摘要:
A method, and circuitry, for choosing the correct equalization curve in adaptive equalization uses a feedback loop in which the incoming high-speed serial data are digitized and deserialized for use in the remainder of the device, and also are used by an adaptive state machine to both extract the reference levels for digitization and to control the equalization curve. Detection of the reference level and selection of the equalization curve may be performed at a different rates to avoid interfering with one another. The state machine preferably is programmable. This is useful in any device, but is particularly well-suited for a programmable device, such as a PLD or other programmable integrated circuit device, where conditions may vary according a user logic design.
摘要:
Equalization of an incoming data signal can be controlled by sampling that signal at times when data values in that signal should be stable (“data samples”) and when that signal should be in transition between successive data values that are different (“transition samples”). A transition sample that has been taken between two successive differently valued data samples is compared to a reference value (which can be one of those two data samples). The result of this comparison can be used as part of a determination as to whether to increase or decrease equalization of the incoming data signal.
摘要:
Equalization of an incoming data signal can be controlled by sampling that signal at times when data values in that signal should be stable (“data samples”) and when that signal should be in transition between successive data values that are different (“transition samples”). A transition sample that has been taken between two successive differently valued data samples is compared to a reference value (which can be one of those two data samples). The result of this comparison can be used as part of a determination as to whether to increase or decrease equalization of the incoming data signal.
摘要:
Equalization of an incoming data signal can be controlled by sampling that signal at times when data values in that signal should be stable (“data samples”) and when that signal should be in transition between successive data values that are different (“transition samples”). A transition sample that has been taken between two successive differently valued data samples is compared to a reference value (which can be one of those two data samples). The result of this comparison can be used as part of a determination as to whether to increase or decrease equalization of the incoming data signal.
摘要:
Equalization of an incoming data signal can be controlled by sampling that signal at times when data values in that signal should be stable (“data samples”) and when that signal should be in transition between successive data values that are different (“transition samples”). A transition sample that has been taken between two successive differently valued data samples is compared to a reference value (which can be one of those two data samples). The result of this comparison can be used as part of a determination as to whether to increase or decrease equalization of the incoming data signal.
摘要:
One embodiment relates to an integrated circuit which includes multiple communication channels, a clock multiplexer in each channel, two low-jitter clock generator circuits, and clock distribution circuitry. Each channel includes circuitry arranged to communicate a serial data stream using a reference clock signal, and the clock multiplexer in each channel is configured to select the reference clock signal from a plurality of input clock signals. The first low-jitter clock generator circuit is arranged to generate a first clock signal using a first inductor-capacitor-based oscillator circuit, and the second low-jitter clock generator circuit is arranged to generate a second clock signal using a second inductor-capacitor-based oscillator circuit The first and second inductor-capacitor-based oscillator circuits have different tuning ranges. The clock distribution circuitry is arranged to input the first and second low-jitter clock signals to each said clock multiplexer. Other embodiments and features are also disclosed.
摘要:
Signal detection circuitry for a serial interface oversamples the input—i.e., samples the input multiple times per clock cycle—so that the likelihood of missing a signal is reduced. Sampling may be done with a regenerative latch which has a large bandwidth and can latch a signal at high speed. The amplitude threshold for detection may be programmable, particularly in a programmable device. Thus, between the use of a regenerative latch which is likely to catch any signal that might be present, and the use of oversampling to avoid the problem of sampling at the wrong time, the likelihood of failing to detect a signal is greatly diminished. Logic, such as a state machine, may be used to determine whether the samples captured s do or do not represent a signal. That logic may be programmable, allowing a user to set various parameters for signal detection.
摘要:
High-speed serial data signal transmitter and/or receiver circuitry is able to dynamically switch between handling data at two (or more) different data rates. Such a switch can be made very rapidly and with no requirement for reprogramming or reconfiguring the circuitry. Circuitry for glitchlessly switching between clock signals having different frequencies is also provided and may be used in the above-mentioned transmitter and/or receiver circuitry.
摘要:
Signal detection circuitry for a serial interface oversamples the input—i.e., samples the input multiple times per clock cycle—so that the likelihood of missing a signal is reduced. Sampling may be done with a regenerative latch which has a large bandwidth and can latch a signal at high speed. The amplitude threshold for detection may be programmable, particularly in a programmable device. Thus, between the use of a regenerative latch which is likely to catch any signal that might be present, and the use of oversampling to avoid the problem of sampling at the wrong time, the likelihood of failing to detect a signal is greatly diminished. Logic, such as a state machine, may be used to determine whether the samples captured s do or do not represent a signal. That logic may be programmable, allowing a user to set various parameters for signal detection.