JPS6339159B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a time division multiplexed telephone switching system.

[Conventional technology]

Time division multiplexing systems, including those operating under the control of digital computers, are very well known and widely used. See US Pat. Nos. 3,115,552, 3,172,956 and 3,401,235 for time division multiplexed systems.

In some cases, time division multiplexing of telephone signals is incorporated into switching equipment, such as a private branch exchange (PBX).

These switching equipment interconnect all classes of trunk sources, station lines, and other lines, perform switching between them, and transmit the signals present on those lines to the public. Aggregation into trunk lines connected to the telephone network.
By doing so, those public telephone line network access trunk lines can be used efficiently. To increase the reliability of such telephone switching systems, a number of practical telephone switching systems can be used, including those that utilize redundant computers. An example of such a time division multiplex system will now be described with reference to FIG. 6. These switching devices typically include an interface unit (including response monitoring, analog-to-digital switching, and digital-to-analog switching functions), a concentrator unit, and a switch unit.

For purposes of this application, these devices and their associated devices will be referred to as "modules" or "replacement modules."

To expand the capacity of these modules, especially PBXs, switching units and concentrating units are interconnected. These interconnections are most often made in multi-tiered, hierarchical architectures. Such conventional interconnection schemes are described with reference to FIGS. 2 and 3. This hierarchical architecture is not always efficient. A single unit used in those telephone switching systems to perform higher order switching must be large enough to handle some maximum number of modules. This high order exchange is not cost efficient when used with fewer than this maximum number of modules. A corollary of this is that these telephone switching systems cannot be easily expanded beyond the maximum number of modules.

A graded multiple architecture is used to interconnect the lines using modular step-by-step switches within the central office. In such architectures, it is well known that direct links efficiently combine deterministic first offered traffic. The telephone traffic that floods those direct links represents a more randomly fluctuating traffic. This overflow traffic can be efficiently handled by a shared link (referred to here as a "bus"). A combination of overflow handling buses and direct links is often used in distributed architectures. An example of such a conventional telephone switching system, which has been in use for many years, is shown in FIG.

Another architecture used in data networks uses multiple time division multiplexing (TDM) switches arranged in a ring. In this ring arrangement, calls are routed along the ring through other modules to a particular module. but,
If time slots are used in each intermediate module through which a call passes, the capacity of the TDM switch in each module will be used with relatively low efficiency. Further, if a ring is broken due to failure of one module, the remaining normal modules in the ring cannot be used to make bidirectional connections. A conceptual diagram of this architecture is shown in Figure 1b.

[Problem that the invention seeks to solve]

As explained below, the present invention discloses a distributed interconnect architecture that combines features of both ring and multiple glazing architectures. The architecture of the present invention is compatible with several commercially available digital
Well suited for interconnecting TDM switches or other modules used in telephones.
In order to effectively route calls through each module without requiring the use of time slots in each module's main TDM bus,
A separate digital interconnection unit is used. Accordingly, the present invention provides a dynamic multiple glazing architecture.

[Means for solving problems]

This specification provides a distributed interconnection scheme for use in a time division telephone switching system consisting of a plurality of similar switching modules. Each module is connected to multiple communication lines, such as trunk switched lines, trunk lines, and other lines. Those modules are called circuit paths.
The signals traveling on those lines are aggregated in a well-known manner into time-shared digital signals for easy completion between the lines. Each module also includes an interconnection unit having input and output ports to enable its associated module to be coupled to other modules.
The improved interconnection scheme of the present invention provides a plurality of double ring interconnection elements connecting modules. Each dual ring system consists of rings that rotate data in opposite directions. This double ring structure is called a carousel.
(carousel). Each carousel is connected to the module by an interconnecting unit. These digital interconnection units have buffers which are connected to the interconnection paths of the carousel. Additionally, each digital interconnection unit is connected to the main TDM of its respective module.
It also has a buffer connected to the bus. The first coupling means in each interconnect unit provides selective transmission between the interconnect buffers without using the main TDM bus of that module. The interconnect unit also includes second coupling means for selectively coupling the interconnect buffers to buffers associated with the TDM buses of their respective modules. Given a number of carousels interconnecting their respective modules, the device of the invention
Split the carousel interconnect structure as needed,
It operates to connect communication lines within the various modules using the shortest available segments.

〔Example〕

Hereinafter, the present invention will be explained in detail with reference to the drawings.

Although the following description is of an efficient and reliable interconnection scheme for interconnecting switching modules in a distributed architecture, it does not obscure the subject matter of the invention in unnecessary detail. To avoid this, detailed explanations of well-known technologies and components will be omitted.

One embodiment of the invention is used to interconnect multiple PBXs. It will be clear that the invention can be used with exchange modules other than PBXs. In the embodiment of the invention described herein,
PBX commercially available from Lorm, located in Clara.
is used. This will be briefly explained with reference to FIG. This PBX was published in the magazine "Business Communication Review".
Communication Review)” March-April 1979 issue 38
"New Rolm VLOBX" listed below the page
(The New Rolm VLCBX)” is fully explained in the article entitled “The New Rolm VLCBX”. This Lorm company
The PBX was transferred to the applicant and on July 24, 1978.
U.S. Patent Application No. 927,185, filed in 1999, is also described with a queuing mechanism for selecting between various classes of trunk lines. This PBX is
Known in the industry as a computerized private branch exchange (CBX), it includes a computer-controlled time division multiplexing private branch exchange. Analog signals appearing on the trunk interface buffer and telephone interface buffer are converted to digital signals by the well-known pulse coded (PCM) digital technique. Other known means are used to reduce errors in these digital signals (see US Pat. Nos. 3,877,022 and 3,999,129). Digital audio signals and other digital
Data is exchanged to enable connections between trunk exchange lines and trunk lines under the control of a central processing unit. Therefore, these PBXs include multiple ports that allow them to be coupled to trunk lines and other lines. The modules for the present invention also include digital interconnect termination units to enable the PBX to couple to other PBXs.

Before describing the present invention, a brief discussion of conventional architecture is helpful in understanding the advantages of the present invention. Two interconnection concepts are commonly employed: direct link and bus or ring. As shown in Figure 1a,
A direct link connects two modules directly. This connection is very cost efficient when used highly, and is especially useful when the traffic between the modules is uniform and stable. However, this type of connection does not allow for dynamic redistribution to overcome unbalanced traffic between multiple switching modules. 1st
In a bus or ring, such as the unidirectional link shown in Figure b, connected in a loop, the line load is equal to the amount of traffic between the modules since the connection is between any two nodes. It is not particularly affected by distribution. However, such a configuration requires many links for one call, and one
Even if only one module fails, communication between all modules may be interrupted. (As will be explained in more detail below, the carousel-like interconnection of the present invention shown in FIG. 5 has two rings that rotate in opposite directions. This allows the carousel to be divided dynamically, i.e. , the carousel interconnection emulates the link structure and operation of the link structure described above.More importantly, as will be explained later, the requirements for the ring shown in Figure 1b, i.e. The requirement for these calls to occupy time slots is eliminated.By providing an interconnect unit, calls can be effectively routed through the module without passing through the module's TDM bus. As noted in the prior art discussion, to increase capacity, multiple PBXs are often combined in a hierarchical structure, as shown in FIG. In FIG. 2, a plurality of lines, such as trunk switched lines and trunk lines 10, are coupled to multiplexer 11. In FIG. These multiplexers select and couple these lines to time division multiplexing (TDM) switches 14 in well-known fashion. one
3 multiplexers 1 to TDM switch 14
1, 12, and 13 are combined. Similarly, three multiplexers are coupled to each of the TDM switches 15 and 16, and a plurality of communication lines are coupled to the multiplexers. Interconnection between multiplexers coupled to the same TDM switch is done via that TDM switch.
For example, the interconnect between the lines coupled to multiplexers 11 and 13 is connected to TDM switch 14.
It is done through. It is clear that some provision must be made for the interconnection lines associated with different TDM switches. for example,
A higher order switch is required to connect the line connected to multiplexer 11 to the line connected to multiplexer 18. In this hierarchical architecture, a higher level TDM switch 17 is used which is coupled to each TDM switch 14, 15, 16. This TDM
The capacity of switch 17 must be such that it can handle the maximum number of lower level switches that can be coupled to it. Therefore, if switch 17 has the capacity to handle six lower level TDM switches, half of the capacity of switch 17 will be unused when only three lower level switches are used as shown in FIG. .
Similarly, if switch 17 is sized to handle three lower level switches, then
It is not easy to extend this system to handle more lower level TDM switches unless higher level TDM switches are used. In other words, this system cannot be easily extended;
Expensive hardware will be idle unless the capacity of the top switch is utilized to its full potential.

FIG. 3 shows another hierarchical architecture that has been implemented. Multiple multiplexers, such as multiplexer 21, are used to select multiple lines 20. These three multiplexers are coupled to a higher level multiplexer 22. These higher level multiplexers 22, 23, 24 are interconnected via a TDM switch 25. Therefore, in this architecture, no switching takes place within the multiplexers 22, 23, 24;
This is done only by the TDM switch 25. Connecting a line from multiplexer 21 to a line associated with multiplexer 26, for example, requires coupling to multiplexer 22 via TDM switch 25 and multiplexer 24. This architecture also has the same drawbacks as the architecture shown in FIG. That is, the system is inefficient unless the TDM switch is used to its maximum design capacity.

For many years, especially in central offices, a distributed architecture such as the one shown in Figure 4, called "multiple glazing", has been used in addition to the hierarchical architecture shown in Figures 2 and 3. I've been exposed to it. In FIG. 4, three switching units 30, 31, and 32 are shown. Each switch is coupled to multiple transmission lines, such as line 33. These exchanges are interconnected by direct links. For example, link 34 interconnects exchanges 30 and 31, and links 35 and 36
interconnects switch 32 with switches 30 and 31, respectively. These direct links allow very efficient use of the hardware, especially if the traffic between each switch can be determined. However, if only direct links are used, as is the case with normal telephone communications,
Fluctuations in call volume occur between exchanges, and a large number of direct links must be added to handle this variation.

Bus 37 is used to avoid the multiple direct links that would be required if any direct link were to overflow. It is clear that this bus 37 does not provide efficient coupling as a direct link, since it requires a port on every switch. For example, if bus 37 is handling a spill between switches 30 and 32, the port corresponding to bus 37 coupled to switch 31 remains unused. However, to achieve the same capacity between any two modules, the bus uses fewer ports than would be required using direct links. The overflow of call traffic from a direct link statistically follows bursts of traffic. The effect in a multiple glazing design is to interleave such overflows on the bus to increase bus utilization.

The distributed architecture shown in FIG. 4 is easier to scale than the hierarchical architecture shown in FIGS. 2 and 3. However, the particular distributed architecture shown in FIG. 4 has certain technical drawbacks, particularly when used in a commercial PBX.

Next, refer to FIG. This figure shows a computer controlled telephone switching system, which is the switching module used in the presently described embodiment of the invention. As you can see from this figure,
The switching interconnections between the lines 48 and the trunk lines connected to the telephone set are established by a central processing unit 57 under the control of a program stored in a memory 58, in particular by a TDM network controller 56. and a module TDM bus 50. Each trunk line 47 is coupled to an encoder 49 and a decoder 51 via a trunk interface 54 and a pair of multicore lines 43,44.
Similarly, each of lines 48 is coupled to an encoder 52 and a decoder 53 via a telephone interface 55 and a multicore line.

Although the present invention can be used with a variety of modules, the module shown in FIG. 6 constitutes a computer-controlled time-division multiplexing private branch exchange known under the trade name computerized private branch exchange (CBX). . In this module,
Analog signals appearing at trunk interface 54 and telephone interface 55 are converted to digital signals by encoders 49 and 52, respectively. These digital signals are passed to a suitable decoder 51 or 53 where they are converted back to analog signals via trunk interface 54 or telephone interface 55 over line 47.
or coupled to 48. All of these actions
This is done under the control of the TDM network controller 56. This TDM network controller 56 is controlled by a central processing unit (CPU) 57 according to a program stored in a memory 58.

A modular TDM bus 50 carries digital information signals (eg, voice signals from line 48), multidigit address signals identifying a particular trunk line, and interfaces 54 and 55, encoders 49 and 52, and decoders 51 and 53. Configure multi-core communication paths for timing and control signals that command sequential operations.

The operation of this module, particularly the PBX shown in FIG. 6, is well known, as is its structure, so a detailed explanation thereof will be omitted here.

In accordance with the objectives of the present invention, the modular TDM bus 50 is connected to an interconnect unit 60. This interconnection unit 60 is used to interconnect modules. This mutual coupling unit 60 has input ports I ₁ , I ₂ for coupling to output ports of other mutual coupling units, and output ports O ₁ , O ₂ for coupling to input ports of other mutual coupling units. include. Generally, this interconnect unit 60 includes first switching means for communicating signals between input port I ₁ and output port O ₂ and between input port I ₂ and output port O ₁ . These interconnections are modular TDM bus 5
This is done without binding through 0. Intercoupling unit 60 includes second switching means that can route signals from the interconnecting ports to modular TDM bus 50.

One embodiment of the interconnect unit 60 shown in FIG. 6 is shown in FIG. This embodiment includes a time division multiplexing (TDM) switch. That is, interconnect unit 60 is physically separate from modular TDM bus 50 and includes an interconnect unit TDM bus 70 that is separate from it. Mutual coupling port I ₁ ,
I ₂ is coupled to bus 70 via decoders 63 and 65, respectively. Output ports O ₁ and O ₂ are encoders 64,
66 to bus 70, respectively. The modular TDM bus 50 is interconnected via a TDM interface switch 62.
Coupled to TDM bus 70. The unit of FIG. 7 can be constructed using well-known TDM components, such as those used in the PBX of FIG.

The signal at port I ₁ can be routed directly to port O ₂ via bus 70 without the need to use time slots on bus 50. Also, calls received on port _I2 are similarly routed to bus 5.
can be sent to port _O1 via bus 70 without requiring a time slot on O1. Ports I ₁ and I ₂ , O ₁ and O ₂ are communicated via switch 62 to modular TDM bus 50 in a well-known manner.

Next, another embodiment of the interconnection unit shown in FIG. 8 will be described. This interconnect unit is also coupled to the modular TDM bus 50. Mutual coupling input ports I ₁ and I ₂ are decoders 63 and 6
output ports O ₁ and O ₂ are coupled to encoders 64 and 66, respectively. Decoder 6
The signal from 3 is sent to buffer 71. This buffer 71 is coupled to bus 50 and multiplexer 74. The signal from bus 50 is module
It can be coupled to the TDM buffer 72.
This buffer 72 is coupled to another input terminal of multiplexer 74. Multiplexer 74, which may be comprised of a conventional digital multiplexer, selects its contents from buffers 71 or 72 and sends signals representative of those contents to buffer 79.
give to This buffer 79 is coupled via encoder 66 to port _O2 . The signal from decoder 65 is sent via buffer 78 to bus 50 or multiplexer 77. Multiplexer 77 also receives input from buffer 80, which receives a signal from bus 50. Multiplexer 77 selects the contents of buffer 78 or 80 and sends a signal representing the selected contents to buffer 73 and encoder 64.
to output port O ₁ via.

Next, the operation will be explained assuming that port _I1 is receiving a call. This call is sent to bus 50 (via buffer 71) or to output port _O2 (via buffer 71, multiplexer 74 and buffer 79). Similarly, a call received at input port I ₂ can be routed to bus 50 or to output port O ₁ . The signal from bus 50 is sent via buffer 80, multiplexer 77 and buffer 73 to output port _O1 . Similarly, calls from bus 50 are routed through buffer 72, multiplexer 74, and buffer 79 to the output port.
Sent to _O2 . Here, it should be noted that port I ₁
and _O2 and between ports _I2 and _O1 do not pass through bus 50 and therefore do not require time slots on this bus. 8th
The illustrated interconnection unit may consist of well-known digital buffers, multiplexers, encoders and decoders.

The buffers and other lines of this interconnection unit, in the embodiment described herein, have a capacity to handle one digital word, with each buffer capable of storing eight digital words. The storage capacity of each buffer is such because time division multiplexing (of eight time slots) is used in the interconnect lines between the interconnect units.

Referring now to FIG. 5, three modules, each of which is equivalent to the module of FIG.
42. In the carousel interconnect configuration used between modules, each module's output port O ₂ is coupled to the input port I ₁ of the next module, and the output port O ₂ of the last module 42 is coupled to the input port I 1 of the first module 40. Coupled to port I ₁ . Similarly, output port O ₁ is coupled to input port I ₂ of the next module, and output port O ₁ of the first module is coupled to input port I ₂ of the last module. As is clear from the figure and the above description, the two buses carry signals in opposite directions.

Generally, in the embodiments described herein, the sixth
All of the central processing units (CPUs) 57 shown are coupled to a common hub to provide a common channel interface signal. This is in contrast to systems that use a main CPU to control the CPUs within each module. Race conditions are avoided by using hubs. The lines interconnecting each module consist of a 16-bit bus. Twelve of these 16 bits are used for the converted analog signal, and the remaining 4 bits are used to indicate whether the channel can be used.
The status of the call path through a switching module is determined by having the coder at the originating end send the identifier of the originating module over the idle channel. This allows each exchange module to monitor idle links. For example, if only call 1 on link 45, indicated by the dashed line in FIG.
Module 42 sends its own identifier via its output port _O2 . Similarly, module 41
It also sends its own identifier through its output port _O1 . In this case, the module 40 in FIG.
It retransmits (transparently) the identifier of the module 42 received at its input port I ₁ via its output port O ₂ and transmits the identifier of the module 41 received at its input port I ₂ via its output port O ₁ . resend. Therefore, any module, e.g. module 40 or 41, has its input ports I ₁ and
Link 4 by examining the identifier received by I ₂ and identifying which module is at play.
6 or whether the link between modules 40 and 41 is idle all the way to module 42. This is because the module's own identifier on the link indicates that the module is being played. After testing, a given module can seize a certain link by sending a command to the control hub. This approach includes
This has the advantage that state information can be continuously updated dynamically. Furthermore, in the link acquisition control sequence, if the terminating module requests intervention, a channel test can be performed using the terminating module's identifier. The acquiring module informs the intermediate module that the channel is now inactive and that upon its release it will be inactive.

Hierarchical fixed routing tables are employed to explore interconnections between modules. In this table, the lowest entry representing most call flows is actually a direct link, such as call 1, shown by dashed line 45 between modules 41 and 42 in FIG. Additionally, the state of each terminating channel in a given module is stored within the module containing that channel. The extent to which the channel activity extends along the carousel is dynamically determined or ascertained at the time of the desired acquisition as described above. This hierarchical approach to routing can guarantee the lowest level path to make a connection. When a call lower in the hierarchy drops, a hole is created that is filled by shifting down calls at a higher level. It is well worth dynamically relocating calls to fill those holes.

For example, the dashed line 4 between modules 41 and 42
Assume that all direct links shown at 5 are in use. In this case, additional calls are routed between modules 42 and 41 via module 40, as indicated by dashed line 46. However, broken line 45
If the link indicated by is determined to be available, the calls are rerouted through this direct link and removed from longer routes. This dynamic rerouting action frees the direct links between modules 40 and 42 and between 40 and 41. It should be noted here that the call is from module 42 to module 4.
1, as shown by dashed line 46, ie via module 40, a TDM bus within module 40 is not required.

The interconnection configuration described above provides standardized hardware for each module that can create links, buses, or rings. Additionally, when connected using a carousel architecture, the carousel can be dynamically divided to effectively create directly linked (through interconnecting modules) and cascaded links. There is no need to use time slots in these modules.

Another important feature of the interconnection arrangement of the present invention is that it provides a more reliable interconnection compared to standard ring architectures. Assume now that module 40 and its associated interconnect unit are not operating. Even in this case, a direct link exists between modules 41 and 42. Note that in a typical unidirectional ring configuration, such as that shown in FIG. 1b, if any one module becomes inoperable, all communication between the modules is disabled. Even with the hierarchical architecture shown in FIGS. 2 and 3, failure of a higher-level TDM switch can disrupt all communication between modules.

The interconnection arrangement of the present invention is particularly advantageous when used with three exchange modules as shown in FIG. If there are three modules,
Direct links exist between each module. However, the principles of the invention are applicable to any number of exchange modules. In FIG. 9, the switching module and its associated interconnect unit are shown as switching units. For example, the combination of switching module (PBX) and interconnection unit 60 in FIG.
It is shown as 2. This figure shows 4
Switches 92, 93, 94, 95 are shown. These exchanges are connected in the same carousel manner as the exchange shown in FIG. For example, one output port of switch 92 is coupled to an input port of switch 93 via line 99, and one output port of switch 93 is coupled to one input port of switch 92 via line 96. The other output port of switch 92 is coupled to the output port of switch 95 via line 97. Exchanges 93 and 94 and 94 and 95 are also coupled in the same manner as above.

Direct links exist between exchanges 92 and 93, between 94 and 95, and between 95 and 92;
There is no direct link between them. Some direct links between these switches cease to exist if more than two switches are used. However, in that case there are two paths through the interconnect unit associated with each switch, without using time slots within the module. For example, considering the possible paths between switches 92 and 94, one connection
It can be seen that the connection is made via switch 95 and another connection is made via switch 95.

Although only three and four exchange modules have been described above, it will be clear to those skilled in the art that the principles of the present invention can be extended and used with a larger number of exchange modules. It will also be apparent that the interconnect unit can be implemented in many different configurations.

Comparing the interconnection configurations shown in FIGS. 5 and 9 with the conventional configuration shown in FIG. 4, the modules and interconnection units 40, 41, 42
It should be noted that the line between serves as both a direct link and bus 37 of FIG. This is because these lines are also used for overflow.

However, unlike the configuration shown in Figure 4, bus 3
7 has no unused ports.

〔Effect of the invention〕

The interconnection configuration of the present invention described above uses a general distributed architecture, without using time slots within each module.
Signals can be effectively conveyed through multiple modules. The interconnection unit is capable of transmitting signals in a carousel or ring. Therefore, according to the present invention, an interconnection architecture for a time division communication system that is highly reliable and advantageous in terms of cost is provided.

[Brief explanation of the drawing]

FIG. 1a is a block diagram illustrating a direct link interconnection between switching modules according to the prior art; FIG.
FIG. 2 is a block diagram showing a ring connection between exchange modules according to the prior art, FIG. 2 is a block diagram showing an example of a hierarchical interconnection configuration according to the prior art, and FIG.
FIG. 4 is a block diagram showing another example of a hierarchical interconnection configuration according to the prior art, FIG. 4 is a block diagram showing an interconnection configuration using a distributed architecture according to the prior art, and FIG. 5 shows modules and their related components. FIG. 6 is a block diagram illustrating the interconnects used in the present invention to interconnect interconnect units.
Figure 7 is a block diagram showing the exchange module and its interconnection to the interconnection unit;
8 is a block diagram of an interconnect unit using a multiplexer; FIG. 9 shows the carousel interconnection of FIG. FIG. 2 is a block diagram illustrating an example of application to an environment consisting of interconnected units. 40, 41, 42...Module and interconnection unit, 62...TDM interface
Switch, 63, 65... Decoder, 64, 66...
... Encoder, 70 ... Mutual coupling TDM bus, 92 -
95...exchange machine.

Claims (1)

[Claims] 1. A plurality of switching modules, each of which interconnects a plurality of communication lines, such as trunk exchange subscriber lines and trunk lines, by time-division multiplexing and digital transmission of signals on the communication lines. A switching module interconnection device in a time-division telephone switching system configured to aggregate signals and switch the digital signals between the communication lines, characterized by comprising the following components: (a) a plurality of interconnection units, one in association with each of said exchange modules; Each of the units has first and second input ports for receiving signals from other interconnect units and first and second output ports for communicating signals to other interconnect units, and , coupled in signal communication with the associated exchange module. (b) connecting a second output port of the first interconnect unit to a first input port of the second interconnect unit; and connecting a second output port of the second interconnect unit to a third interconnect unit; a first interconnect for connecting to a first input port of the coupling unit and likewise for connecting a second output port of the last interconnection unit to a first input port of said first interconnection unit; line. (c) connecting a second input port of said first interconnection unit to a first output port of said second interconnection unit; a second input port for connecting to the first output port of the third interconnect unit, and so on for connecting the second input port of the last interconnect unit to the first output port of the first interconnect unit; interconnection lines. 2. each of said interconnection units has first switching means for switching a signal from said first input port thereof to said second output port thereof and one of said associated switching modules; a second input port for switching a signal from the second input port to the first output port and its associated one of the switching modules;
switching means, the connection between the first input port and the second output port by the first switching means and the connection between the second input port and the first output port by the second switching means. 2. The switching module interconnection device according to claim 1, wherein the connection between the switching module and the output port of the switching module is made without going through the associated switching module.