Abstract: The invention relates to a method for processing call signalling messages during a communications switch overload condition. The method integral to the invention comprises a dual threshold test, one based on message latency and the other based on queue occupancy. The method begins with the receipt of a plurality of call signalling messages comprising dispensable, indispensable and essential messages. The processing then continues by comparing a new message expected latency with a latency-based threshold. If the new message expected latency compares unfavourably, the queue occupancy is then compared with a queue occupancy based threshold. If this comparison is also unfavourable the call signalling messages are processed based on call type. Using the expected latency as a threshold ensures that no dispensable message has a latency greater than a tunable threshold.

Abstract: A network management protocol for the efficient retrieval of items of information or objects from managed elements or other network management systems in a digital communications network is described. The new protocol introduces one or more qualifier bits appended to an information retrieval command wherein the qualifier bit specifies the subclasses of the objects requested.

Abstract: A base station power control for dynamically introducing transmission power to compensate for rain fade in a broadband wireless system operating at millimeter wave frequencies. The base station periodically polls a selected network interface unit at a customer site to obtain status information such as the power level of the received signal. Variations in the power level of the received signal, indicative of variations in the attenuation of the wireless link, are used by the base station power controller to appropriately adjust transmission power. Status information collected by the base station can also include signal to noise level data and the bit error rate.

Abstract: A method is disclosed for establishing a switched virtual circuit in a digital network having network nodes with static routing tables. The static routing tables contain at least primary and alternate routing data. When a node is unable to forward a call over its outgoing primary route due to congestion or physical failure and its alternate route is the same as the route on which a call setup request arrived, it clears the call at that node and sends a crankback message to the preceding node, which responds to the crankback message to attempt to dynamically re-route the call over the alternate route stored in the routing table of the preceding node. If the attempt is unsuccessful, it sends the message back to the next preceding node and so on.

Abstract: A method is disclosed for traffic policing in packet-based, broadband networks, such as ATM networks. The method comprises the steps of monitoring the arrival of incoming cells, and advancing a reference clock on a step of I if an arrival is late but not later than I, where I is an increment, and if an arrival is later than I assuming a new busy period and setting the reference clock at the arrival time.

Abstract: A method and apparatus for transporting data include processing that begins by receiving a bit stream of data at a rate of at least a bit per input interval. The bit stream of data corresponds to ingress packetized data that has been packetized based on one of a plurality of data packetizing protocols (e.g., HDLC encapsulated data, frame relay, PPP, and/or SMDS). As the bit stream is received, the data is extracted from the bit stream to produce extracted data, which is stored in ingress local memory. When a sufficient amount of extracted data is stored in the ingress local memory (i.e., an ingress data word is stored), it is transported to a non-local memory, where the transporting is based, at least in part, on content of the ingress data word. The processing also includes transporting stored egress data words from a non-local memory to an egress local memory based, at least in part, on content of the egress data words.

Abstract: A rail tracking system and method for providing precise tracking of voltage levels to a dual supply voltage Integrated Circuit. A switch mode DC—DC voltage regulator is used to derive the lower of the two voltage levels from the higher level. The switch mode regulator employs a pulse width modulator (PWM) to derive the lower voltage level. A separate supply source is utilized to power the PWM and the timing of the supply voltage is such that the PWM has reached steady state before the higher voltage level is provided to the regulator.

Abstract: A method of clock recovery from a fast packet switched asynchronous network wherein a time stamp is transmitted over a network that has variable delay, involves maintaining an input counter at a sending device. The counter has a value dependent on an input clock. The value of the input counter is periodically transmitted over the network to a receiving device. A local counter at a receiving device has a value dependent on a local clock. The received value sample x.sub.i is compared with the local sample y.sub.i to derive the difference e.sub.i. The local clock frequency f.sub.j is adjusted such that the average of Emin.sub.j, where Emin.sub.j is the minimum of a block of error samples e.sub.i, remains close to a predetermined value, preferably zero. The clock recovery method is particularly suitable for MPEG2 video transmitted over ATM.

Abstract: A system and method for adaptively assigning queue service weights to an ATM traffic management controller in order to reduce service weight update calculations. The implementation of a fair weighted queuing scheme on a network is divided into two functions: queue arbitration; and queue weight configuration. Queue arbitration is performed via a virtual time-stamped fair queuing algorithm. The queue weight configuration is performed via a reconfigurable weighted fair queuing controller wherein bandwidth requirements are calculated in response to connection setup and release values.

Abstract: An apparatus and method for processing multiple cell streams of differing service categories at a queuing point in an ATM network. The apparatus includes two or more cell processors for monitoring bi-directional ABR traffic at each queuing point in order to efficiently calculate and mark explicit rate values and/or congestion information in resource management cells in either VSVD or non-VSVD ABR traffic.