Abstract: A switch for transmitting data from a first device to a second device includes a port in communication with the second device, queues which store slices of data received from the first device, and circuitry for selectively outputting a slice of the data from at least one of the queues to the port for transmission to the second device.

Abstract: A packet transfer device that can be easily realized even when a number of input ports is large. Each input buffer temporarily stores entered packets class by class, and outputs packets of a selected class specified by the control unit, while the control unit determines the selected class of packets to be outputted from the input buffers according to a packet storage state in the packet storage units of the input buffers as a whole for each class. Each input buffer can temporarily store entered packets while selecting packets to be outputted at a next phase, and the control unit can specify packets to be selected in the input buffers according to an output state of packets previously selected in the input buffers as a whole. Packets stored in the buffer can be managed in terms of a plurality of groups, and each packet entered at the buffer can be distributed into a plurality of groups so that packets are distributed fairly among flows.

Abstract: A method for measurement-based connection admission control in a packet data network where traffic flow of arriving data packets is forwarded via a switch with a delay depending on the priority of a queue associated to the data packets of a respective connection and the amount of prevailing traffic. The method comprises providing an estimated maximal rate envelope (Rk) of the traffic flow based on traffic measurements; approximating the envelope to a piecewise linear traffic constraint curve (A(Ik)); modifying the piecewise linear traffic constraint curve such that the curve assumes a concave shape; and checking whether a predetermined delay bound for the priority queue of a new connection requesting to be admitted, and the delay bounds for all queues having a lower priority than the requesting new one, are not violated, and granting the requested new connection, if the predetermined delay bounds are not violated.

Abstract: A method and system is disclosed for providing packetized voice (PV) call admission control (CaC), which delivers uninterrupted voice network services concurrently with data network services. Utilization of PV CaC prevents voice service interruption caused by congestion in PV networks, which can occur when the network has insufficient bandwidth resources to ensure the Quality of Service (QoS) of all calls. As the network's bandwidth capacity diminishes and approaches a predefined congestion onset threshold, an appropriate congestion indicator is “piggybacked” on the existing network service (or services) offered by the service provider. For PV, this piggybacked congestion notification will prevent additional calls from occurring by using a fast busy, or some other, signal to alert users (attempting to place a new call) that the service is unavailable to them. Upon the abatement of the diminished network capacity (i.e.

Abstract: Quality of Service (QoS) support is provided by means of a Tiered Contention Multiple Access (TCMA) distributed medium access protocol that schedules transmission of different types of traffic based on their service quality specifications. In one embodiment, a wireless station is supplied with data from a source having a lower QoS priority QoS(A), such as file transfer data. Another wireless station is supplied with data from a source having a higher QoS priority QoS(B), such as voice and video data. Each wireless station can determine the urgency class of its pending packets according to a scheduling algorithm. For example file transfer data is assigned lower urgency class and voice and video data is assigned higher urgency class. There are several urgency classes which indicate the desired ordering.

Abstract: A method for subcarrier allocation and loading for a multi-carrier, multi-subscriber system is described. At least one cluster in a first and second set of clusters of subcarriers is associated for use in communication with a first and second subscriber, respectively. Then, for each cluster associated for use in communication with the first subscriber and the second subscriber, usage of that cluster is multiplexed between the first subscriber during a first time division and the second subscriber during a second time division.

Abstract: In a system for switching high-capacity and variable length packets from m input ports to n output ports, k Banyan networks are connected into a ring by means of at least n parallel lines, said input ports are connected to inputs of said Banyan networks in a dispersed manner by means of m input modules, misroute tag check parts are connected to outputs of said Banyan networks as well as to inputs of n output modules whose outputs are connected to said output ports. Each of plural unit switches constituting a Banyan network is constructed such that when an input packet cannot be sent to a desired output, the relevant packet is outputted to a non-occupied output of the relevant Banyan network. When the misroute tag check part finds that the misroute tag is set, the relevant packet is sent to a next stage Banyan network, but when the misroute check tag is not set, the relevant packet is outputted to an output module connected to a desired output.

Abstract: A way of managing bandwidth in a network supporting variable bit rate is provided. An apparatus is provided that comprises an interface to transmit data to a receiving device. The apparatus comprises a controller that is communicatively coupled to the interface, the controller to detect a bit rate change event and transmit a first portion of the data using reserved bandwidth and a second portion of the data using unreserved bandwidth in response to detecting the bit rate change event.

Abstract: Methods and systems for controlling scheduling in a packet switching node in a network are provided which enable the scheduling of packets from different sources in an earliest deadline first order. The packets are assigned timestamp deadlines and placed in input queues. The timestamps are determined according to maximum delay or minimum throughput quality of service requirements specified for the packets. The packets are scheduled in the earliest deadline first order in an output packet store. The packet closest to its timestamp deadline is selected from the output packet store by using an index.

Abstract: A queue control method used in a relay apparatus which includes a plurality of ports for sending and receiving and performs relay of flows between networks is provided, in which the queue control method includes the steps of: performing queue control in which queue control parts which are connected to one or more ports enqueue flows from the one or more ports to a plurality of queues each having different priority, and dequeue the flows; switching flows dequeued from the queue control parts and sending the flows dequeued from the queue control parts from the ports; and sending and receiving control signals between the queue control parts such that queue control for arbitrary queue in the queue control parts is performed.

Abstract: An integrated QoS control system is provided that transmits, in real time, the stream data between a bandwidth-guaranteed network and a bandwidth-not-guaranteed network. The QoS manager 102 records the remaining bandwidth of the bandwidth-guaranteed network 2 captured by the remaining bandwidth table capturer 103 and the remaining bandwidth of the bandwidth-not-guaranteed network 1 calculated with traffic information notified by the network status monitor 107 on the use bandwidth registration table 104 (for comprehensively managing the bandwidth of an integrated network). The Qos manager 102 converts a QoS parameter received via the QoS request receiver 106 into a traffic parameter and controllably adapts the value of the traffic parameter to a service quality required by the application 109.

Abstract: A new format for asynchronous transfer mode defines a priority unspecified bit rate format; which uses the same memory structure as that used by the existing UBR and CBR traffic protocols. This format is also used for a loopback operation.

Abstract: A method of controlling packet traffic in an IP network of originating, receiving and intermediate nodes to meet performance objectives established by service level agreements. Traffic statistics and performance data such as delay and loss rates relating to traffic flows are collected at intermediate nodes. A control server processes the collected data to determines data flow rates for different priorities of traffic. A static directory node is used to look up inter-node connections and determine initial traffic classes corresponding to those connections. The rates are combined with the initial traffic classes to define codes for encoding the headers of packets to determine their network priority.

Abstract: A method for controlling a flow of packet data in a memory management unit of a network switch fabric is disclosed. A first portion of a data packet is received at a port on an ingress bus ring of the network switch fabric. A class of service for the data packet is determined based on the first portion and the portion is stored in a packer RAM of the port based on the class of service. Subsequent portions of the data packet are stored in the packer RAM. Once the predetermined number of portions have been received, the predetermined number of portions is sent to a packet pool RAM. A reference pointer to a first predetermined number of portions is sent to a transaction queue once an end of packet is detected and an egress scheduler detects a presence of a ready packet in the transaction queue and notifies an unpacker of the ready packet. The unpacker puts the ready packet into a FIFO and the ready packet is sent to an ingress/egress module.

Abstract: A queue sort selection part 1 uses as a search key a sort tag included in an inputted packet to decide a queue to be allocated to the packet. A utilization computation part 7 computes utilization of each queue by computing the packet storage number of the each queue. An allocation coordinating part 8 computes a new allocating result in which the inputted packets are divided and allocated among all queues. An information setting storage part 2 holds information used for allowing the queue sort selection part 1 to decide a queue to be allocated to the inputted packet. The packet stored into the queue held in a holding storage part 4 is transmitted in the stored time order. This can prevent deterioration of transmission quality such as a delay characteristic, packet loss percentage and a jitter characteristic due to packet concentration onto one queue and realize enhancement of the resource utilization of all queues and finer priority control.

Abstract: A communication bus snooper switch includes an inbound cell queue coupled to receive ATM cells from a number of ATM physical layer interfaces across a common inbound bus. An out-bound cell queue (which may be the same queue as the inbound cell queue when a dual port queue is used) is coupled to provide the ATM cells to separate ATM termination points according cell address information across separate out-bound busses. Each of the out-bound busses is associated with one of the ATM termination points. The snooper switch is configured to operate as an ATM bus master when communicating with the physical layer interfaces and as a ATM bus slave when communicating with the ATM termination points. During transmit operations, cells from the various ATM termination points are queued in corresponding transmission cell queues within the snooper switch and, thereafter, provided to the ATM physical layer interfaces according to an arbitration scheme implemented at the snooper switch.

Abstract: A communication system sorts packets or messages at a base station such that differential delays and absolute delays are reduced, further resulting in less needed memory at the packet selector and reduced jitter. The sorting is based on a packet identifier that is associated with each packet or message, which packet identifier includes data specific to the packet.

Abstract: A routing system is disclosed for routing communication events. The system comprises at least one data queue for queuing incoming events; at least one interaction mechanism for enabling bi-directional communication with authors of events in queue; and a processor for processing events in queue according to routing rules. In a preferred embodiment authors of the communication events are solicited through the at least one interaction mechanism to make a value contribution in exchange for advancement in the data queue.

Abstract: In the communication method according to the invention, the user data are passed on iteratively between subscribers (A, B), with the user data communication being the stimulus for second-order communication (explorer communication), by means of which network structure data are determined and/or refreshed. The explorer data are transmitted iteratively on the same or on a different communications infrastructure as the user data, with the data being passed on in unchanged or processed form during transmission. The explorer data contain explicit requests relating to network structure data and can thus actively initiate the process of obtaining network structure data. Suitable communication paths (a, b, c) for passing on user data iteratively are calculated on the basis of the network structure data which are already available locally, in particular on the basis of previous communications processes, or which are obtained by requests.

Abstract: A method for encoding digital video with low delay for transport on a multi-channel QoS packet-switched network using multiple reference pictures with specially designated “key frames” to minimize the undesired effects of packet loss on video quality.

Abstract: The invention proposes a switching arrangement for transporting data packets which comprise a data packet destination information and a payload, to one or more output ports. The switching device is able to route the arriving data packets according to the data packet destination information, to at least one dedicated of the output ports. It comprises at each input port an input buffer with at least as many single input queues as there are output ports, and an input controller for each input port, serving for controlling the order of multiplexing the data packets from the input queues of the corresponding input buffer to the switching device. The total of input ports is divided up into several subsets of input ports. Each subset in the switching device has its separate output buffer for storing at addresses therein at least the payload of each data packet arriving at the input port. At least one set of as many output queues as the switching arrangement has output ports are arranged.

Abstract: A switching system connectable to a communications network and configured to handle logical multicast data without degrading the flow of other data through the system. The switching system includes a plurality of input ports, a plurality of output ports, and a plurality of queues for buffering streams of data units being routed through the system. The switching system further includes at least one meter for metering data units to ensure that the data units conform to predetermined Quality of Service criteria. The logical multicast data is metered once at the meter and subsequently replicated and transmitted onto the communications network a predetermined number of times at an output port so as not to degrade the flow of other data through the system.

Abstract: Using CDMA encoding and spectrum spreading with a factor N, more than N messages are transmitted although N is a theoretical limit for this type of encoding. Encoding uses orthogonal sequences for a first group of N messages and random or pseudo-random sequences PN for a second group of M-N additional messages. On reception, the N messages of the first group are decoded and detected. Their interference is subtracted from the second group of M-N messages before detecting the M-N messages. In a second iteration, the interference of the M-N messages using PN sequences is estimated and subtracted from the N first messages before a second detection on the first group of N messages. Detection is then repeated for the second group of M-N messages after subtracting the interference of the first group of N messages and the mutual interference of the M-N messages.

Abstract: A head of line blockage avoidance system for use with network systems that employ packets having an associated priority and a method of operation thereof. In one embodiment, the head of line blockage avoidance system includes: (1) m inputs, m numbering at least two, configured to receive the packets, (2) n packet first-in-first-out buffers (FIFOs), n numbering at least three, each of the packet FIFOs configured to receive at least one of the packets from the m inputs, (3) a priority summarizer configured to generate a priority summary of the packets within the m inputs and the n packet FIFOs, and (4) a scheduler configured to cause one of the n packet FIFOs to be queued for processing based on the priority summary.

Abstract: The present invention relates to an AAL switch for multicast and a switching method for performing a flexible multicast connection. An AAL2 switch for multicast and a switching method for performing a flexible multicast connection can be used in applications like video conference and video on demand. An AAL2 switch can include a plurality of receiver circuits that convert an AAL2 packet into a CPS packet by demultiplexing, a plurality of storing devices that store a CPS packet separated from the AAL2 packet, and a plurality of transmitter circuits that periodically search the plurality of storing device and convert the searched CPS packet into an outgoing AAL2 packet by multiplexing. The transmitter circuits can transmit the outgoing AAL2 packet.

Abstract: In a communications arrangement where services requested by users are provided over a network, communication flow arising from the services are shared by multiple links in the network. The users are required to pay for use of the links to realize the respective services. A target operating point defining the desired amount of communication flow on each link is established. A pricing scheme in accordance with the invention is used to price each link on a per unit of flow basis. Each user elects use of the capacity of those relatively inexpensive links to fulfil the requested service to minimize the cost. Such optimization actions by the individual users give rise to a “routing game” leading to a Nash equilibrium which coincides with the target operating point. Thus, with the inventive pricing scheme, the target operating point is attained over time by virtue of the users' cost optimization actions.

Abstract: A multicast network system utilizes a high speed link, such as a satellite link, to multicast multimedia information from the Internet to a plurality of receivers, such as personal computers. Information from selected web sites is organized into “channels” and provided to a multicast network for multicast transmission to the receivers. Updated channel information is also periodically provided. The receivers store the received channel such that a user can access the web page content in the channel at hard disk speed. Preferably, a conditional access system ensures that only authorized receivers receive the channels. The present invention also preferably includes a lower speed two-way connection to the Internet (such as dial up modem) which is used to report usage information and/or subscription information back to the web sites.

Abstract: Data switching systems for use in ATM and other packet and cell switching networks can more smoothly switch data if incoming data packets to be switched are re-ordered to avoid overwhelming an output port with too many consecutive packets to a particular destination. An apparatus and method for simply re-ordering data samples in a nonconsecutive manner is computationally efficient and effective.

Abstract: A cross-connect switch (12) uses a matrix (40) for providing connection paths. The matrix (40) includes prioritization logic (42) for receiving connection requests from restoration state machines (50) (for changing connections responsive to line conditions) and an administrative processor (for changing connections responsive to operator commands). Connection requests from both the restoration state machines (50) and the administrative processor (52) are cached in FIFO memories (54). When a FIFO memory (54) stores one or more connection requests, the switching control circuitry is notified by a data ready signal. Logic (56) inhibits passing of the data ready signal from the FIFO memory (54) to the switching control (26) until all restoration connection requests have been serviced.

Abstract: A packet switching device having a central shared memory and a number of medium access controllers each coupled to a communications medium to exchange data packets therewith, and a controller coupled to each medium access controller via a data path to exchange data packets with the media access controller. The controller has a number of data path controllers each connected to each medium access controller via a separate and like portion of the data path to exchange a corresponding portion of the data packets with the medium access controller. The data path controllers each have a number of buffers each connected to one of the medium access controllers to which the data path controller is connected, to hold the portion of the data packets exchanged with the corresponding medium access controller.

Abstract: Broadband switching including the implementation of and control over a massive sub-microsecond switching fabric. To effect the attributes of the switching fabric, conditionally nonblocking components are used a building-blocks in an interconnection network which is recursively constructed. The properties of the interconnection network are preserved during each recursion to thereby configure the massive switching fabric from scalable circuitry.

Abstract: Broadband switching including the implementation of and control over a massive sub-microsecond switching fabric. To effect the attributes of the switching fabric, conditionally nonblocking components are used a building-blocks in an interconnection network which is recursively constructed. The properties of the interconnection network are preserved during each recursion to thereby configure the massive switching fabric from scalable circuitry.

Abstract: A method and apparatus for managing contention in a self-routing switching architecture based on a set of n×n individual switching structures that are connected in a port expansion mode by means of fan-out and fan-in circuits providing access of the Switch Core Access Layer (SCAL) to the different input and output ports of the switching core. The fan-in circuits use an arbitration mechanism for providing a token to the switch that is allowed to deliver the next cell and the arbiter operates from a detection of a special comma character in accordance with the 8B/10B coding scheme which is introduced in the data flow between the individual switching structures and the fan-in circuits. This provides a compensation for the difference in transfer delays of the cells even when high switching speed and long length of the physical media are involved.

Abstract: This invention describes a method for transmitting and forwarding packets over a switching network using time information. The network switches maintain a common time reference, which is obtained either from an external source (such as GPS—Global Positioning System) or is generated and distributed internally. The time intervals are arranged in simple periodicity and complex periodicity (like seconds and minutes of a clock). A data packet that arrives to an input port is switched to an output port based on its order or time position in the time interval in which it arrives at the switch. The time interval duration can be longer than the time duration required for transmitting a data packet, in which case the exact position of a data packet in its forwarding time interval is predetermined.

Abstract: A communication device integrates traffic monitoring, policy configuration and enforcement, and proxy services within the device. The policies can affect both prioritization of data as well as allocation of communication bandwidth. Data packets that are accepted at the device are classified according to a set of defined classes and are transmitted through the device according to a policy for prioritization and bandwidth allocation for the classes. Optionally, some of the packets are passed to a proxy application hosted in the communication device. The device functions as a link-layer bridge, thereby allowing the device to be inserted into a data path without reconfiguring network layer software at devices that use that data path. The device includes operating modes in which packets are passed through the device without modification.

Abstract: A system for broadcasting data (30) that includes a message flow optimizer (150) that forms a single concatenated message (180) from a number of discrete messages (190, 200), and a number of broadcast message receivers (250) operable to isolate a portion of the single concatenated message (180).

Abstract: A packetized-data processing apparatus includes a memory configured to store core groups of packetized data, a channel coupled to the memory and having a total bandwidth for transferring packets of data from the core groups, and a scheduler operatively coupled to the memory and the channel and configured to allocate amounts of the total bandwidth of the channel to each of the core groups that is backlogged, while limiting the amount of allocated bandwidth, and a corresponding transfer rate of packets of data, for each core group to a maximum allowable bandwidth for each core group, to schedule transfer of packetized data of the core groups from the memory to the channel in accordance with the respective amounts of allocated bandwidth for the core groups.

Abstract: Contention losses are minimized in path searching, circuit-switched networks by adding intelligence to the last stage of the network. A count of number of bytes remaining to be transferred in each active message is maintained in real time for each output port of the network. If contention arises at the last stage switch being requested in camp-on mode to make a connection to a busy output port, the switch checks the bytes remaining count and responds differently depending on how the bytes remaining count compares to a preset threshold register. If the count remaining is below the threshold, the last stage switch accepts the camp-on request, because the desired output port will be available shortly. If the count remaining is above the threshold or below the threshold but another user is camped-on, the switch rejects the camp-on request to the last stage, because the desired output port will not be available shortly. In this case further path searching would be meaningless.

Abstract: A systematic method for creating layouts for various sorting and switching networks on VLSI chips is proposed. These sorting and switching networks include butterfly switching networks, benes networks, odd-even sorting networks, and bitonic sorting networks. The present invention utilizes the existing VLSI grid-model to create various layouts. A rectangular layout is proposed which introduces the use of an additional row to create an area-efficient layout for various switching networks. A diamond layout is proposed which further improves rectangular layout by reducing the number of columns required. A suitable combination of diamond and rectangular layouts is also proposed.

Abstract: An improved system for sorting a data set of 2.sup.m keys in a parallel processor environment is disclosed. The system operatively maps a shuffle-exchange network onto a Batcher sorting network in order to produce a monotonically sorted output set. Complete and efficient sorting is achieved by shuffling a control mask applied to multiple compare-exchange elements when the data set is redundantly shuffled. A further improvement is achieved by employing an unshuffle operation to the data set and/or the control mask when the use of an unshuffle operation would be more expedient than use of a shuffle operation.

Abstract: A switch network and method of an asynchronous transfer mode switch network, includes an input buffer; an input switch network performing fetch and add function; a copy switch network performing fetch and add function; a feedback switch network receiving cells fedback and maintaining them until the next cell cycle; a routing table for storing data for translation and replacement of data for the cells' routing; a routing/feedback controller routing-controlling the input cells and fedback cells with reference to the routing table; routing switch networks switching-outputting the cells input; a cell splitter splitting the cells routing-output into the number of the routing networks and transferring them to the routing switch networks, and a cell merger merging and outputting the cells split and output from the output ports of the routing switch networks

Abstract: Improved layouts for sorting networks having .smallcircle.(N.sup.2) comparison circuits, a method therefor, and improved switches incorporating such layouts, are disclosed. An improved layout reduces grid-area requirements for Batcher's odd-even sorting network to an area of 3N.sup.2 grid units. The improved layout results from (i) arranging the comparison circuits comprising the network in one or more "cones," (ii) "slanting" a substantial portion of the interconnections or links so that they extend in a substantially non-orthogonal direction to the axes of a coordinate system such as is typically used in defining a grid for laying out a sorting network, and (iii) using channel linking for exiting a cone and for combining appropriate outputs from the exited cone in subsequent cones.

Abstract: A packet router for a data packet transmission network, wherein routers offer priority services of the type required for isochronous handling of data representing real-time voice, includes a Quality of Service (QoS) management system for ensuring that guarantees associated with such priority service can be met with a high degree of certainty. This management system provides prioritized queues including a highest priority queue supporting reservations for the priority service suited to isochronous handling. The highest priority queue and other queues are closely monitored by a QoS manager element for states of near congestion and critical congestion. While neither state exists, filler packet flows are promoted from lower priority queues to the highest priority queue, in order to keep the latter queue optimally utilized. If all lower priority queues are empty at such times, dummy packets are inserted as filler flows.

Abstract: Improved layouts for sorting networks having O[N log.sup.2 N] comparison circuits, a method therefor, and improved switches incorporating such layouts, are disclosed. An improved layout reduces grid-area requirements for Batcher's bitonic sorting network to an area of 4N.sup.2 grid units. The improved layout results from (i) slanting a substantial portion of each link between comparison circuits 45.degree. with respect to a conventional layout of the network, (ii) incorporating "flip-links," and (iii) not restricting all comparison circuits having the same network depth to placement in a single grid column.

Abstract: An augmented ring-banyan network utilizing the adaptive self-routing control algorithm in the ATM switching system comprises qth(.gtoreq.1, interger) augmenting stage having prescribed switch elements each connected to the switch elements of the final stage in said ring-banyan network respectively, and q+1st augmenting stage having prescribed switch elements each connected to said prescribed switch elements of said qth(q.gtoreq.1, interger) augmenting stage respectively, wherein the topology for the output links of each of said switch elements is represented as follows:.beta..sub.1 [(P.sub.l,P.sub.l-1, . . . ,P.sub.1).sub.i ]=(P.sub.l,P.sub.l-1, . . . ,P.sub.1).sub.i+1 connectedwithlink( P.sub.l,P.sub.l-1, . . . ,P.sub.1,1).sub.i.beta..sub.0 [(P.sub.l,P.sub.l-1, . . . ,P.sub.1).sub.i ]=(P.sub.l,P.sub.l-1, . . . ,P.sub.1).sub.i+1 connectedwithlink( P.sub.l,P.sub.l-1, . . . ,P.sub.1,0).sub.iProvided:l=(log.sub.2 N)-1, i.gtoreq.l+1.

Abstract: A switch that has a plurality of inputs in which cells are received at these inputs. Each cell received at the inputs of the switch contain routing information. A routing means is employed to route the cells received at the inputs of the switch to outputs using routing information in which a number of the cells are misrouted by the routing means during the process of routing the cells to the outputs. Bus means is employed to route a cell to the destination in which the bus means is connected to the routing means. The bus means routes misrouted cells that are misrouted from the destination by some selected amount.

Abstract: Sorting networks having enhanced layouts, and a method for developing such layouts, are disclosed. The improved layouts use 5/4 N grid rows, where N is the number of network inputs. Additionally, up to log (N) permutations of interconnections within the network are combined into a single permutation, thereby reducing grid column requirements for network implementation. The improved layout, as applied to Batcher's bitonic and odd-even networks results in respective upper bound grid-areas of 11.25N.sup.2 +o(N.sup.2) and 9.375N.sup.2 +o(N.sup.2) grid units for a network of N inputs.

Abstract: A circuit switched switching system and method includes a plurality of matrix transposition memories. The memories permute the order of incoming and outgoing data, and thereby increase the efficiency of the switch. The switch and matrix transposition memories may be disposed in a satellite. The switch may further include batching circuits for grouping together data from a single terminal device for more efficient transmission through the switch and for more efficient error correction.

Abstract: An apparatus for routing a plurality of cells. The apparatus includes a plurality of inputs, wherein the plurality of cells are received at the inputs. Each of the cells contains routing information. The apparatus also includes a plurality of outputs, wherein the outputs are associated with a plurality of destinations such that each output corresponds to a destination within the plurality of destinations. The apparatus employs a routing means for routing the cells received by the inputs through the apparatus to the outputs using the routing information contained in the cells. The routing means includes a number of single path switching matrixes in which each single path switching matrix has a plurality of stages. A portion of the these stages are dilated such that two or more cells may be routed to the same destination.

Abstract: An automatic path setting apparatus for a synchronous communication system equipped with a network management system (NMS) is provided which relieves the operator of the labor associated with routing and which permits prompt and reliable line settings. A line connection information storing element in the NMS stores information about line connection states of individual NEs. A specified information receiving element receives specified information about new path setting, and a path searching element searches for a path conforming to the specified information received by the specified information receiving element by referring to the information stored in the line connection information storing element.