Field of the Invention
Our present invention relates to central-office equipment of a telephone or other telecommunication system, more particularly to a call monitor for an exchange serving a multiplicity of telecommunication lines through associated line links which can be temporarily interconnected with the aid of a switching network under the control of a processor.
Background of the Invention
In such telecommunication systems, the momentary state of a connection between a calling and a called station -- and therefore between two local line links temporarily allotted to the respective lines -- is conventionally registered within the processor, e.g. with the aid of several circulating memories as described in U.S. Pat. No. 3,581,016. The data thus stored in the processor include not only ad hoc information, such as the addresses of the intercommunicating lines and the instantaneous activities thereof, but also standard and recurring functions relating to signal generation and transcoding, for example. This requires memories of high storage capacity within the processor, particularly in central offices or exchanges serving a large number of telecommunication lines.
Object of the Invention
The object of our present invention, therefore, is to provide relatively simple circuitry in such an exchange for monitoring the progress of a call independently of the centralized processor and calling that processor into service only in certain stages of the progress of a connection requiring intervention of its dynamic storage facilities.
Summary of the Invention
We realize this object, in accordance with the present invention, by providing in the exchange a first memory (which may be of any suitable conventional type) with a number of phases corresponding to the number of time slots in a scanning cycle, each phase being allocated to a respective line link, and a preferably static second memory of the read-only type containing codes which identify all possible states of a connection including those which are of interest to the processor and those intervening conditions which do not require more than the execution of certain switching operations. A comparator receives, during each time slot, stored data from the first memory and updating information from the corresponding line link on one set of inputs and a succession of evolutionary codes from the second memory on another set of inputs, a detection of a match halting the readout of codes from the second memory while initiating the readout of an accompanying instruction to the processor and/or to the switching network.
Brief Description of the Drawing
The above and other features of our invention will now be described in detail with reference to the accompanying drawing in which:
FIG. 1 is a block diagram showing the basic components of a central office equipped with a call monitor according to our invention;
FIG. 2 is a more detailed diagram of the call monitor shown in FIG. 1; and
FIG. 3 is a set of graphs serving to explain the operation of the system of FIG. 2.
Specific Description
In FIG. 1 we have shown a set of subscriber lines L which can be selectively connected with one another and are coupled, at least temporarily, to a multiplicity of local line links G.sub.1, G.sub.2 . . . G.sub.n. These line links may include the usual conductors for incoming and outgoing signals along with other equipment such as tone generators and (in case of PCM data transmission) synchronization-signal extractors. A line scanner S, periodically stepped under the control of a clock circuit included in a programmer CK (FIG. 2), samples the several line links in respective time slots of a scanning cycle. Incoming signals are fed to the centralized equipment by way of a lead 1 terminating at a signal receiver RS which evaluates the voltage samples from active lines, e.g. in the manner described in our commonly owned application Ser. No. 517,017 filed Oct. 22, 1974, and forwards the resulting information over a lead 2 to a call monitor LS more fully described hereinafter with reference to FIG. 2. Monitor LS communicates via a lead 3 with a central processor EL and can also transmit instructions by way of a lead 6 to a switching network RC. Unit LS further cmmunicates by way of a lead 4 with a transmitter TS for outgoing signals which are delivered by way of a lead 5 to the scanner S. It will be understood that leads 1-6 may each consist of multiple conductors. Signal transmitter TS may include a conventional coder for converting outgoing binary signals into specific waveforms. Subscriber lines L are also connected to switching network RC by way of an analog/digital interface unit I.
We shall now describe, with reference to FIG. 2, details of the call monitor LS embodying our invention. The units of this call monitor are controlled, in a manner more fully explained below, by the aforementioned programmer CK generating output pulses T1, T2, T3, T4, T5, T6, TG, C, W1 and W2 which are energized at different times in the course of any time slot of duration T.sub.i (see FIG. 3). Output signal TG is a binary code word, transmitted over a multiple 7 to a multiplexer MX, which changes from one time slot to the next and represents the address of the line links G.sub.1 -G.sub.n (FIG. 1) which are sampled during these time slots. Multiplexer MX works via another set of conductors 8 into a primary memory MG with a multiplicity of phases r, as described in the aforementioned U.S. Pat. No. 3,581,016, for the storage of information relating to the activity of the respective line links. Two such phases, designated r.sub.i and r.sub.j, are assigned to a pair of these line links which are assumed to have established communication with each other.
Each phase r of memory MG consists of a row of storage elements subdivided into four sectors, i.e. a first sector containing the address code CGI.sub.i, CGI.sub.j of the assigned line link, a second sector containing a coded information SGI.sub.i, SGI.sub.j on the activity of that line link, a third sector SEL.sub.i, SEL.sub.j carrying similar information on the state of an ongoing connection, if any, and a fourth sector which upon initiation or establishment of such a connection contains the address code of the communicating line link, i.e. a code IGC.sub.j in the case of phase r.sub.i and a code IGC.sub.i in the case of phase r.sub.j under the conditions here assumed.
Another memory MP, of the read-only type, consists of a multiplicity of pages storing, by conventional microprogramming technique, the codes of the various evolutionary states of a connection as discussed above. These evolutionary states can by cyclically read out, in the course of a single time slot, at an output X of memory MP which is addressed via a lead 20 (also a multiple) by a binary counter CP adapted to receive stepping pulses SI from an AND gate P1. Another output Y of memory MP carries instructions to be read out onto a lead 23 with the appearance of respective evolutionary codes in outlet X. These evolutionary codes are transmitted by a lead 22 to a set of inputs of a comparator CF having another set of inputs tied by a lead 24 to a buffer register RI. Another buffer register RO is divided into four sections RO1, RO2, RO3, RO4 having inputs linked via respective leads 9, 10, 11, 12 with corresponding sectors of whichever phase r of memory MG is in scanning position during a given time slot. A branch 14 of lead 10 extends to the input side of register RI, as does a lead 15 originating at register section RO3. Register RI further receives signals from units RS and TS of FIG. 1 by way of lead 2 and a lead 17 forming an incoming branch of the 2-way connection 4 between call monitor LS and signal transmitter TS. Register section RO1 has an output lead 18 extending to memory MP whereas section RO2 preloads the counter CP via a lead 19. A lead 13 extends from section RO4 back to multiplexer MX which also receives signals from processor EL via a lead 34 forming part of the connection 3. Other leads 32 and 33 carry information from the processor to the third and fourth sectors, respectively, of memory MG.
A branch 30 of lead 8 extends to a further buffer register RE. A fourth buffer register RU receives the instructions read out over lead 23 from memory MP and delivers them to switching network RC via lead 6; a branch 29 of that lead forms the outgoing part of connection 4 whereas another branch 26 thereof terminates at an input 27 of register RE and has an extension 28 leading to the second sector of the scanned phase of memory MG. The contents of register RE can be discharged via a lead 31, forming part of connection 3, into processor EL upon the energization of an output lead 35 of an AND gate P2 having an input tied by a lead 21 to an output of register RU. Lead 35 also extends to a setting input of a flip-flop DP with a set output 36 and a resetting input 37 tied to respective conductors of multiple 3.
The establishment of connections between output 8 and inputs 7, 13 and 34 of multiplxer MX is controlled by programmer pulses T1 and T2. Pulse T3 enables the loading of register RO from memory MG. Pulse T4 causes the simultaneous discharge from register RI, via lead 24, of all the data fed in by way of leads 2, 14, 15 and 17; thus, lead 24 consists of as many conductors as the combination of leads 2, 14, 15 and 17, the same being true of lead 22 inasmuch as circuit CF is to perform a bit-by-bit comparison between the data appearing on leads 22 and 24, respectively. Pulse T5 unloads from register RU the information fed in via lead 23 from memory MP. Pulse T6 reads the count of circuit CP into the address input 20 of memory MP. Pulse C opens the AND gate P1 for the passage of an output signal from comparator CF to the stepping input of counter CP. Pulse W1 enables the writing of the contents of register RU, via leads 26 and 28, in the second circuit of the operatively positioned phase of memory MG simultaneously with an unblocking of AND gate P2 for the discharge of register RE and the setting of flip-flop DP if, and only if, an instruction destined for processor EL is present in register RU. Pulse W2 enables the writing of information from the processor, arriving over leads 32 and 33, in the third and fourth sectors of the operative memory phase.
We shall now describe the operation of the system of FIG. 2 with reference to the several graphs of FIG. 3 showing a time slot T.sub.i, allocated to the line link identified by memory phase r.sub.i, which starts at an instant t.sub.O following by the usual guard interval the termination of the preceding time slot allocated to a different line link.
At instant t.sub.O, the simultaneous absence of pulses T1 and T2 sets the multiplexer MX in a position in which the address code of the corresponding line link is transmitted from multiple 7 to multiple 8 as an instruction for the reading of the contents of phase r.sub.i. The information stored in the several sectors of this memory phase is thus transmitted via leads 9-12 to the several sections RO1-RO4 of register RO, respectively.
Immediately thereafter, at instant t.sub.1, register RO is loaded by pulse T3 so as to energize its output leads 18, 19, 15 and 13 with the information received from memory MG.
At instant t.sub.2 the pulse T6 is generated along with pulses T1 and T2. The appearance of the latter two pulses switches the multiplexer MX into a position in which it activates its input lead 13 in lieu of lead 7 whereby the address IGC.sub.j of the communicating line link is read into the input lead 8 of memory MG as an instruction for the readout of the information stored in the second sector of phase r.sub.j, i.e. the activity code SGI.sub.j, via lead 14 to register RI. At the same time the corresponding data code SGI.sub.i, previously read out from phase r.sub.i into register section RO2, is loaded into counter CP by way of lead 19 under the control of pulse T6. Since pulse C is absent, the transmission of a stepping pulse SI to the counter is blocked so that no further change takes place at this time in the contents thereof. The data from register section RO3 are introduced into register RI by lead 15. Memory MP receives the count on lead 20 together with the address code on lead 18 from counter CP and register section RO1, respectively. The address code on lead 18 selects in memory MP the page assigned to the corresponding line link whereas the count on lead 20 identifies on that page a particular revolutionary code determined by the activity code SGI.sub.i.
At instant t.sub.3, the appearance of pulse T4 simultaneously discharges into input 24 of comparator CF all the information loaded into register RI via leads 2, 14, 15 and 17. Input 22 of the comparator receives at the same time the evolutionary code read out from memory MP under the control of counter CP for bit-by-bit comparison with the contents of register RI. If the comparator fails to detect a match, it energizes its output lead 25; otherwise that lead carries zero voltage and no significant changes take place in the position of counter CP and memory MP for the remainder of the time slot.
At instant t.sub.4 the programmer CK emits a series of pulses C to unblock the AND gate P1. This unblocking is ineffectual if lead 25 is de-energized because of an identity between the codes on comparator inputs 22 and 24. Otherwise, a stepping pulse SI advances the counter CP with each pulse C and thereby causes the readout of successive evolutionary codes from memory output X until the comparator CF responds and cuts off the stepping pulses.
The train of unblocking pulses C, equal in number to the evolutionary codes stored on any page of read-only memory MP, comes to an end at instant t.sub.k.sub.-6 in the latter part of the time slot. Next, at instant t.sub.k.sub.-5, pulse T5 is generated to load the buffer register RU with the instruction (if any) appearing on output Y of memory MP; that instruction represents the updated activity code which is to appear in phase r.sub.i during the following scanning cycle.
At instant t.sub.k.sub.-4, the setting of multiplexer MX is restored to its initial condition with the disappearance of pulses T1 and T2. Since programmer CK still emits on lead 7 the address code of the line link associated with phase r.sub.i, that address is now transmitted on lead 30 to an input of register RE.
At instant t.sub.k.sub.-3 the writing pulse W1 causes the inscription of the code from register RU in the second sector of phase r.sub.i and simultaneously unblocks the AND gate P2 for the passage of a discriminating bit, if present, from register RU via lead 21 to lead 35 for the setting of flip-flop DP and the discharge of register RE into lead 31 extending to the processor EL. The instruction from register RU, if thus read out to the processor on lead 31, is accompanied by the address from multiplexer MX.
At instant t.sub.k.sub.-2 the pulse T1 reappears at multiplexer MX to switch same into its third position for transmittal of information from the processor, arriving over lead 34, to the address input 8 of memory MG. Processor EL is not synchronized with the scanning cycle of programmer CK so that the address information arriving at that point is not necessarily related to the line link identified by the phase being scanned, i.e. memory row r.sub.i in the present instance. That address controls the inscription of information in the third and fourth sectors of a memory phase r, via leads 32 and 33, upon the occurrence of writing pulse W2 at instant t.sub.k.sub.-1. Thus, the processor may modify the contents of these two sectors of any phase r in the terminal period of any time slot of a scanning cycle.
If the presence of a discriminating bit on lead 21 causes the setting of flip-flop DP, output 36 thereof transmits a "data ready" signal to the processor which responds with a resetting pulse for the flip-flop on its input 37. If necessary, a further buffer register (not shown) may be inserted between the processor and the register RE to allow the feeding of the contents of the latter register to the processor at a time independent of the scanning cycle.
It will thus be seen that the call monitor LS according to our invention decides what information, if any, should be forwarded to the processor EL, to the switching circuit RC and/or to the signal transmitter TS on the basis of signals arriving from signal receiver RS. This minimizes the number of interventions of the processor in the establishment and dissolution of a connection. The system is highly flexible since any change in the structure of the network can be accommodated simply by the storage of new data in read-only memory MP.