Abstract: A method of arranging and fabricating parallel primary and secondary coils of a wideband planar transformer is provided. The spacing and width of the coils are disposed to extend the bandwidth from DC to GHz and allow for high frequency coupling when the core permeability dramatically drops and achieves low reflected energy and low loss over a wide bandwidth. A bottom mold having a pattern of hole-pairs with conductive elements inserted vertically couples to a top mold such that a middle portion of the conductive elements spans between the top and bottom molds. Dielectric material envelopes the middle portion and a displacement feature of the mold creates a vacancy. A ferrite element is deposited to the vacancy. A second top mold spans the bottom mold and dielectric material is deposited to create a molded assembly. A deposited patterned conductive coating connects the element ends to define the transformer coils.

Abstract: An adaptive transversal filter having tap weights Wj which are products of corresponding tap coefficients Cj and tap gains Mj is provided. A filter control loop controls all of the tap coefficients Cj such that an error signal derived from the filter output is minimized. One or more tap control loops controls a tap gain Mk such that the corresponding tap coefficient Ck satisfies a predetermined control condition. For example, |Ck| can be maximized subject to a constraint |Ck|?Cmax, where Cmax is a predetermined maximum coefficient value. In this manner, the effect of quantization noise on the coefficients Cj can be reduced. Multiple tap control loops can be employed, one for each tap. Alternatively, a single tap control loop can be used to control multiple taps by time interleaving.

Abstract: An improved decoder and decoding method for low density parity check (LDPC) codes is provided. Decoding proceeds by repetitive message passing from a set of variable nodes to a set of check nodes, and from the check nodes back to the variable nodes. The variable node output messages include a “best guess” as to the relevant bit value, along with a weight giving the confidence in the guess (e.g., weak, medium or strong). The check node output messages have magnitudes selected from a predetermined set including neutral, weak, medium and strong magnitudes. The check node output messages tend to reinforce the status quo of the input variable nodes if the check node parity check is satisfied, and tend to flip bits in the input variable nodes if the check node parity check is not satisfied. The variable node message weights are used to determine the check node message magnitudes.

Abstract: A finite impulse response filter having tap weight rotation is provided, where each tap has a corresponding coefficient selector. Each coefficient selector includes N coefficients, where N is the number of taps. Each coefficient selector provides one of its corresponding coefficients as an input to a multiplier. Each multiplier also receives an input from a triggered track and hold tap. The tap coefficients are selected according to the time delay since the corresponding track and hold tap was most recently triggered. In this manner, the tendency of multiplier gain nonuniformity to degrade filter operation in the presence of tap weight rotation is reduced. In another embodiment, an offset selector is provided, to reduce the tendency of component offsets to degrade filter operation in the presence of tap weight rotation.

Abstract: High speed data transmission schemes often use differential lines to reduce the effect of noise on the data signal. Unfortunately, the signal propagation on the positive and negative lines may be different, which leads to a signal skew problem. This document describes a novel way of compensating for differential line skew in data transmission lines.