of count units corresponding to the quotient of time interval/time division derived from that time interval and from the time division.

Claim 28

A method as claimed in claim 8, in which a two-core cable is used as the common signal path, an impedance, preferably an ohmic resistance and a voltage source in series being connected to the two condUctors in the cable at the reception point, and the summation signal that arrives at the reception point is picked up via the impedance; and in which, at each sending point, for each separate signal channel over which information is supplied at the sender concerned, there is connected to each conductor in the cable a separate series arrangement for the signal channel concerned, consisting of a further impedance, preferably also an ohmic resistance and an electronic switch; and in which the electronic switch, when the carrier signal allotted to the signal channel concerned is not applied to the common signal path, is moved to one or other of its two switching positions ''''open'''' and ''''closed'''' - preferably to the ''''open'''' position - and, when the carrier signal consisting of a Walsh function allotted to the signal channel concerned is applied to the common signal path, is moved to one of its two switching positions when the Walsh function constituting the carrier signal assumes one of the two function values which that Walsh function can assume, and is moved to the other of its two switching positions when the Walsh function constituting the carrier signal assumes the other of the two function values which that function can assume.

Claim 29

A method as claimed in claim 27, in which the further impedances in the series arrangements connected to line at the individual sending points are such that when any electronic switch is closed in the series arrangements connected to line at the individual sending points, the same increase in voltage occurs in each instance at the impedance forming part of the series arrangements at the reception point, ohmic resistances preferably being used as impedances in all the series arrangements and the magnitudes chosen for those ohmic resistances being equal in each case to the difference between a pre-determined resistance value, higher than the ohmic line resistance from the reception point to the sending point most distant along the line, and the value of the ohmic line resistance from the reception point and that sending point at which the series arrangement incorporating the ohmic resistance concerned is connected to line.

Claim 30

The application of the method as claimed in claim 1 to the automatic monitoring, from a central point, of a number of measuring points situated at different locations, a sender being provided at each measuring point and a receiver at the central point, the measuring values obtained at the various measuring points being transmitted to the receiver in each instance by the sender associated with the measuring point concerned.

Claim 31

Application as claimed in claim 29 to cable monitoring and in particular to the monitoring of a high tension cable in regard to its temperature by temperature sensing and in regard to possible leaks by detecting any leakage of cable oil and by detecting any penetration of water into the cable.