Abstract: A switched integrated wideband and narrowband multiservices digital network (FIGS. 1 and 2) is an ISDM providing universal information services based on wideband and narrowband voice, data, and video communications. It comprises a plurality of service areas (100, 101), each served by a central switching node (110). The central node is connected to a plurality of remote nodes (103) by feeder optical fibers (107) and a control bus extension (106). Network interface equipment (104) at subscribers' (102) premises is connected to remote nodes by distribution optical fibers (105). Each distribution fiber is wavelength-division multiplexed and carries modulated (pulse-analog, pulse-code, or differential pulse-code) wideband digital channels (205) and a multiplexed channel (206) comprising 32 time-division-multiplexed narrowband digital channels (207). One narrowband channel (207D) carries all signaling messages. Feeder fibers are wavelength-division multiplexed and carry modulated wideband digital channels (305), and multiplexed channels (306) each comprising a plurality of time-division-multiplexed distribution multiplexed channels. Each remote node comprises a digital space-division switch (505) for wideband channels, and a digital time-division multiplexer and demultiplexer (506) for multiplexed channels. Each central node comprises a digital space-division switch (606) for wideband channels, and a digital time-division switch (607) for narrowband channels. All switches are controlled by a central node control complex (612) over a control bus (116) and its extensions. Signaling messages are transferred between the signaling-message-carrying narrowband channels and the central node complex by a subscriber signaling subsystem (613) via the narrowband switch and the control bus. A central node optionally includes interfaces (615-618) to other communication systems, and trunk communication fiber (112) and CCIS signaling fiber (117) connections to other central nodes of the network.

Abstract: Improved alternate routing in a packet switching system is provided by inserting alternate routing control information into each packet and by storing alternate routing information at each network node. The stored information at each node includes a list of the available paths extending from the node towards all other nodes together with a list of available algorithms that can be used to select one of the available routes. The alternate routing control information in each packet contains postage information specifying the maximum number of nodes through which the packet is to travel. The alternate routing control information also includes a destination node index code identifying the destination node. The destination node index is used as address information by each node receiving a packet to read out the stored information at the node identifying the available paths and the algorithm to be used in selecting one of these paths for use in transmitting the packet towards the destination node. The identified algorithm is then executed to select the path to be used.

Abstract: Method and apparatus for transfer of packet-type information from the memory (24B) of one node (14) in a computer network to the memory (24C) of another node (16) in the network The invention is of particular utility in transfers over serial buses (eg, 18) Packets are sent from a named memory buffer (25A) at a first node (14) to a named memory buffer (25C) at a second node (16), allowing random access by the first node to the memory of the second node without either node having to have knowledge of the memory structure of the other, the source and destination buffer names are contained right in the transmitted packet The first node (14) can both write to and read from the second node (16) An opcode (40A) sent in each packet signifies whether a read or write operation is to be performed For reading from the second node, the opcode actually causes the second node to write back to the first node; in this situation, the second node, upon detecting the appropriate opcode, places the remainder of the received packet on a command queue (202), to be executed with the commands locally generated at the second node, without need for host interruption

Abstract: The difficulty of interconnecting networks or network components when they obey different architectures is an important inhibitor to computer communication growth and flexibility. This paper reviews the state of the protocol conversion art from the architectural point of view. The aim is to state the problem succinctly, propose consistent definitions, review current ad hoc solutions, and speculate about more effective and general methodologies. To do this, we first analyze what happens on an access path between two end users at the discontinuity point where the path passes from a portion of the network obeying one architecture into a portion obeying another. This discontinuity can be designed to occur at a choice of nodes along the access path and at a choice of protocol layer level at each such node. The process of conversion to complete the path at one of these points can range in complexity from the trivial to the impossible. The procedure available today starts by analyzing the set of "atomic protocol functions" that support the services normally provided at the conversion point by the two architectures in terms of their interface, peer and control protocols (both transient and steady state), and then identifying a usable common subset of the two sets. Known solutions then include defining useful partial resolution of protocol mismatches, complementation of one or both protocols with missing services, and the use of Open System Interconnection as an intermediary in the conversion process.

Abstract: A message passing parallel processing system capable of executing different types of intercommunication protocols simultaneously, as well as plural instances of each type of protocol. Each processing node of the system contains a message passing protocol apparatus comprising a protocol processor and a memory. Messages passed between nodes contains, inter alia, a protocol type and an identification of an instance of the protocol type. Examples of protocol types might be shared memory, hierarchical tree connect, distributed sets, etc. Messages are passed to processing nodes from other processing nodes in fixed sized blocks and only when input buffers are empty. Flow control of the network is self-regulating and the network is deadlock free.

Abstract: In a distributed database system the partitioning and allocation of the database over the processor nodes of the network can be a critical aspect of the database design effort. In this paper we develop and evaluate algorithms that perform this task in a computationally feasible manner. The network we consider is characterized by a relatively high communication bandwidth, considering the processing and input output capacities in its processors. Such a balance is typical if the processors are connected via busses or local networks. The common constraint that transactions have a specific root node no longer exists, so that there are more distribution choices. However, a poor distribution leads to less efficient computation, higher costs, and higher loads in the nodes or in the communication network so that the system may not be able to handle the required set of transactions.Our approach is to first split the database into fragments which constitute appropriate units for allocation. The fragments to be allocated are selected based on maximal benefit criteria using a greedy heuristic. The assignment to processor nodes uses a first-fit algorithm. The complete algorithm, called GFF, is stated in a procedural form.The complexity of the problem and of its candidate solutions are analyzed and several interesting relationships are proven. Alternate benefit metrics are considered, since the execution cost of the allocation procedure varies by orders of magnitude with the alternatives of benefit evaluation. A mixed benefit evaluation strategy is eventually proposed.A model for evaluation is presented. Two of the strategies are experimentally evaluated, and the reported results support the discussion. The approach should be suitable for other cases where resources have to be allocated subject to resource constraints.

Abstract: An arrangement is disclosed for providing station number portability to stations ported from an original switching node to a new switching node which allows the ported station to maintain its original assigned station number. A ported station retains its original assigned number when it's original and new node have a common data base. The common data base identifies what station number is currently associated with which node. The station can be ported from one node to another node by altering the data base to indicate the new association of the ported number and the node to which the station has been ported.

Abstract: A programmable controller is adapted to communicate in a token passing logical ring network in a peer-to-peer fashion. The programmable controller is a station on the network and contains an interface which implements the network protocol. The protocol requires that each station maintain an Active Node Table which stores a complete map of the logical ring and the order in which the token is passed. The Active Node Table is automatically updated by monitoring each token pass on the network, where each token pass identifies the source and destination of the token pass. The Active Node Table is used in performing all routine network maintenance functions, including passing the token, adding stations, and dropping stations. The Active Node Table allows the formation of an Active Node Status Word for use by high level processes within the station. The Active Node Table also facilitates the inclusion of data in the token pass message.

Abstract: A method and apparatus for flexibly interconnecting the nodes of a local data network to achieve reliable internodal data transmission while minimizing the extra data processing load on the host processors of each node. An interface processor is provided at each node which controls transmission and reception of data packets and the communication of data from and to the location in node storage associated with the program processes which generate and receive the data. Different protocols are provided for different types of messages and are controlled by the interface processor in order to provide high reliability data transmission where needed. Destination addresses are associated with each data packet to provide flexible routing of data.

Abstract: A communication method and packet switching network (101) in which self-routing packets are communicated via multipaths through the network while maintaining the sequence of the packets. The switching network has a plurality of stages (0, 1, 2, 3) the stages are interconnected by inter-node links (204, 205, 206). Each inter-node link comprises a pair of sublinks (204-0, 204-1, 205-0, 205-1) thus establishing multipaths through the switching network in response to the self-routing packets. Each stage has a plurality of switch nodes. A switch node is responsive to packets received on one inter-node link destined for a second inter-node link to maintain the sequence of packets as they are communicated out on the second inter-node link. Each node comprises a plurality of input circuits (700-703) each individually connected to one of the incoming sublinks and a plurality of output circuits (736, 737) each connected to one of the outgoing inter-node links. Each output circuit has an arbiter (734, 735) which properly sequences the packets onto the outgoing link in response to requests for communicating packets received from the input circuits.

Abstract: A network of microprocessors, or nodes, are interconnected in an n-dimensional cube having bidirectional communication links along the edges of the n-dimensional cube. Each node's processor network includes an I/O subprocessor dedicated to controlling communication of message packets along a bidirectional communication link with each end thereof terminating at an I/O controlled transceiver. Transmit data lines are directly connected from a local FIFO through each node's communication link transceiver. Status and control signals from the neighboring nodes are delivered over supervisory lines to inform the local node that the neighbor node's FIFO is empty and the bidirectional link between the two nodes is idle for data communication. A clocking line between neighbors, clocks a message into an empty FIFO at a neighbor's node and vica versa. Either neighbor may acquire control over the bidirectional communication link at any time, and thus each node has circuitry for checking whether or not the communication link is busy or idle, and whether or not the receive FIFO is empty. Likewise, each node can empty its own FIFO and in turn deliver a status signal to a neighboring node indicating that the local FIFO is empty. The system includes features of automatic message rerouting, block message transfer and automatic parity checking and generation.

Abstract: A method for rounting user packets in a multinode interconnect network is disclosed. Each packet has a destination node (10). Each node (10) is connected by input and output lines (12) and has a plurality of packet storage links (34) and (56-60). Each link (34) and (56-60) is connected to an input line (18-24) and has a number of packet storage buffers (38-44) and is connectable to any of the node's output lines (26-32). During operation node - (10) uses a set of rounting instructions and a look-up table (192) to determine which user packets are sent out which output lines (26-32). The rounting rules determine which output line is connected to the shortest path to the destination node, and ensure that the node (10) will always have an empty storage buffer (38-44) for each line (18-24) for receiving further user packets.

Abstract: A method for transmitting data in packet switching networks provides a Collision-Eliminating Multiple Access protocol in which nodes desiring to transmit over the network channel transmit reservation requests during a plurality of contention slots, the number of contention slots being dynamically controlled according to network load. A node designated to next obtain control of the channel receives the identifiers of nodes transmitting reservation requests and, prior to transmitting application data, transmits network control data consisting of the identifiers of nodes from whom successful reservation requests were successfully received. The transmitted identifiers are received and stored by each node in an identical queue, whereby subsequent control of the channel is rotated based on the order of node identifiers appearing in an identical queue on each node. The transmitted network control data includes reservation requests received during a previous contention slot period, queue correction information, and the identifiers of nodes from which the controlling node expects to receive data.

Abstract: Local area networks having a high level of network integrity, error detection and correction, message security, expandability and flexibility are disclosed. The system is comprised of a loop controller which may communicate with a host through a standard interface such as an RS232 interface, and to which a plurality of nodes are coupled in a reconfigurable network. Each node has a plurality of modes controlled by the network controller to serve various functions depending upon requests from the controller and/or the needs or condition of the respective node from time to time. In particular, the first node is coupled to the controller through a pair of wires, with each successive node being coupled to the preceding node through a similar pair of wires. In one mode, the respective node monitors the incoming signal from the controller as repeated through any preceding nodes, and itself acts as a signal repeater for the next downstream node. The node also acts as a signal repeater to any return signal from the downstream node to pass that signal on toward the loop controller. In another mode the respective node provides its own signal on the return line for transmission toward the controller, and at the same time monitors the incoming return signal received from the next downstream node for repeating that signal on the upstream return line as soon as it completes its own transmission. In still another mode, the incoming signal is not repeated for the next node downstream but rather is looped back toward the controller.

Abstract: A computer aided dispatch system is provided for use in a trunked communication system. The system comprises at least a master file node and a plurality of user nodes. The master file node contains a data record for each subscriber on the trunked communication system. Each data record has a plurality of fields that may assume various values. Each dispatcher "attaches" to a particular value of at least one field in any of the data records. The master file node maintains the records for each subscriber and automatically transmits an updated record to each dispatcher attached to the subgroup in which the subscriber operates. In this way, dispatchers are continuously provided the latest subscriber status even though the responsibility for monitoring the status of a particular subscriber unit may dynamically pass from dispatcher to dispatcher.

Abstract: A grid-based mesh network with even numbers of alternatingly directed rows and columns. Devices are attached by way of nodes (100) which exist at the intersections of rows and columns. The extremes of the grid are connected to form a three-dimensional network on the surface of a sphere. The strategy at each node requires that every packet arriving on an incoming link must leave on an outgoing link, so there is no need for any packet buffering in the node. Described herein are several routing algorithms, an alternative to sequential addressing, a method for adding nodes to the network, ways to deal with failed nodes and links, a hierarchically structured network, and a scheme for efficiently transferring files of more than one packet.

Abstract: A digital data message transmission network of interconnected network elements (12,14,16,18,20; 32,34,36,40,44), maintains topology data bases (112) recording the potentially available communication routes through the network and the status of the network elements thereof, each node originating a message interrogating base to ascertain a suitable route; when connected into a route, whether completed or not, recording the adjacent elements; when detecting the failure of any next adjacent link or node sending a ROUTE FAILURE MESSAGE to the route end(s) available to it to update the topology data base(s) for such route(s), whereby the route maintenance message traffic within the network is limited while ensuring that the topology data base associated with the origin of a route that does not establish and the topology data base or bases associated with both ends of all failed established route incorporating that failed element are correspondingly updated.

Abstract: In this paper we study a message resequencing problem in a store-and-forward computer network where messages may go out of order while traversing logical channels. The logical channels are assumed to consist of multiple physical links which may be of different capacities. A message is dispatched to the fastest available link. Resequencing methods suggested in the literature [3] (resequencing at the channel level and resequencing at the virtual circuit level) are investigated for this link selection rule. The analysis is done on a two-node network connected by multiple links. The source node together with the set of outgoing links are modeled as an M/M/m queue with servers of different rates. The resequencing delay distribution and the average resequencing delay are derived. On multihop networks, the effect of message length, link numbers, link service rates, and the resequencing methods on resequeucing delay are investigated by simulation.

Abstract: A modular, hierarchical local area network is realizable by using existing twisted pair wiring, for example, in a building together with several elements known as network access units and hub units. A network access unit is connected to a device (microprocessor, display terminal, peripheral device, other local area network, information systems network or the like) in the local area network to give that device access to and from the network. The network access device is designed for simple daisy chaining with other devices to form a node when devices are colocated in the same section or room in the building. The hub unit is the network building block which permits extensive expansion of the local area network. It provides a connection point for access units or other hub units, performs collision detection operations, if necessary, and serves as a loop-back point for the network. Switchably connectable termination impedances and loop-backs are automatically passed to the appropriate points in the local area network. While most small local area networks are considered for intra-building applications, the present network is applicable to inter-building applications, also.

Abstract: A mapping system and method which establish a number of navigation nodes and paths among them for a vehicle in an environment to be navigated by directing the vehicle to a selected location in the environment, designating that location as a first node, identifying the initial direction of a path to a second node such as by locating a navigation beacon proximate the second node, and moving the vehicle in the initial direction from the first node while measuring distance travelled along the path to the second node This system and method further include defining the second node by at least the distance travelled along the path and by the initial path direction

Abstract: Researcher have long conjectured upon the possibility of constructing large, massively-parallel computing engines by interconnecting many conventional processing elements to form an integrated supersystem. The rapid expansion in very large scale integration, or VLSI, circuit technology during the past decade has accelerated research in this direction. As advances in VLSI push basic component or chip functionalities to the processor level and beyond, it becomes natural to view complex processing elements as the basic components of much larger systems. Several names for such systems have been proposed, including network computers, multicomputers, and distributed multiprocessors. Despite the naming differences, these systems have the following salient features: (1) A large number of basically autonomous processing elements interconnected by a structure that allows high-bandwidth communication between them. At the system level, these processing elements and interconnection facilities are viewed as the basic components of the system. Each processing node has its own local memory and there is no sharing of memory between nodes. (2) A high degree of distribution of control or operating system functions among the processing elements. (3) Highly parallel computation performed by constructing applications as collections of several or many distinct tasks. These tasks may execute concurrently on different processors, withmore » necessary intertask communication carried out over the communication facilities linking the nodes. The collection of cooperating tasks comprising an application is sometimes referred to as a task force.« less

Abstract: We consider distributed solution of the classical linear minimum cost network flow problem. We formulate a dual problem which is unconstrained, piecewise linear, and involves a dual variable for each node. We propose a dual algorithm that resembles a Gauss-Seidel relaxation method. At each iteration the dual variable of a single node is changed based on local information from adjacent nodes. In a distributed setting each node can change its variable independently of the variable changes of other nodes. The algorithm is efficient for some classes of problems, notably for the max-flow problem for which it resembles a recent algorithm by Goldberg [11].

Abstract: A user-controlled interactive computer display system and method is disclosed for manipulating a hierarchy of information (a topmost node and a number of subordinate nodes, each with only one superior node). This display system and method allows a hierarchical arrangement of information to be constructed and changed with a minimum of steps and errors because positional information about each node is handled implicitly by the system.

Abstract: A system (20, 22) for compensating for varying attenuation of an uplink signal from a local node to a satellite. The system (20, 22) monitors two beacon signals and the local downlink signal to determine fade. An error signal, indicating the uplink fade, is generated and utilized to adjust the gain of the uplink transmitter (42) to compensate for the fade.

Abstract: A data communication network comprising a plurality of nodes interconnecting transmission paths in a ring structure arranged to propagate data messages between data systems coupled to the network by the nodes in opposite directions around the network ring structure. An executive node having ring interface units interconnecting ones of the transmission paths is arranged to selectively couple a network control processor with various sectors of the transmission paths to enable the network control processor to control the traffic flow of data messages on the data communication network.

Abstract: A model for distributed systems with failing components is presented. Each node may fail and during its recovery the load is distributed to other nodes that are up. The model assumes periodic checkpointing for error recovery and testing of the status of other nodes for the distribution of load. We consider the availability of a node, which is the proportion of time a node is available for processing, as the performance measure. A methodology for optimizing the availability of a node with respect to the checkpointing and testing intervals is given. A decomposition approach that uses the steady-state flow balance condition to estimate the load at a node is proposed. Numerical examples are presented to demonstrate the usefulness of the technique. For the case in which all nodes are identical, closed form solutions are obtained.

Abstract: : This report summarizes Distributed Sensor Networks research conducted during the period 1 April through 30 September 1978. One specific model for function decomposition and distribution is presented. A top level design of a strawman DSN for detecting and tracking low flying aircraft is described. High resolution acoustic signal processing algorithms for the strawman are specified and sized. Additional features of a multisite target search algorithm are presented. A moderately detailed model for simulation of an acoustic node under various scenarios is developed. New model simulation and experimental data analysis software are described and demonstrated. Technical issues involved with and plans for a digital acoustic data acquisition system are presented.

Abstract: An optical-fiber ring network is capable of operating in the face of the failure of any single node regardless of the particular node failure mechanism, including stuck "on" and stuck "off" transmitters. Each node in the network (302) comprises a main receiver (312), an alternate receiver (322), and a transmitter (332). The main receiver (312) receives data from the immediately adjacent upstream node (301), while the alternate receiver (322) monitors transmissions from the next preceding upstream node. Each node diagnoses the transmitter in its immediately adjacent upstream neighbour and its own main receiver. If either fails, the node switches from its main receiver to its alternate receiver to bypass the immediately adjacent upstream node, while the rest of the ring remains functional.

Abstract: A communication system for the transfer of small digital message blocks and large digital message blocks. A plurality of nodes are coupled to a common wideband communication channel capable of facilitating the transfer of small digital message blocks and large digital message blocks. One of the nodes is a master node controlling the communication of the communication channel by selectively polling the other of the plurality of nodes. The communication system has a first protocol for the transmission of small digital message blocks and a second protocol for the transmission of large digital message blocks. The master node establishes a polling priority for each of the plurality of nodes and polls those with a higher polling priority more frequently than those with a lower polling priority. The master node allows the one node being polled to communicate over the channel of either a small digital message block or a large digital message block. Further the master node adapts the polling priority for each of the nodes based upon a predetermined algorithm.