Abstract:
A network switch includes at least one port processor and at least one switch element. The port processor has an SONET OC-x interface (for TDM traffic), a UTOPIA interface (for ATM and packet traffic), and an interface to the switch element. In one embodiment, the port processor has a total I/O bandwidth equivalent to an OC-48, and the switch element has 12×12 ports for a total bandwidth of 30 Gbps. A typical switch includes multiple port processors and switch elements. A data frame of 9 rows by 1700 slots is used to transport ATM, TDM, and Packet data from a port processor through one or more switch elements to the same or another port processor. Each frame is transmitted in 125 microseconds; each row in 13.89 microseconds. Each slot includes a 4-bit tag plus a 4-byte payload. The slot bandwidth is 2.592 Mbps which is large enough to carry an E-1 signal with overhead. The 4-bit tag is a cross connect pointer which is setup when a TDM connection is provisioned. The last twenty slots of the frame are reserved for link overhead. Thus, the frame is capable of carrying the equivalent of 1,680 E-1 TDM signals. For ATM and packet data, a PDU (protocol data unit) of 16 slots is defined for a 64-byte payload. The PDUs are self-routed through the switch with a 28-bit routing tag which allows routing through seven switch stages using 4-bits per stage. Bandwidth is arbitrated among ATM and Packet connections while maintaining TDM timing.
Abstract:
An asynchronous data transfer and source traffic control system includes a bus master and a plurality of bus users coupled to a bidirectional data bus. The bus master provides two clock signals to each bus user, a system clock and a frame clock. The frame clock designates the start of a frame. A frame format preferably includes fifteen or sixteen system clock cycles, the first of which is designated the request field and the last of which includes a grant field. One or more other cycles may be assigned control and/or routing information and the remainder of the cycles comprise a data field of fixed length. During the request field, any number of bus users may request access which is received by the bus master. During the grant field, the bus master grants access to a selected bus user for the entire data portion of the next frame. Which user is granted access to the next frame is determined according to an arbitration algorithm in the bus master which may be unknown to the bus users. The asynchronous data transfer and source traffic control system has particular application in accommodating the transfer of the contents of ATM cells used in BISDN systems.
Abstract:
A DS1 dejitter circuit has a control circuit for generating six pulses over a one hundred and ninety three 1.544 Mb/sec clock cycle, and a clock circuit for tracking the frequency of a jittered incoming DS1 signal, and based on that frequency, and utilizing the six pulses, generating a clean DS1 signal at the nominal rate of the jittered incoming signal. The control circuit preferably includes a divide by 28 or 29 circuit which receives a 44.736 Mb/sec (DS3) input clock signal, a mod 193 counter, and a count decoder for providing the six control pulses over the 193 count. Logic circuitry is provided for taking the outputs from the count decode and controlling the divide block to guarantee that the divide block divides the DS3 signal by 29 one hundred eighty-eight times for every five times the divide block divides the DS3 signal by 28. In this manner an average clock of 1.544 Mb/sec (the standard DS1) rate is obtained from the divide block. The clock circuit includes a FIFO, a clock rate control circuit and another divide by 28 or 29 block. The FIFO receives the incoming jittered DS1 signal. The clock rate control circuit senses how full the FIFO is, and using that information along with the pulses from the control circuit regulates the divide block to divide the DS3 signal by 28 either four, five, or six times in a one hundred ninety-three clock cycle. Where it is desired to dejitter a plurality of DS1 signals, a single common control circuit can be used to supply the six control pulses to a plurality of clock rate control circuits.
Abstract:
There is disclosed an apparatus for enabling one to increase the bandwidth of a multi-channel digital switching system. Essentially, the system allows one to set up multi-channel link paths to obtain a total band width of N times the bandwidth of a typical switching path in a digital switching network. To do this, a common frame counter is located in the unused bit positions of the sample in each of the paths, forming a wideband channel. At the receive end, the apparatus de-skews the received samples to properly reassembly the output. This operation is provided by utilizing one content addressable memory and an associated random access memory (CAM/RAM) (C/R) for each channel to be joined. The width of each C/R is twelve bits, eight for data, the read-out section and four for frame counter, which is the associative section. The length of each C/R is a function of the maximum skew to be counted. A four bit output frame counter provides the associative addresses for the C/R which, when given an associative address corresponding to a frame number, will respond with an eight bit data word for that frame or with an Empty signal, meaning that no word exists for that frame number. At initialization the frame counter presets to one and all C/Rs are interrogated to see if they contain frame one information. If they do not, then the output frame counter is indexed and all C/Rs are re-interrogated at a fast rate until they all indicate not Empty. At this point, the outputs of each C/R for the frame count indicated are assembled into an output word. Thus as long as all the C/Rs indicate not Empty, frame integrity is insured, and hence the distributed control switching system can provide wideband operation via multi-channel linked paths while insuring frame integrity.
Abstract:
A system is disclosed which permits information received from any data terminal or telephone subscriber line at any interface to a digital network to be transmitted to any number of other terminals at any or all of the interfaces to the digital network. The system herein described also permits any terminal at any interface of the digital network to be conferenced, with up to N other terminals at any or all of the interfaces to the digital network where N can be made to depend solely upon the topology and single path transmission delay of the digital network itself. A facility is provided which is independently provisioned at every interface to the digital network and which is independently and concurrently available to any number of combinations of terminals and simplex paths at those interfaces. A broadcast capability is also provided whereby telecommunication information can be simultaneously transmitted from a source terminal to a plurality of destination terminals.
Abstract:
A network switch includes at least one port processor and at least one switch element. The port processor has an SONET OC-x interface (for TDM traffic), a UTOPIA interface (for ATM and packet traffic), and an interface to the switch element. In one embodiment, the port processor has a total I/O bandwidth equivalent to an OC-48, and the switch element has 12×12 ports for a total bandwidth of 30 Gbps. A typical switch includes multiple port processors and switch elements. A data frame of 9 rows by 1700 slots is used to transport ATM, TDM, and Packet data from a port processor through one or more switch elements to the same or another port processor. Each frame is transmitted in 125 microseconds; each row in 13.89 microseconds. Each slot includes a 4-bit tag plus a 4-byte payload. The slot bandwidth is 2.592 Mbps which is large enough to carry an E-1 signal with overhead. The 4-bit tag is a cross connect pointer which is setup when a TDM connection is provisioned. The last twenty slots of the frame are reserved for link overhead. Thus, the frame is capable of carrying the equivalent of 1,680 E-1 TDM signals. For ATM and packet data, a PDU (protocol data unit) of 16 slots is defined for a 64-byte payload. The PDUs are self-routed through the switch with a 28-bit routing tag which allows routing through seven switch stages using 4-bits per stage. Bandwidth is arbitrated among ATM and Packet connections while maintaining TDM timing.
Abstract:
Methods for retiming and realigning SONET signals include demultiplexing STS-1 signals from an STS-3 signal, buffering each of the three signals in a FIFO, determining the FIFO depth over time, determining a pointer leak rate based in part on FIFO depth and also based on the rate of received pointer movements. For a 28-byte deep FIFO, if the depth of a FIFO is 12-16 bytes, no pointer leaking is performed. If the depth is 0-4 bytes, an immediate positive leak is performed. If the depth is 24-28, an immediate negative leak is performed. If the depth is 5-11 bytes a calculated positive leak is performed. If the depth is 17-23 bytes, a calculated negative leak is performed. The calculated leak rates are based on the net number of pointer movements (magnitude of positive and negative movements summed) received every 32 seconds (256,000 frames).
Abstract:
A two stage desynchronizer is provided to receive a gapped data component of an STS-3C (STM-1) signal and provide therefrom an ungapped DS-4NA (E4) data signal. The first stage includes a data byte formation block which takes the gapped STS-3C payload data and formulates the data into bytes, a first FIFO which receives the bytes, and a first FIFO read controller which utilizes the STS-3C clock signal and causes bytes of data to be read out according to a schedule which reads bytes eight or nine times out of every ten STS-3C clock cycles. For each row (270 byte times) of the STS-3C frame, either 241 or 242 bytes are read out of the FIFO according to a slightly gapped schedule where the reading of the 242nd byte at least partially depends upon the number of stuffs in the signal and the pointer movements received. The second stage of the desynchronizer includes a second FIFO, a FIFO fullness measurement block, and a VCXO. The FIFO fullness measurement block uses the incoming slightly gapped byte clock and the ungapped DS-4NA output clock as inputs for effectively measuring the relative fullness of the second FIFO, and provides a control signal based on the relative fullness. The control signal is fed to the voltage controlled crystal oscillator (VCXO) which generates the ungapped DS-4NA or E4 clock in response thereto.
Abstract:
A digital clock dejitter circuit has a RAM for receiving an incoming gapped signal, a digital, fractional RAM fullness gauge for tracking the average input and output rates to and from the RAM and for generating therefrom a control indication, and a controllable digital frequency generator for receiving a fast clock signal and the control indication, and for providing therefrom a substantially jitter-free clock signal at the same nominal rate as the incoming gapped signal. The RAM fullness gauge has write and read counters which track the movement of data into and out of the RAM, and a subtractor for taking the difference of the counters to obtain the integer value of the RAM depth. The controllable digital frequency generator has an adder, a register, and a fast clock counter (FCC) which provides the fullness gauge with a fractional digital indication of the RAM depth. The adder has a carry output fed to the FCC to control whether the FCC divides by x or x+1, and a remainder output fed to the register and then fed back as an input to the adder. The adder also receives the control indication from the fullness gauge as an input. FCC inputs include the fast clock, and the carry output of the adder. The FCC outputs are a read signal for causing a byte to be read from the RAM at the end of a count cycle, and the fast clock count used for fractional fullness.
Abstract:
Clock dejitter circuits are provided and comprise control circuits for generating a plurality of pulses over a clock cycle, and clock circuits for tracking the speeds of jittered incoming data signal and based on those speeds, and utilizing the plurality of pulses generating substantially unjittered data signals at the nominal rates of the jittered incoming signals. A control circuit broadly includes a divide by value x-divide by value x+1 circuit which receives a fast input clock signal, a modulus y counter, and a count decode for providing z control pulses over the count of y, and a logic gate for taking the outputs from the count decode and controlling the divide block to guarantee hat the divide block divides the fast input clock signal by value x q times for every r times the divide block divides the fast input clock signal by value x+1; wherein q plus r equals y, and z equals either q+1 or r+1.