Abstract: A power-saving method for a first-in-first-out (FIFO) buffer implemented in a memory. The memory is segmented into a plurality of logical segments. For each logical segment, for each power saving mode, a recovery time and recovery overhead to an operational mode, and a transition overhead for transitioning the logical segment into the power saving mode, are determined. During each clock cycle, a determination is made as to whether a net power saving will result by entering each logical segment into a power saving mode based on a minimum time before a read or write pointer will enter the logical segment as well as the recovery time, the recovery overhead, and the transition overhead. The logical segment is transitioned to the power saving mode only if a net power saving will result, and is returned to the operational mode when the minimum time is no longer greater than the recovery time.

Abstract: A method and apparatus for timing optimization are disclosed, which rely on information gathered from a timing detection circuit to find the optimal sampling point of a data recovery system. In an implementation, a timing shift is optimized based on Gardner detector data. In an example, a Gardner detector, or an early and late extraction portion thereof, is added to the data path, and the data path clock is shifted so that it is centered on the data transition mean. In an implementation, the sampling point of the data path is optimized for better horizontal eye opening using a Gardner detector's output for centering the average crossing point of different paths. In an example embodiment, an apparatus is provided with a second clock recovery that is not completely independent of a first clock recovery.

Abstract: A distributed electrostatic discharge (ESD) protection circuit is provided. At frequencies beyond 10 GHz, the parasitic capacitance of primary ESD protection voltage clamping devices, such as diodes, hampers adequate insertion and return loss, in spite of lumped inductor tuning. An ESD protection circuit according to an embodiment of the present disclosure solves the problem by distributing the diode, or voltage clamping device, capacitance among several sections of an artificial transmission line. The power and ground ESD return paths are also distributed to ensure a constant current density in the voltage clamping segments, even for fast charged-device model (CDM) discharge events. By sharing the ESD return paths between differential inputs (or outputs), these return paths have no impact on differential return or insertion loss.

Abstract: A SerDes receiver comprising: an input for receiving a signal, the signal having a baud rate; an Analog Finite Impulse Response equalizer (AFIR) for equalizing the received signal, the AFIR comprising: a pre-cursor tap having a pre-cursor coefficient; a cursor tap having a cursor coefficient, the cursor coefficient being constrained to a non-negative value; and a post-cursor tap having a post-cursor coefficient; an adaptation block coupled to the AFIR, the adaptation block configured to adjust the pre-cursor coefficient and the post-cursor coefficient based on the received signal, the adaptation block further being configured to constrain the values of the pre-cursor and post-cursor coefficients to be non-positive.

Abstract: A system and method for automatically generating a test script includes receiving a test case flow that includes steps, nodes, and sub-nodes, wherein each sub-node is associated with a use-case based application programming interface (UC-API), for each sub-node of the test case flow retrieving a template array corresponding to the UC-API associated with the sub-node, generating a test array wherein for each node in a step, generating a node array wherein the elements of each node array includes the sub-node arrays associated with the sub-nodes of the node, for each step, generating a step array wherein the elements of each step array include references to the node arrays of the nodes in the step, and populating the test array wherein each element of the test array includes one of the step arrays.

Abstract: A high efficiency search table is implemented with a multiple hash algorithm. The search table allows for exact match searching of arbitrary data sets with fixed latency. The probability of collisions from the hash algorithms is reduced through the use of oversized pointer tables allowing for a level of indirection between hash values and table entries. In the event of a collision in all hash functions, a firmware assisted cuckoo algorithm is employed to resolve the collision.

Abstract: A transmit (TX) signal path circuit in a multiple-input, multiple-output (MIMO) transceiver responsive to a digital front end (DFE) for generating receive (RX) path phase alignment signals is disclosed. A digital up-conversion block uses a first numerically-controlled oscillator (NCO) for generating digital intermediate frequency (IF) signals for ordinary TX signal generation, and a different, second NCO for generating digital IF signals for RX phase alignment signal generation. An RF up-conversion block uses a TX local oscillator (LO) for generating analog RF signals for ordinary TX signal generation, and a different feedback (FB) LO for generating analog RF signals for RX phase alignment signal generation. Thus, phase alignment of the circuitry used for ordinary TX signal generation is left undisturbed by RX phase alignment signal generation.

Abstract: A system and method are disclosed for a flexible routing engine in a PCIe switch. The system may include a switch manager that is enabled, through firmware, to configure one or more routing tables associated with a switch stack of a PCIe switch. To enable non-transparent bridging and non-standard routing, the method may include receiving a transaction layer packet at a translated routing port of a PCIe switch, and performing translation of the address and requester ID of the packet utilizing tables that are updated by the firmware of the switch manager to route the packet through the switch.

Abstract: Methods and circuits for equalizing a linear response in an observation path of a digital pre-distorter. A method comprises generating observed signals in an observation path based on observing a transmit signal; down-converting the observed signals into intermediate frequencies using different LO frequencies; calculating a ratio using the intermediate frequencies; and equalizing the linear response of the observation path on the observed signals using the ratio. An apparatus comprises a directional coupler for observing a transmit signal and generating observed signals; a down-converter for converting the observed signals into intermediate frequencies using different LO frequencies; and an adaptive estimator for calculating a ratio using the intermediate frequencies and using the ratio to equalize a linear response from the observation path on the observed signals.

Abstract: A method and apparatus are provided for generating Generic Mapping Procedure (GMP) stuff/data decisions, which avoids brute force modulo arithmetic and is efficient for hitless adjustment of ODUFIex (G.7044) in an Optical Transport Network (OTN). Addition operations are used, rather than multiplication operations, to facilitate faster and less computationally expensive calculation of data/stuff decisions, based on calculated residue values. Residue values are logically arranged in rows to facilitate residue calculation, such as based on relationships with previously calculated residue values. This method is also applicable for mapping and de-mapping Constant Bit Rate (CBR) clients into and from an ODUk carrier.

Abstract: A method for estimating a carrier frequency offset over a dispersive but spectrally flat channel comprises determining an autocorrelation of a received oversampled complex baseband digital signal, and estimating the carrier frequency offset based on an angle of the determined autocorrelation.

Abstract: A low-density parity check (LDPC) decoder is provided for decoding low-density parity check (LDPC) encoded data wherein a layer specific attenuation factor is provided for each layer of the LDPC parity check matrix. An attenuation factor matrix comprising a plurality of coefficients specifies the specific attenuation factor for each layer and each iteration of the decoding process. A check node processor performs check node processing for each layer of the parity check matrix associated with the LDPC encoded codeword utilizing the normalized layered min-sum algorithm wherein the attenuation factor of the min-sum algorithm is determined by the coefficients of the attenuation factor matrix.

Abstract: A low-density parity check (LDPC) decoder is provided that eliminates the need to calculate customized check node codeword estimates by considering the check node processor and the variable node processor as a single processer having a shared memory for storing common variables to be used during both the check node processing and the variable node processing of the iterative decoding method.

Abstract: A low-density parity check (LDPC) decoder is provided for decoding low-density parity check (LDPC) encoded data wherein the processing order of the layers of the LDPC parity check matrix are rearranged during the decode process in an attempt to avoid error mechanisms brought about by the iterative nature of the LDPC belief propagation decoding process, such as stopping sets and trapping sets.

Abstract: A self-synchronous scrambler/descrambler and method for operating same are disclosed. A self-synchronous scrambler/descrambler comprises an M-bit Scrambler State memory for retaining M previously scrambled/descrambled bits, a SOP/EOP Zero Inserter for receiving replacing certain bytes of the bus word with a value of zero, a Mid-Packet Word Logic for scrambling/descrambling the received bits using the previously scrambled/descrambled bits from the M-bit Scrambler State memory; and a Barrel Shifter for rotating the M-bit Scrambler State memory backwards. The method for scrambling/descrambling bits, comprising receiving a bus word, replacing certain bytes of the bus word, scrambling/descrambling bits of the bus word by exclusive-ORing with previously scrambled/descrambled bits, retaining the scrambled/descrambled bits of the bus word; and rotating the scrambled/descrambled bits of the bus word backwards an amount.

Abstract: A method and system are provided for performing Decision Feedback Equalization (DFE) and Decision Feedback Sequence Estimation (DFSE) in high-throughput applications that are not latency critical. In an embodiment, overlapping blocks of samples are used to allow for the parallelization of the computation and the breaking of the critical path. In addition, the overlap of the windows addresses issues associated with performance loss due to what is termed “ramp-up” and “ramp-down” BER loss.

Abstract: Methods and systems to generate control signals for timing recovery of a signal received over baseband communications systems are disclosed. The timing control circuit uses a multi-rate DSP structure for the implementation of the DSP functions in the control loop for use in an ASIC and requires a reduced DSP clock rate, which in turn reduces the need for pipelining and/or high-speed libraries. Thus lower latency, better tracking performance and lower power consumption are achieved. An example embodiment involves splitting the timing error signal, supplied at a given update rate, into a sum and a difference component, and processing each component in separate circuit chains at half the update rate. The resultant half-rate control signals from each separate circuit chain are joined to provide a control signal at the full update rate. Thus, implementations of the present disclosure perform like a full-rate structure, but require a halved DSP clock rate.

Abstract: A method of background calibration of aperture center errors in a data communication system is provided. In an implementation, in response to detection of a low sampler output (“0”) in between two high sampler outputs (“1”), the method includes: determining a direction of an ADC output signal at the time of the detected low output; and adjusting timing at a selected sampler based on the determined signal direction. In an example implementation, the method includes watching for bubbles in the thermometer code output, and estimating the first derivative of the signal at the time of the bubble, then estimating the sign of the errors. In an example implementation, the errors are used in a control loop to reduce the aperture center error.

Abstract: A method and apparatus to digitally remove in-band non-linear signal distortion caused by a radio frequency (RF)/intermediate frequency (IF) receiver circuit that has non-linearities, which are further affected by low-IF ADC sample aliasing.

Abstract: A link negotiation method for enabling communication between first and second Serial Attached Small Computer Interface (SAS) storage devices operably coupled by an optical cable. The method includes continuously transmitting a non-SAS data pattern between the first and second SAS storage devices. In response to successful exchange of the non-SAS data between the first and second SAS storage devices, a SAS data pattern is continuously transmitted between the first and second SAS storage devices. In response to successful exchange of the SAS data pattern between the first and second SAS storage devices, an initial frame is continuously transmitted between the first and second SAS storage devices. Communication between the first and second SAS storage devices is enabled in response successful communication of the initial frame between the first and second SAS storage devices.