Abstract: Unicast endpoint clients (110, 111, 115) on an IP Unicast network (107, 108) are provided access to Multicast sessions on an IP Multicast network (101) through a Multicast-Unicast gateway server (120, 121). The server obtains information about sessions on the Multicast network and makes such information available to a Unicast client on the Unicast network upon request by the client. Upon being presented with a list describing the subject matter of each session, the user at the Unicast client selects the session to which he or she wants to join, which causes the Multicast-Unicast server to join the appropriate session on behalf of the requesting client for each media type in which the joining client wants to be a participant. The server then sets a bi-directional Unicast User Datagram Protocol (UDP) stream between itself and the client. All packets then received by the server from the Unicast client are address-translated to the appropriate Multicast session address.

Abstract: Paging, cellular telephone or other messaging functionality is incorporated into a virtual contact center system to permit logged-off customer service representatives (CSRs) to be alerted to a need to logon to the system, and to permit CSRs to alert supervisors or specialists to logon to the system. Matching at a contact control server of existing conditions to predetermined criteria, e.g., failure to login at an assigned time, or specified network conditions (such as length of customer queues), triggers a command for a communication to one or more logged-off CSRs. Results of follow-up testing at the contact control server for matches causes page requests to be renewed, redirected or canceled.

Abstract: A method and apparatus provide improved cache coherency and more effective caching operations without placing an undue burden on network links. A proxy receives a request for a resource and then, depending on information in the proxy cache, generates a resource request for transmission to a resource server. The proxy appends a proxy filter to the request. The resource server maintains one or more volumes of resources based on some predetermined criterion that can be either static or dynamic in nature. Upon receipt of the request and the proxy filter the resource server generates a request response and a piggybacked list of additional resources selected from the volume with which the requested resource is associated.

Abstract: The present invention, in illustrative embodiments, implements systems and methods for adjusting the timing of access requests in mobile communications systems, such as IS-95-A mobile telephone systems, to reduce the likelihood of request collisions—especially during periods of very high calling rates. These embodiments employ measurements of access channel occupancy, as well as system operator inputs, to develop thresholds and parameter translations at a base station. Execution of control algorithms at a base station advantageously yields modified persistence test parameters sent to mobile stations. When overload conditions are present, these modified parameters cause the mobile stations to experience variable delays through judicious control of persistence testing at the mobile stations.

Abstract: A new architecture and operational techniques for supporting high quality live and on-demand streaming multimedia on a data network. By using Helper machines inside the network as data forwarding, caching, and buffering agents, and by forming meshes among Helper machines, advantages of homogeneous, synchronous multicast transmission and of heterogeneous, asynchronous reception are achieved. The architecture provides essentially transparently support to the receivers for near-zero start-up latency, improved playback quality, improved CR-like operations. These are achieved while reducing network and server load compared to today's multimedia networking techniques.

Abstract: High-speed wireless data systems and methods are described for simultaneously supporting multiple data users, including flexible mixed-traffic services, using an illustrative CDMA architecture. Embodiments are compatible with current and third generation CDMA systems, while achieving spectral efficiencies that are in some cases an order of magnitude or more higher. Embodiments advantageously employ multiple antennas at the transmitter and receiver and code spreading to realize multiple access systems. Some illustrative receiver embodiments advantageously employ a decorrelating detector, while others feature a decorrelating decision-feedback detector. In typical operation, each of a plurality of high-speed data streams directed to respective data users is demultiplexed into multiple (G>1) lower rate substreams, which are modulated by a CDMA spreading code and transmitted from M antennas at a base station.

Abstract: Automatic Speech Recognition (ASR) is achieved in wireless communications systems in which reliable ASR feature vector sequences are derived at a base station directly from digitally transmitted speech coder parameters, with no additional processing or signal modification required at the originating handset. No secondary channel need be provided for the transmission of ASR feature vectors. In operating on received speech coder parameters prior to conversion to a voice signal the present system and methods avoid the lossy conversion process and associated voice distortion. Since the received voice parameters are error protected during transmission they are received with greater accuracy. All, or a subset, of speech coding parameters, including, in appropriate cases, spectral envelope parameters, reflection coefficients, LSPs, LSFs, LPCs, LPCCs, and weighted LPCCs may be processed at a receiving base station or forwarded to another location for processing.

Abstract: A method and system for controlling overload at a network processor in a communications system. Messages relating to a plurality of types of calls, functions or processes (collectively, “calls”) received at a processor seeking access to processor resources are granted normal access when the delay in processing such messages does not exceed threshold values. When such delay exceeds a threshold, messages are blocked in numbers relating to, among other things, the number of messages allowed in a reference interval, message delay, the total number of blocked calls, the number of blocked calls of each type, the number of calls allowed in one or more preceding intervals and preferences between types of calls.

Abstract: A method for providing real-time multimedia conferencing services is described over a hybrid network, consisting of packet-switched nonguaranteed quality of service local area networks and a cell-switched wide area network having a multimedia bridge connected to the wide area network. The local area networks and the multimedia bridge are interconnected to the wide area network via routers. New architectural entities within the multimedia bridge execute a multimedia performance and resource analysis algorithm and determine whether the network and the bridge have sufficient resources to satisfy the requirements of the multimedia conference call. If sufficient resources are not available, it is determined whether the requirements for the higher priority services can be provided by preempting the lower priority resources.

Abstract: A real-time method for routing subject to an acceptable delay constraint between nodes in high-speed data networks, such as PNNI protocol networks, uses an illustrative two-phase algorithm employing Dijkstra's algorithm at each phase. In an illustrative first phase, the Dijkstra SPF algorithm is used in seeking the shortest cumulative delay from the destination to the source, thereby generating cumulative delay labels from a node j to the destination node k. The delay results are then employed in the second phase, where the Dijkstra SPF algorithm is illustratively employed for determining administrative weight (AW) as the link metric subject to modification in accordance with results obtained in the first phase.

Abstract: Methods for provisioning optical communications networks employing wavelength division multiplexing (WDM) technology provide for routing and wave-length assignment of a set of connection requests. Avoiding many of the pitfalls of exactly optimum solutions, the present heuristic methods provide much shorter execution times while providing near-optimum performance. Adaptations of the basic solutions provide design of restorable networks capable of handling a specified set of failures. This approach advantageously is based on considering all failures simultaneously, and performs better than solutions in which independent designs are developed for each failure. Networks designed using these methods store configuration information for primary network configurations and for one or more restoration networks.

Abstract: A finite state machine in which n discrete states of the machine are encoded using a set of n bistable elements permitting simplified decoding of states. Transitions between discrete states are illustratively executed by selectively directing a bistable device to a 1 state using transition circuits activated by positive transitions of asynchronous spiking inputs. In one illustrative embodiment the positive transition is capacitively coupled to a switch that connects the complement output of a bistable element to ground, thereby causing the nominal output of that element to approach a voltage corresponding to a 1 state. In a second illustrative embodiment a pass transistor combination maintains a capacitor at the nominal output voltage of the bistable element in a 1 state until a spiking input signal arrives. A combination of the spiking input signal and the voltage on the capacitor connects the complement output of another bistable element to ground, thus switching that other bistable element.

Abstract: A system and method for discovering association rules that display regular cyclic variation over time is disclosed. Such association rules may apply over daily, weekly or monthly (or other) cycles of sales data or the like. A first technique, referred to as the sequential algorithm, treats association rules and cycles relatively independently. Based on the interaction between association rules and time, we employ a new technique called cycle pruning, which reduces the amount of time needed to find cyclic association rules. A second algorithm, the interleaved algorithm, uses cycle pruning and other optimization techniques for discovering cyclic association rules with reduced overhead.

Abstract: A/D conversion of a current input is performed with integrate-and-fire spiking neurons. Techniques that upcount or downcount the number of spikes fired by one neuron in a time period established by another neuron yield quantized estimates of analog charge residues created by the input current. Recursive application of alternate upcounting and downcounting operations yields successively finer quantization estimates that are terminated by an error-correction operation to obtain the least significant bit of the conversion. A spike-based hybrid state machine (HSM) employing both analog and digital elements is configured to create a 2-step or a successive-subranging analog-to-digital converter. The speed of the conversion is augmented in a pipelined topology. In the HSM, a spike-triggered finite state machine (FSM) controls the input currents to the spiking neurons and is in turn controlled by spikes arising from these neurons.

Abstract: A stable, reliable, op-amp circuit overcomes the adverse affect of input offset voltages, VOSI, present at the input of op-amp. In one application, such VOSI-compensated op-amps employ a standard bandgap voltage input to achieve an improved voltage regulated reference source. A new circuit combination includes an auto-zero circuit arrangement intermediate an input network and the op-amp exhibiting the input voltage offset. In an auto-zero mode, the new auto-zero circuit arrangement samples the op amp's VOSI and redistributes the sampled signal in a storage network for use in compensating for the VOSI while operating in a normal mode. As the potential for VOSI problems again develops, another auto-zero pulse causes the circuit arrangement to switch again to the auto-zero mode and the cycle continues. Because the auto-zero arrangement employs the actual VOSI encountered in a specific op-amp, and that value for VOSI is used to null the error-causing input, the circuit arrangement is self-calibrating.

Abstract: A spiking neuron circuit providing a spiking output signal in response to an input current that causes a first capacitor to charge to a threshold voltage. In response to achieving such threshold, an output terminal is connected to a voltage, illustratively VDD, for a period determined by an applied voltage, Vpw. Rapid switching of the output to its spiking level is achieved using a positive feedback path, and deactivation of such feedback rapidly terminates the spiking period.

Abstract: A plurality of users communicate in a plurality of real-time text conversations (e.g., “chat sessions”) in a client-server message processing environment using messages including a conversation index, a conversation-initiator ID and a list of message recipients. Each conversation is maintained at client terminals in an individual window. Dropping and controlled adding of conversation participants is attended by message updates to other participants. Alternative peer-to-peer message handling reduces the processing burden on servers while allowing clients to perform control and display functions. Voice or other non-text messages are also communicated using described techniques.

Abstract: A system an method for efficiently providing access by a large number of clients to objects located at a large number of information servers. A non-bottleneck solution is provided to sharing load among servers by migrating or replicating objects over from highly loaded servers to less highly loaded servers. Objects that experience low loading are deleted to make room for more highly used objects and to permit make space for new objects. A naming service is provided to provide rapid access to a replica of a requested objects, while avoiding directing access requests to servers from which replicas of requested objects have been deleted. Hierarchical ordering of replication and naming functions permits a variety of particular access methods to be realized.

Abstract: A new method for real time implementation of rate-distortion optimized coding mode selection is disclosed that can be efficiently applied to H-263-compatible video codecs and other codecs of similar type. A normalized rate-distortion model is used to efficiently compute the rate and the distortion when encoding motion-compensated prediction error signals, instead of performing DCT, quantization and entropy-encoding. A fast algorithm is also disclosed that determines sub-optimal values of coding parameters such as the quantization parameter and the Lagrangian multiplier, .lambda., for the trellis search. Very good rate control and picture quality are achieved, especially when the disclosed techniques are applied in very low bitrate video coding.

Abstract: A method for providing real-time multimedia conferencing services is described over a hybrid network, consisting of packet-switched nonguaranteed quality of service local area networks and a cell-switched wide area network having a multimedia bridge connected to the wide area network. The local area networks and the multimedia bridge are interconnected to the wide area network via routers. New architectural entities within the multimedia bridge execute a multimedia performance and resource analysis algorithm and determine whether the network and the bridge have sufficient resources to satisfy the requirements of the multimedia conference call. If sufficient resources are not available, it is determined whether the requirements for the higher priority services can be provided by preempting the lower priority resources.