scispace - formally typeset
Search or ask a question

Showing papers on "Network switch published in 1992"


Patent
18 Dec 1992
TL;DR: In this paper, a variable is maintained at the network interface device which stores information relating to the success or failure of transmission of packets in the channel, and if the variable is less than a certain limit, the delay ζ is selected according to a first basic bandwidth allocation protocol.
Abstract: In a local area network segment, the bandwidth in the channel between a network interface device, e.g., a port, switching hub, router, bridge or server, and one or more associated station is shared as follows. At the network interface device, a variable is maintained which stores information relating to the success or failure of transmission of packets in the channel. If the channel is sensed idle, the network interface device transmits the packet with probability 1. If the channel is sensed busy, the network interface device senses the channel for an end-of-carrier signal and, thereafter, waits a period of time equal to the interframe gap g. In addition, after an additional delay period ζ, the network interface device reattempts to transmit the packet in the channel. If the variable maintained by the network interface device is less than a certain limit, the delay ζ is selected according to a first basic bandwidth allocation protocol. If the variable is greater than or equal to the limit, the delay is selected according to a second basic bandwidth allocation protocol.

69 citations


Patent
22 Jan 1992
TL;DR: In this paper, a communication network uses a network switch number (NSN) and a class-of-service classification (COS) code to route calls through the network.
Abstract: A communication network uses a network switch number (NSN) and a class-of-service classification (COS) code to route calls through the network. When a call is received at an originating switch of the network, the dialed number is used to select the NSN which identifies a network terminating switch which connects to the called party. The network uses received call signaling information and stored trunk information and possibly network database information to determine the COS code. The COS code is also used at the terminating switch to identify the type of service trunk to connect to the called party. In another feature, only the network switch number is used to route the call over the network. In accordance with yet another feature, the COS code may be used to select specific network trunks without requiring the specification of the NSN.

59 citations


Patent
18 Sep 1992
TL;DR: In this paper, a multi-tone signal processor and associated control is provided such that when signaling is detected, such signaling is validated against a provisioned list, and a valid signal is reported to a central control of the network switch, which takes action in response to the signal.
Abstract: A network monitor and associated control system is provided to monitor a toll network for customer signaling. A multi-tone signal processor and associated control is provided such that when signaling is detected, such signaling is validated against a provisioned list. A valid signal is reported to a central control of the network switch, which takes action in response to the signal.

58 citations


Proceedings ArticleDOI
01 May 1992
TL;DR: The authors provide an overview of the architecture of the Xunet 2 testbed, which has three principal components: routers, switches and network management systems.
Abstract: Xunet 2 is an experimental wide-area network that serves as a testbed for research on data communications techniques. The authors provide an overview of the architecture of the Xunet 2 testbed. Xunet 2 has three principal components: routers, switches and network management systems. Long-haul transmission between them will initially use DS3 facilities. Performance objectives for the backbone have been set to cover a wide range of transmission speeds. Each router attaches a local area network to the backbone network. The backbone network provides virtual circuit connectivity between the routers. Circuit setup and teardown is done by a control computer associated with each switch, and the network as a whole will be managed by a network management station in the network operations center. >

53 citations


Journal ArticleDOI
TL;DR: The issues involved in bridging 802.5 (token-ring) and 802.3 (Ethernet) networks at the media access control (MAC) layer are discussed and an approach incorporated in the IBM 8209 LAN bridge for resolving the incompatibilities is described.
Abstract: The issues involved in bridging 802.5 (token-ring) and 802.3 (Ethernet) networks at the media access control (MAC) layer are discussed. The issues concern incompatibilities between the two networks in six areas: routing schemes, station addresses, frame formats, and local area network speeds. An approach incorporated in the IBM 8209 LAN bridge for resolving the incompatibilities is described. The IBM 8209 LAN bridge, a two-port MAC bridge, supports source routing on the token ring/802.5 side and transparent bridging on the Ethernet/802.3 side. Recently proposed changes in the token-ring source routing standards are discussed. >

11 citations


Patent
29 Oct 1992
TL;DR: In this article, a method performs arbitration between network devices (3-9) connected to one of a first network segment (10,70) and a second network segment(30,80), where the first segment is connected to the second segment by a long interconnection medium (41,42,63).
Abstract: A method performs arbitration between network devices (3-9) connected to one of a first network segment (10,70) and a second network segment (30,80), where the first network segment (10,70) is connected to the second network segment (30,80) by a long interconnection medium (41,42,63) over an extended distance. When one or more network devices (3-9) connected to the first network segment (10,70) senses the network is free and desires control of the network for a data transfer, each of these network devices (3-9) will assert a first network control signal. When the first network control signal is first asserted, this marks the beginning of an arbitration period for the first network segment (10,70). The first network control signal is forwarded to the second network segment (30,80). Upon the second network segment (30,80) receiving the first network control signal, any arbitration currently in progress upon the second network segment (30,80) is aborted. Upon completion of the arbitration period and provided arbitration on the first network segment (10,70) has not been aborted, one of the network devices (3-9) connected to the first network segment (10,70) is granted control of the network.

5 citations


Patent
09 Jan 1992
TL;DR: In this paper, a voltage switching arrangement for an electrical power control system including a plurality of switches for connecting and disconnecting voltage from the input to a power transfer network which provides power for a load is presented.
Abstract: A voltage switching arrangement for an electrical power control system including a plurality of switches for connecting and disconnecting voltage from the input to a power transfer network which provides power for a load. A time delay circuit prevents the voltage connected to the input to the power transfer network from being disconnected for a predetermined time period after the control system has been turned off. The power transfer network switches from an on-mode to the off-mode during said time delay period. The voltage connecting switches include an electromechanical DC relay, and the time delay circuit includes a capacitor cooperating with the relay to maintain the relay energized for the time delay period after the control system has been turned off. Safeguards are provided to prevent simultaneous connection of voltage in the forward and reverse directions to the input of the power transfer network.

5 citations


Patent
17 Mar 1992
TL;DR: In this paper, a back-bone network using bridges 1-4 as components is logically configurated to implement source routing processing without a limit on LAN configuration and the increase in bridge processing time.
Abstract: PURPOSE:To prevent a limit on LAN configuration and the increase in bridge processing time by regarding an equipment as a network so as to provide a network number to the equipment and configurating a back bone network using bridges as components logically and implementing source routing processing. CONSTITUTION:A port is regarded as a bridge operated as a MAC bridge and an equipment is regarded as a network to obtain physical configuration as shown in figure (a), and a back bone network using bridges 1-4 as components is logically configurated logically as shown in figure (b). Thus, a network number is given to the back bone network to implement source routing processing. Then the processing of each bridge in the back bone network is to execute the bridge function between its own connection ring and the back bone ring. Thus, the source routing processing is implemented without a limit on LAN configuration and the increase in bridge processing time.

4 citations


Patent
21 Sep 1992
TL;DR: In this article, a line adaptor connecting to lines, a low-speed data switch, a high speed data switch and a processing unit implementing changeover control of the switches in the exchange is provided.
Abstract: PURPOSE:To remarkably improve the man-machine interface in the maintenance operation by providing a line adaptor connecting to lines, a low speed data switch, a high speed data switch and a processing unit implementing changeover control of the switches in the exchange. CONSTITUTION:Data from a telephone set 2, a low speed data terminal equipment 3 and a LAN (high speed data) terminal equipment 12 are multiplexed to data for one line 16 with a multiplexer/demultiplexer 15 and the line 16 is accommodated to a line adaptor section 20A of an electronic exchange 1. The multiplexed data are demultiplexer by the line adaptor section 20A, voice and low speed data are switched by a low speed data switch 21 and high speed data are switched by a high speed data switch 22. Then the low speed data switch 21 and the high speed data switch 22 are subjected to changeover control by a processing unit 33. Thus, the man-machine interface at maintenance operation is considerably improved.

3 citations



Journal ArticleDOI
15 Nov 1992-EPL
TL;DR: A network which has the ability to retrieve not only a sequence of patterns but also their rhythms and the duration of each resting period can be independently adjusted to an arbitrary length is proposed.
Abstract: We propose a network which has the ability to retrieve not only a sequence of patterns but also their rhythms. The network is a deterministic system in which neither delay nor noise is required. The elements constituting the network are bursting neurons modelled in a relatively simple way. In the retrieval process, the network at first exhibits a retrieval pattern for a finite time. It is then followed by a period in which all neurons are quiescent until the network switches to the next retrieval pattern. The duration of each pattern and the duration of each resting period can be independently adjusted to an arbitrary length. We confirm the excellent retrieval performance of the network by numerical simulation.

Journal ArticleDOI
TL;DR: In this article, an optically amplified distributive switch network based on a simplex passive optical network is described, and the capacity is 2.488 Gbit/s and the loss budget is 72 dB.
Abstract: An optically amplified distributive switch network based on a simplex passive optical network is described. The capacity is 2.488 Gbit/s and the loss budget is 72 dB, sufficient for more than 1000 customers to be served with an individual capacity of 2 Mbit/s. Effectively only a single stage of amplification is used with no optical filtering.

Proceedings ArticleDOI
13 Sep 1992
TL;DR: The switched multi-megabit data service (SMDS) is a high-performance, public, packet-switched data service and a customer network management (CNM) service has been defined in conjunction with SMDS, based on SNMP.
Abstract: The switched multi-megabit data service (SMDS) is a high-performance, public, packet-switched data service. Offered as public network equivalent of a MAC-level service, the datagram transport offered by SMDS is easily integrated into networking protocol architectures using local area network (LAN) or metropolitan area network (MAN) technologies such as Ethernet and fiber distributed data interface (FDDI). A customer network management (CNM) service has been defined in conjunction with SMDS, based on SNMP. SMDS and the accompanying CNM service were successfully demonstrated. These demonstrations and how SMDS and its CNM service can be integrated into existing corporate networks are discussed. >

Patent
18 Nov 1992
TL;DR: In this article, an automatic programmable control unit (APCU) is driven by a clock using a protocol registered in a memory and task sequences in a second memory, where the tasks are executed by an automatic dialler, which switches the telephone channel in use at that time.
Abstract: The automatic programmable control unit (3) is driven by a clock (6) using a protocol registered in a memory (4), and task sequences in a second memory (5). The interface can be a man-machine interface (2), a radio receiver (1) or a communications interface (2). The tasks are executed by an automatic dialler (7), which switches the telephone channel (8) in use at that time. The system is particularly suited to producing a three-way call and reversing charges automatically. ADVANTAGE - Less complex, with more reliable access.

Patent
25 Sep 1992
TL;DR: In this paper, the broadcast system of the switching network provides for the broadcast or multicast of interface switch status broadcasts and interface switch event broadcasts and the broadcast messages are sent into the switch network in a non-broadcast manner until they reach the fold point.
Abstract: The broadcast system of the switching network provides for the broadcast or multicast of interface switch status broadcasts and interface switch event broadcasts. An interface switch feature processor initiates a broadcast message made up of two bytes of data on a first channel, with particular command codes to identify the type of data contained within the broadcast message, paired with another two bytes of data in the same frame on a second channel. Broadcast messages are sent into the switching network in a non-broadcast manner until they reach the fold point of the switch network. From the fold point the broadcast messages are broadcast on every available broadcast channel directed out of the switch network towards the interface switches.

Patent
04 Aug 1992
TL;DR: In this paper, the authors propose to optimize a path of a clock supply system and the arrangement and design of a switch circuit group by employing the plural delay means, which is defined as 20-1-20n delaying a clock signal CLKin for a prescribed time and applying the delayed clock signal to next-stage of switch circuit groups 10- 1-10n.
Abstract: PURPOSE:To optimize a path of a clock supply system and the arrangement and design of a switch circuit group by employing plural delay means. CONSTITUTION:Plural delay means 20-1-20-n delaying a clock signal CLKin for a prescribed time and applying the delayed clock signal to next-stage of switch circuit groups 10-1-10-n are provided to a multi-stage switch network in which the plural switch circuit groups 10-1-10-n which fetches an input data from plural channels based on a clock signal CLKin and selects and outputs a channel of an output destination according to the content of the above input data are connected in cascade.

Patent
08 Dec 1992
TL;DR: In this paper, the authors proposed a protocol to prevent processing performance from being degraded even when relay of plural protocols and relay in a data link layer take place simultaneously by using transceivers 33-36.
Abstract: PURPOSE:To prevent processing performance from being deteriorated even when relay of plural protocols and relay in a data link layer take place simultaneously. CONSTITUTION:A MAC bridge interface 13 having a MAC bridge function and protocol interfaces 14-16 implementing routing in response to each specific protocol (TCP/IP, OSI, DECnet) are connected to a branch line LAN 30 by using transceivers 33-36. The connector is provided with an internal bus 11 being a data relay line of the interfaces 13-16 and a trunk line LAN interface 12 controlling data transfer between the internal bus 11 and the trunk line LAN 20. A microprocessor muP built in each of the interfaces 13-16 implements specific relay processing respectively.

Patent
Eiji Kamagata1, Yasuro Shobatake1
08 Jan 1992
TL;DR: In this paper, the authors provide a simplified cell switch testing framework for a cell switch and a network in which the cell switches are interconnected in a form of a multi-step configuration.
Abstract: A cell switch and a cell switch network are provided with a simplified cell switch testing such that the testing processes can be carried out quickly and easily. The cell switch includes a dummy cell generation circuit for generating dummy cells in correspondence to the output transmission paths; and an output control circuit for selectively outputting the dummy cells to the output transmission paths when a test of the cell switch is indicated by an externally provided control signal. In a cell switch network in which the cell switches are inter-connected in a form of a multi-step configuration, the cell switches belonging to each step in the multi-step configuration of the cell switch network is tested sequentially, by supplying the control signal to each of the cell switches belonging to each step in the multi-step configuration of the cell switch network sequentially, in a reverse order of steps in the multi-step configuration of the cell switch network.

Proceedings ArticleDOI
10 May 1992
TL;DR: Describes the design of an interconnection network switch VLSI for parallel digital signal processing (DSP) systems that features 6*400 Mb/s bidirectional communication ports and 1.44-Gb/s internal switching capability for realizing efficient packet-switched communication among many processing elements.
Abstract: Describes the design of an interconnection network switch VLSI for parallel digital signal processing (DSP) systems. This VLSI features 6*400 Mb/s bidirectional communication ports and 1.44-Gb/s internal switching capability for realizing efficient packet-switched communication among many processing elements. To achieve this performance with existing CMOS technology, the transport line transfers 8 b in parallel at 50 MHz; input data are first multiplexed up to 64 b and switched at 25 MHz. An automatic routing function is implemented on each switch using a distributed routing table scheme. A high-level behavioral description based CAD (computer-aided design) system called PARTHENON is used to design the functions and logic circuits of this VLSI. The suitability and effectiveness of PARTHENON for switch design are also shown in terms of parallel operation and finite state machine design. >

01 Jan 1992
TL;DR: LesLeslie, McAuley, and Mullender as mentioned in this paper explore how emerging 64-bit processors can be used to implement shared address spaces spanning multiple machines, and conclude that DAN-based architectures allow the exploitation of shared memory on a wider scale than just a single (multi)processor.
Abstract: In a recent issue of Operating System Review, Hayter and McAuley [1991] argue that future high-performance systems trade a traditional, bus-based organization for one where all components are linked together by network switches (the Desk-Area Network). In this issue of Operating System Review, Leslie, McAuley and Mullender conclude that DAN-based architectures allow the exploitation of shared memory on a wider scale than just a single (multi)processor. In this paper, we will explore how emerging 64-bit processors can be used to implement shared address spaces spanning multiple machines.

Proceedings ArticleDOI
14 Jun 1992
TL;DR: The software design and implementation of a packet switching node for a data communication network in China, the first Chinese-made packet switch based on international standards, compatible and interoperable with similar Western products is discussed.
Abstract: The software design and implementation of a packet switching node for a data communication network in China is discussed. Network management is emphasized. A scheme for centralized and localized network management is also briefly introduced. It is the first Chinese-made packet switch based on international standards, compatible and interoperable with similar Western products. The packet-switching node was connected to the data network in 1989. Operational results are discussed. >