296results about How to "Effective bandwidth" patented technology

A method and system to increase the throughput of a communications system that uses an electrical power distribution system as a communications pathway, determines the phase of the power distribution power cycle and compares this determined phase to predetermined regions of the power cycle. If the power cycle is within a predetermined region, a particular communication scheme is used for transmitting and receiving information. The power cycle can have two or more predetermined regions. Optionally, the throughput for any or all regions can be determined so as to provide for modification of the associated communication scheme if the throughput is determined to be outside some predetermined range.

The present invention is directed to a call admission controller that is operable to: (a) determine at least one of (i) a bandwidth utilization level for a first path including a first link; (ii) an available bandwidth level for the first path; and (iii) one or more Quality of Service or QoS metrics for the first path; (b) compare the at least one of (i) a bandwidth utilization level; (ii) an available bandwidth level; and (iii) one or more Quality of Service or QoS metrics to one or more selected thresholds to determine whether a new live voice communication may be set up with a first selected codec; and (iii) when a new live voice communication may not be set up with the first selected codec, perform at least one of the following operations: (i) select a second different codec from among a plurality of possible codecs for the new live voice communication, wherein the second codec has a lower bit rate than the first codec; (ii) change an existing live voice communication from the first codec to the second codec; and (iii) redirect the new live voice communication from the first path to a second different path, wherein the second path does not include the first link.

A method and system for electronically delivering files over a public network is disclosed. The network includes a plurality of computers including at least one server node and multiple client nodes. In a first aspect of the present invention, the method and system enable secure and reliable peer-to-peer file sharing between two client nodes. First, a digital fingerprint is generated and associated with a file in response to the file being selected for publication on a first client node. An entry for the file is then added to a searchable index of shared files on the server node, and the fingerprint for the file is also stored on the server. In response to a second client selecting the file from the search list on the server node, the file is automatically transferred from the first client node directly to the second client node. The second client node then generates a new fingerprint for the file and compares with the new fingerprint with the fingerprint from the server node, thereby verifying the authenticity of the file and publisher. In a second aspect of the present invention, the method and system also enable subscription-based decentralized file downloads to the client nodes. First, the client nodes are allowed to subscribe with the server node to periodically receive copies of one of the files. To provide a current subscribing client node with the file, the geographically closest client node containing the file is located, and the file is transferred from the closest node directly to the current subscribing node, thereby efficiently utilizing bandwidth.

Data center network architectures that can reduce the cost and complexity of data center networks. The data center network architectures can employ optical network topologies and optical nodes to efficiently allocate bandwidth within the data center networks, while reducing the physical interconnectivity requirements of the data center networks. The data center network architectures also allow computing resources within data center networks to be controlled and provisioned based at least in part on a combined network topology and application component topology, thereby enhancing overall application program performance.

A method and apparatus for adaptively obtaining bandwidth requests in a broadband wireless communication system. The method and apparatus includes dynamically varying technique combinations enabling a plurality of users to efficiently request bandwidth from a shared base station. A user may “piggyback” a new bandwidth request upon, or set a “poll-me bit” within, presently allocated bandwidth. A base station may poll users, individually or in groups, by allocating unrequested bandwidth for new requests. Polling may respond to a “poll-me bit,” and/or it may be adaptively periodic at a rate based on communication status parameters, such as recent communication activity and connection QoS levels. Group polling permits a possibility of collisions. Polling policies may be established for dynamically varying user groups, or may be determined for each user. Dynamic selection of appropriate polling techniques makes use of efficiency benefits associated with each technique.

A DMA engine, includes, in part, a DMA controller, an associative memory buffer, a request FIFO accepting data transfer requests from a programmable engine, such as a CPU, and a response FIFO that returns the completion status of the transfer requests to the CPU. Each request includes, in part, a target external memory address from which data is to be loaded or to which data is to be stored; a block size, specifying the amount of data to be transferred; and context information. The associative buffer holds data fetched from the external memory; and provides the data to the CPUs for processing. Loading into and storing from the associative buffer is done under the control of the DMA controller. When a request to fetch data from the external memory is processed, the DMA controller allocates a block within the associative buffer and loads the data into the allocated block.

A method and system for use with wireless communication systems having a cellular architecture with at least a first and a second cell. The method and system provided ensure near uniform capacity and quality of channels within the second cell via the following steps. The noise signal power in unused data channels within the second cell is monitored. When a request for channel access is received, a determination is made whether the request for channel access is either a request for handoff from the first cell into the second cell, or not. In the event that the request is not a request for handoff, a determination is made whether idle channels exist to satisfy the request for channel access. In the event of a determination either that the request for channel access is a request for handoff, or both that the request is not a request for handoff and that idle channels exist to satisfy the request, a measured received signal power of a mobile unit subscriber unit making the request is determined. One of the unused channels in the second cell is then preferentially assigned to the mobile subscriber unit where such preference in assigning is to assign a channel, provided that a signal to noise ratio calculated upon the monitored received signal power and the monitored noise signal power of the preferentially assigned noisy channel meets or exceeds a required signal to noise ratio.

The present invention includes novel techniques, apparatus, and systems for optical WDM communications. Various wavelocker apparatus and methods are disclosed that measure the frequency offsets between signal lasers and reference lasers. The measured offsets are used to adjust the signal laser frequencies to meet their target frequencies. The absolute accuracy of the reference laser frequency is improved by measuring the absorption of the reference laser by a gas cell with known fixed absorption lines versus the reference laser frequency. Apparatus and methods are disclosed to cover scenarios in which the reference laser polarization is aligned with the signal lasers, as well as those in which the reference laser polarization is not aligned with the signal lasers. The wavelocker apparatus may or may not be located at the same network site as the signal lasers.

An apparatus for achieving an accelerator of a sparse convolutional neural network is provided. The apparatus comprises a convolution and pooling unit, a full connection unit and a control unit. Convolution parameter information and input data and intermediate calculation data are read based on control information, and weight matrix position information of a full connection layer is also read. Then a convolution and pooling operation for a first iteration number of times is performed on the input data in accordance with the convolution parameter information, and then a full connection calculation for a second iteration number of times is performed in accordance with the weight matrix position information of the full connection layer. Each input data is divided into a plurality of sub-blocks, and the convolution and pooling unit and the full connection unit perform operations on the plurality of sub-blocks in parallel, respectively.

The present invention relates to systems and methods for efficient bit rate alteration of a bitstream to match an available channel capacity. The efficient bit rate alteration includes selective re-quantization of the compressed bitstream. Selective re-quantization according to the present invention applies multiple re-quantization schemes to different portions of a bitstream. In one embodiment, the multiple re-quantization schemes each have a different computational load. By selectively choosing which type of re-quantization is performed on each portion, efficient bandwidth scaling and data transmission may be achieved both when computational capacity is limited and when video data integrity is important.

A dynamic content routing network routes update messages containing updates to properties of live objects from input sources to clients. The clients receive a web page having live objects, identify the object IDs associated with the objects, and contact an object state storage to obtain update messages specifying the objects'initial properties. The clients register the object IDs with the routing network. The routing network maintains a registry of object IDs and clients. The input source provides an update message to the routing network containing the object ID and data for updating a property of the object. The routing network routes update messages from the input source to the clients registered for the object ID contained in the message. Upon receipt of the message, a client updates the specified property of the live object. The update messages are also provided to, and stored by, the object state storage.

A method and apparatus for allocating bandwidth in a broadband wireless communication system is disclosed. One embodiment uses a self-correcting bandwidth request/grant protocol. The self-correcting bandwidth request/grant protocol utilizes a combination of incremental and aggregate bandwidth requests. CPEs primarily transmit incremental bandwidth requests to their associated base stations, followed by periodic transmissions of aggregate bandwidth requests. The use of periodic aggregate bandwidth requests (that express the current state of their respective connection queues) allows the bandwidth allocation method and apparatus to be “self-correcting”. Another embodiment utilizes an abridged bandwidth request/grant protocol to allocate bandwidth. The abridged bandwidth request/grant protocol system utilizes padding packets to request a reduction in bandwidth allocation to a CPE. A base station modem alerts a base station CPU when the BS modem receives a padding packet from a CPE. After alerting the BS CPU the method can reduce the associated CPE's bandwidth allocation.

A method and apparatus for adaptively obtaining bandwidth requests in a broadband wireless communication system. The method and apparatus includes dynamically varying technique combinations enabling a plurality of users to efficiently request bandwidth from a shared base station. A user may “piggyback” a new bandwidth request upon, or set a “poll-me bit” within, presently allocated bandwidth. A base station may poll users, individually or in groups, by allocating unrequested bandwidth for new requests. Polling may respond to a “poll-me bit,” and/or it may be adaptively periodic at a rate based on communication status parameters, such as recent communication activity and connection QoS levels. Group polling permits a possibility of collisions. Polling policies may be established for dynamically varying user groups, or may be determined for each user. Dynamic selection of appropriate polling techniques makes use of efficiency benefits associated with each technique.

A method and apparatus for adaptively obtaining bandwidth requests in a broadband wireless communication system. The method and apparatus includes dynamically varying technique combinations enabling a plurality of users to efficiently request bandwidth from a shared base station. A user may “piggyback” a new bandwidth request upon, or set a “poll-me bit” within, presently allocated bandwidth. A base station may poll users, individually or in groups, by allocating unrequested bandwidth for new requests. Polling may respond to a “poll-me bit,” and/or it may be adaptively periodic at a rate based on communication status parameters, such as recent communication activity and connection QoS levels Group polling permits a possibility of collisions. Polling policies may be established for dynamically varying user groups, or may be determined for each user. Dynamic selection of appropriate polling techniques makes use of efficiency benefits associated with each technique.

Various types of passive optical networks operate simultaneously in one passive optical network system comprising an optical line terminal, a passive remote node, and multiple optical network units. Downstream data is orthogonal frequency division multiplexed onto a single wavelength optical carrier transmitted on a primary downstream optical beam from the optical line terminal to a splitter in the passive remote node. The primary downstream optical beam is split into multiple secondary downstream optical beams; each is transmitted to a specific optical network unit. Upstream data is orthogonal frequency division multiplexed onto a single wavelength optical carrier transmitted on a secondary upstream optical beam from each optical network unit to a coupler in the passive remote node. The upstream wavelength for each optical network unit is different. The wavelength division multiplexed optical beam is transmitted from the passive remote node to a parallel signal detector in the optical line terminal.

A method, system and device for enabling efficient bandwidth utilization of a multipoint network service over an arbitrary topology network that includes a plurality of network elements (NEs). In a preferred embodiment, the method comprises the steps of setting up a full connectivity between the NEs and providing the multipoint network service using the full connectivity, whereby data packets of the multipoint network services are transmitted from a source NE to at least one edge NE through at least one intermediate NE, and whereby data packets that need to be flooded are not replicated or at the source NE. The full connectivity includes an arbitrary combination of a first plurality of point-to-multipoint connections between each source NE and each edge NE and a second plurality of point-to-point connections between each source NE and every edge NE.

A system and method for switching between transmission modes provides more efficient use of available bandwidth. A content delivery system determines, based upon a predetermined limit, whether to broadcast content to a plurality of mobile platforms or unicast the data content via a point-to-point communication link. Acknowledgment signals from the mobile platforms are used to determine if the predetermined limit has been exceeded. A specific number or percentage defines the predetermined limit within a specified time period. The number of acknowledgment signals received is compared to the limit to determine if an exceedance condition has occurred.

A dynamic content routing network routes update messages containing updates to properties of live objects from input sources to clients having the objects. Certain objects served to clients by a server are indicated as “live.” When the clients receive live objects, the clients identify the object IDs associated with the objects and register the object IDs with the routing network. The routing network maintains a registry of object IDs and clients. An input source provides an update message to the routing network containing the object ID and data specifying an update to a property of the object. The routing network routes the message to each client that has registered for the object ID contained in the message. Upon receipt of the message, a client updates the specified property of the live object.

A method, device, and system for providing multipoint network services over a ring topology network. According to the disclosed method point-to-point connections and point-to-multipoint connections are established between network elements (NEs) providing the same network services. These connections eliminate the need for replicating packets at a source NE and, as a result, the bandwidth utilization is significantly improved. In one embodiment of this invention the multipoint network services provided by the present invention is a virtual private LAN service (VPLS).