Abstract: A technique for synchronizing clocks in a network is disclosed. In one exemplary embodiment, the technique may be realized as a method for synchronizing clocks in a network. The method comprises receiving a first timestamp and a second timestamp, each indicating a respective time instance as determined by a first clock signal within the network. The method also comprises measuring a first time interval between the first timestamp and the second timestamp. The method further comprises generating a difference signal representing a difference between the first time interval and a second time interval, and generating a second clock signal based upon the difference signal such that the second clock signal is synchronized with the first clock signal.

Abstract: A system and method for synchronizing a local clock to a reference clock using a linear model of the clock error between the local clock and the reference clock is disclosed. In one embodiment, a double-exponential smoothing process is used in conjunction with the linear model to estimate a frequency offset by which the frequency of an oscillator of the local clock is adjusted. Also disclosed herein is a phased-lock loop (PLL) adapted to synchronize a local clock with a reference clock using the double-exponential smoothing process, as well as a system implementing the PLL for timing the playout of data received from a transmitter.

Abstract: A first level of control over operation of slave Digitally Controlled Frequency Selectors (DCFSs), such as DCOs or DDSs, may occur by periodic transmission of control words from the master clock to the slave clocks. To allow enhanced control over the output of the slave clocks, the frequency of the local oscillator used to generate the synthesized output of the master clock may also be conveyed to the slave clocks to allow a second level of control to take place. The second level of control allows the local oscillators at the slave clocks to lock onto the frequency of the master local oscillator to thereby allow the slave local oscillators to operate the slave DCFSs using the same local oscillator frequency. The first level of control synchronizes operation of the DCFSs while the second level control prevents instabilities in the local oscillators from causing long term drift between the slave and master clock outputs. Timestamps may be used to synchronize the master and slave local oscillators.

Abstract: Network elements may be synchronized over an asynchronous network by implementing a master clock as an all digital PLL that includes a Digitally Controlled Frequency Selector (DCFS), the output frequency of which may be directly controlled through the input of a control word. The PLL causes the control word input to the master DCFS to be adjusted to cause the output of the master DCFS to lock onto a reference frequency. Information associated with the control word is transmitted from the master clock to the slave clocks which are also implemented as DCFSs. By using the transmitted information to recreate the master control word, the slaves may be made to assume the same state as the master DCFS without requiring the slaves to be implemented as PLLs. The DCFS may be formed as a digitally controlled oscillator (DCO) or as a Direct Digital Synthesizer (DDS).

Abstract: A method of encoding a plurality of pre-defined codes into a search key and a method of using the search key to locate a longest matching pre-defined code to a given code is disclosed. Encoding the pre-defined codes into a search key involves producing a prefix node bit array (PNBA) having a plurality of bit positions corresponding to possible bit combinations of a bit string having a length equal to or less than the longest predefined code in said plurality of said pre-defined codes such that said bit positions are arranged by the lengths of said possible bit combinations and by numeric value of said possible bit combinations and to setting bits active in bit positions which correspond to bit combinations identified by said pre-defined codes.

Abstract: A method and apparatus are provided for adjusting the volume of data communicated between a transmitter and a receiver on a network, in a time interval. The method involves producing a desired volume value in response to a volume value specified by the receiver and a difference between a target departure volume and an estimate of arrival volume of data at a queue through which data passes from the transmitter to the receiver, and communicating the desired volume value to the transmitter in response to an acknowledgement signal produced by the receiver.

Abstract: A technique for managing a queue so as to distribute losses among different service classes is disclosed. In one embodiment, the technique is realized by classifying an incoming packet into one of a plurality of classes. Each class has an associated weighting factor. The system continuously monitors a queue size and determines an overall packet drop probability based on the actual queue size and a target queue size and calculates a target class drop probability based on the overall packet drop probability and the weighting factor. Finally, the system makes a comparison based on the target class drop probability and a selected value and decides whether to drop the incoming packet based on a result of the comparison. If losses are unavoidable in the system, the technique ensures that the losses will be distributed among the different service classes in inverse proportion to the service price of each class.

Abstract: An apparatus, method, computer-readable medium and signals for connection-class parameter control of packet flow, involves indicating a desired packet flow from a packet source to a network element having a queue, by communicating to the packet source an indication of a desired number of packets to be received from the source in a time period, in response to a class parameter associated with a class of connections supported by the network element and queue conditions at the network element.

Abstract: A method and apparatus for locating in a list of pre-defined codes, a longest code matching a given code. The method and apparatus involve producing a search mask encoding at least one portion of said given code and comparing the search mask to a search key having a Prefix Node Bit Array (PNBA) in which a bit is set active in at least one of a plurality of bit positions corresponding to possible bit combinations of bits in a bit string having a length equal to or less than the longest predefined code in the plurality of pre-defined codes and arranged by the lengths of possible bit combinations and by numeric values of the bit combinations, to identify a common active bit position in the search key and the search mask corresponding to one of the pre-defined codes having a length greater than all others of said pre-defined codes which correspond to common active bit positions.

Abstract: A technique for an improved active queue management scheme which dynamically changes its threshold settings as the number of connections (and system load) changes is disclosed. Using this technique, network devices can effectively control packet losses and TCP timeouts while maintaining high link utilization. The technique also allows a network to support a larger number of connections during congestion periods.

Abstract: A technique for adaptively controlling a gain of an optical amplifier is disclosed. In one particular exemplary embodiment, the technique may be realized by a method comprising the steps of measuring an output power signal of the optical amplifier; computing a command signal from an input power signal; computing a model state signal of a reference model; computing a model output signal of the reference model based at least in part on the model state signal; computing an error signal between the model output signal and the output power signal of the optical amplifier; adjusting at least one adaptive control gain value to minimize the error signal; and computing a control input signal for driving the optical amplifier based at least in part on the at least one adaptive control gain value, the error signal, the model state signal and the command signal.

Abstract: A rotator switch including more tandem buffers than inputs is disclosed. An input data conditioner formats data to be transferred from the multiple inputs to the tandem buffers. Excess tandem buffers allow data to be transferred from inputs to tandem buffers at a rate less than the rate at which data arrives at the inputs. Excess capacity of the switch fabric may be used to carry overhead, or slow the rate at which data is transferred to the switch fabric.

Abstract: An improved buffer management process is disclosed wherein the buffer is shared among a plurality of packet queues. The improved buffer management process comprises computing a common queue threshold value based upon the aggregate size of the plurality of packet queues and a predetermined buffer threshold value. The common queue threshold value is then used to manage the size of each of the plurality of packet queues and thereby manage the buffer.

Abstract: A method and apparatus for controlling a rate at which packets are transmitted to a network node having a queue and a network element employing same is described. The method and apparatus cooperate to generate a probability value in response to queue occupancy and desired queue occupancy and to control the packet in response to the probability value and a reference value to cause a source of the packet to change its transmission rate. Control of the packet is achieved by dropping the packet or by modifying the packet to include a message informing the source that it should adjust its transmission rate. In one embodiment, the probability value is converted into an inter-drop/mark value representing the number of packets to be admitted into the queue between dropped or marked packets.

Abstract: A method of allocating bandwidth for transmitting upstream cells from a CPE unit to a BTS. The CPE determines arrival time information associated with each cell and sends arrival time information associated with at least one cell to the BTS. The BTS allocates future bandwidth to the CPE as a function of the arrival time information received from the CPE. For example, the BTS estimates the number of cell arrivals occurring in the current scheduling period based on past cell arrival times. The BTS then sends information to the CPE in which the allocated future bandwidth is specified. Finally, the CPE groups a number of cells into a burst packet occupying the allocated bandwidth and sends the burst packet to the BTS. Rather than respond in a delayed manner to a bandwidth deficiency or surplus at an individual CPE unit, the BTS produces an estimate of a CPE unit's bandwidth demands, resulting in more efficient bandwidth utilization and reduced cell loss ratio.

Abstract: A flow control system and methodology controls and avoids congestion in an Asynchronous Transfer Mode (ATM) network. The congestion state of a network switch is determined based on incoming Available Bit Rate (ABR) traffic, Constant Bit Rate (CBR) traffic, and Variable Bit Rate (VBR) traffic. High-frequency fluctuations due to VBR traffic are filtered out of the CBR/VBR traffic to determine the underlying trend of network traffic. A filtering mechanism is based on wavelet transforms. The switch congestion state is determined based on the current ABR traffic, the filtered CBR/VBR traffic, and the current status of buffers in the switch. The congestion state can then be used to adjust the ABR rate according to a feedback scheme such as Explicit Forward Congestion Indication (EFCI).

Abstract: Flow control in a network is implemented based on aggregate traffic measurements. For example, in an ATM network only the aggregate background (CBR/VBR) traffic rate and the aggregate ABR traffic rate are used, in contrast with other schemes that require per-connection rate measurements or variables. An explicit rate is calculated recursively at discrete time instances using a scaled error value which is generated in response to an aggregate ABR input rate and a desired traffic rate. Explicit rate computations can be performed entirely by software, and the interval between computations is large enough to keep the processing overhead required of the switch very low. In addition, methods consistent with the present invention achieve max-min fairness and MCR plus equal share in a natural way without any additional computation or information about bottleneck rates.

Abstract: A rotator switch includes active and redundant tandem buffers used to switch information units through the switch. The tandem buffers are interconnected to inputs and outputs by way of commutators. Information provided to one or more of the active buffers is provided to a redundant buffer. If an active tandem buffer fails, the switch may use the redundant buffer in place of the failed active buffer. A further rotator switch is formed using a plurality of switching blocks. Each switching block includes its own set of tandem buffers, and ingress and egress commutators to interconnect the tandem buffers to inputs and outputs. The plurality of switching blocks is interconnected using additional commutators. At least one of these switching blocks is redundant and is provided with information provided to an active switching block. In the event of failure of an active block, the switch may use a redundant switching block in place of the failed or replaced active block.

Abstract: A technique for synchronizing clocks in a network is disclosed. In one exemplary embodiment, the technique may be realized as a method for synchronizing clocks in a network. The method comprises receiving a first timestamp and a second timestamp, each indicating a respective time instance as determined by a first clock signal within the network. The method also comprises measuring a first time interval between the first timestamp and the second timestamp. The method further comprises generating a difference signal representing a difference between the first time interval and a second time interval, and generating a second clock signal based upon the difference signal such that the second clock signal is synchronized with the first clock signal.

Abstract: A technique for an improved active queue management scheme which dynamically changes its threshold settings as the number of connections (and system load) changes is disclosed. Using this technique, network devices can effectively control packet losses and TCP timeouts while maintaining high link utilization. The technique also allows a network to support a larger number of connections during congestion periods.