1. Field of the Invention
The present invention relates to a method and apparatus of clock recovery, and more particularly, to a method and apparatus using a serial to parallel converting unit to convert a serial phase error signal into a plurality of parallel phase error signals.
2. Description of the Prior Art
As is well known in the IC design industry, a clock recovery circuit is commonly used in optical communications. Please refer to
U.S. Pat. No. 6,442,225 further discloses another kind of clock recovery circuit. For more details, please refer to this patent.
One objective of the present invention is to provide a method and apparatus of clock recovery using a serial to parallel converting unit to convert a serial phase error signal into a plurality of parallel phase error signals, and generate a clock signal.
According to an embodiment of the present invention, a clock recovery circuit for generating an output clock signal with respect to an input signal is disclosed. The clock recovery circuit is for generating an output clock corresponding to an input signal. The clock recovery circuit includes: a phase detection unit for receiving the input signal and the output clock and generating a serial phase error signal according to the input signal and the output clock; a serial-to-parallel converting unit coupled to the phase detection unit for converting the serial phase error signal into a plurality of parallel phase error signals; a plurality of charging/discharging units coupled to the serial-to-parallel converting unit for generating an adjustment signal according to the parallel phase error signals; and an oscillator for generating the output clock according to the adjustment signal.
These and other objectives of the present invention will no doubt become obvious to those of ordinary skill in the art after reading the following detailed description of the preferred embodiment that is illustrated in the various figures and drawings.
Please refer to
Please refer to
Step 700: Generate a serial phase error signal Se according to a phase difference between the input signal Din and output clock signal CLKout;
Step 702: Convert the serial phase error signal Se into a plurality of parallel phase error signals P1˜Pn;
Step 704: Generate an adjustment signal Ie according to the plurality of parallel phase error signals P1˜Pn, wherein the plurality of parallel phase error signals P1˜Pn have frequencies lower than the frequency of the serial phase error signal Se.
Step 706: Filter the adjustment signal Ie to generate a control signal C; and
Step 708: Generate the output clock signal CLKout according to the control signal C.
In this embodiment, output signal Din can be a coded random signal with a range from 27-1 to 231-1, in other words, a non-return to zero (NRZ) signal, which is usually utilized in very high speed optical communications. The phase detection unit 310 takes the output clock CLKout as a sample clock for sampling the input signal Din, and converting the sampled input signal Din into a serial phase error signal Se (Step 700), wherein the serial phase error signal Se represents a phase difference between the input signal Din and the output clock signal CLKout. The serial-to-parallel converting unit 320 then converts the serial phase error signal Se into a plurality of parallel phase error signals P1˜Pn (Step 702). Please note that, in this embodiment, the number n of the plurality of parallel phase error signals depends on the operating speed a system can stand. For example, when designing the clock recovery circuit 300 to be operated at 2 Gbps, and if the charging/discharging unit 330 operates at 700˜800 Mhz, the serial phase error signal Se is converted into “four” parallel phase error signals P1˜P4 under considerations of manufacturing process and temperature variations. Since the frequencies of the parallel phase error signals P1˜P4 (corresponding to 700˜800 Mhz) are lower than the frequency of the serial phase error signal Se (corresponding to 2 Ghz), designing the charging/discharging unit 330 becomes significantly more simple.
After being processed by the serial-to-parallel converting unit 320, the serial phase error signal Se is converted to the plurality of parallel phase error signals P1˜Pn. Therefore the plurality of charging/discharging units 330 are required to separately receive the plurality of parallel phase error signals P1˜Pn in order to generate an adjustment signal Ie comprising a plurality of charging/discharging signals I1′˜In′ to be input to the loop filter 340 (Step 704). In this embodiment, the loop filter 340 is a kind of low-pass filter used for filtering the adjustment signal Ie to generate the control signal C (Step 706). The control signal C is further fed back to the controllable oscillator 350 (e.g. a voltage-control oscillator or a current-control oscillator), to drive the controllable oscillator 350 to output the needed output clock signal CLKout (Step 708). Since the implementations of the loop filter 340 such as the loop filter disclosed by U.S. Pat. No. 6,442,225 and the controllable oscillator 350 such as a VCO or an ICO are well known in the art, unnecessary details are not provided in this disclosure.
A clock recovery circuit in this embodiment applies a serial-to-parallel converting unit to lower the operating frequency of the plurality of charging/discharging units. Moreover, this circuit also enables a controllable oscillator to operate at a high frequency and generate a needed clock recovery signal. The present invention thus has the advantages of both low clock jitter, and simplicity of designing a charging/discharging unit.
Those skilled in the art will readily observe that numerous modifications and alterations of the device and method may be made while retaining the teachings of the invention. Accordingly, the above disclosure should be construed as limited only by the metes and bounds of the appended claims.
Number | Date | Country | Kind |
---|---|---|---|
094133201 | Sep 2005 | TW | national |