WO2011079430A1 - 一种时钟频率调整的系统及方法 - Google Patents

一种时钟频率调整的系统及方法 Download PDF

Info

Publication number
WO2011079430A1
WO2011079430A1 PCT/CN2009/076157 CN2009076157W WO2011079430A1 WO 2011079430 A1 WO2011079430 A1 WO 2011079430A1 CN 2009076157 W CN2009076157 W CN 2009076157W WO 2011079430 A1 WO2011079430 A1 WO 2011079430A1
Authority
WO
WIPO (PCT)
Prior art keywords
clock
local
adjustment
level
phase
Prior art date
Application number
PCT/CN2009/076157
Other languages
English (en)
French (fr)
Inventor
郑伟
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to PCT/CN2009/076157 priority Critical patent/WO2011079430A1/zh
Publication of WO2011079430A1 publication Critical patent/WO2011079430A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/04Generating or distributing clock signals or signals derived directly therefrom
    • G06F1/08Clock generators with changeable or programmable clock frequency
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/04Generating or distributing clock signals or signals derived directly therefrom
    • G06F1/12Synchronisation of different clock signals provided by a plurality of clock generators

Definitions

  • the present invention relates to frequency adjustment techniques, and more particularly to a system and method for achieving high precision fast clock frequency adjustment. Background technique
  • Clocks exist in almost all digital products. The requirements and application scenarios are different, and the accuracy and stability performance of the clock are completely different. In most applications, the local clock of the device does not need to be functioning externally, and the performance of the clock can meet the demand. Relatively speaking, the industrial measurement and communication field requires high clock performance indicators. In most cases, it is necessary to use an auxiliary means to calibrate the clock, for example, using a cesium atomic clock or satellite timing as a reference to adjust the clock frequency.
  • Slave clocks typically use a pulse per second (PP1S, Pulse Per one Second) signal recovered from an Ethernet message as a reference for clock frequency adjustment. Due to the inherent drawbacks of the technology, under the current technical conditions, the recovered PP1S signal will have hundreds of billions of seconds (ns, nanosecond) of jitter. At the same time, the biggest difference between the recovered PP1S signal and the PP1S signal sent by satellite timing is: The jitter distribution is completely different. The jitter distribution of the PP1S signal is close to the normal distribution, and the jitter distribution of the recovered PP1S signal has a great relationship with the synchronization algorithm and the application scenario. The jitter distribution is almost random.
  • P1S Pulse Per one Second
  • the clock frequency In order to ensure that the recovered PP1S signal is consistent with the PP1S signal transmitted by the satellite, the clock frequency must be corrected.
  • the current correction technology only provides a local clock for adjustment, which will bring about a problem of long lock-in time and low adjustment accuracy, and cannot accurately and quickly fine-tune the correlation established between local multi-clocks. Therefore, the use of the current Some correction techniques do not achieve good performance specifications, limiting the application of the above-mentioned clock reference technology in systems that are extremely sensitive to clock frequency jitter.
  • a clock frequency adjustment system includes: an adjustment unit, configured to: when a plurality of clocks enter a fine adjustment state from a coarse adjustment state, and a correlation between a non-reference clock and a reference clock, the non-reference clock is based on a non-reference clock Fine-tuning the non-reference clock by adjusting the difference between the reference clock and the adjustment range of the reference clock;
  • the multiple clocks include: the reference clock and the non-reference clock.
  • the adjusting unit further includes: a phase detecting module and a clock adjusting module; wherein, the phase detecting module is configured to: the reference clock is a local first level clock, and the non-reference clock is a local second level clock In the case of the coarse adjustment state, the phase discrimination value between the input reference and the local first-level clock, and between the input reference and the local second-level clock are respectively calculated according to an input reference;
  • a clock adjustment module configured to: when the reference clock is a local first-level clock, where the non-reference clock is a local second-level clock, in the coarse adjustment state, according to a local first output by the phase-detecting module Calculating the first adjustment value according to the phase-detection value of the clock, and calculating the second adjustment value according to the phase-detection value of the local second-level clock output by the phase-detecting module; respectively, according to the first adjustment value and the second adjustment value, respectively, to the local first level
  • the clock and the local second-level clock are coarsely adjusted.
  • the condition that the correlation between the non-reference clock and the reference clock is specifically: The phase-detection value of the first-level clock is less than the duration of the first threshold, and the phase-detection value of the local second-level clock is less than the duration of the first threshold, which is greater than the second threshold.
  • the clock adjustment module is further configured to calculate, according to an adjustment value of the local first-level clock, and an adjustment value of the local second-level clock, a compensation difference between the two clocks; according to the compensation difference and the local number
  • the adjustment range of the primary clock is performed on the local second-level clock.
  • a method for clock frequency adjustment comprising:
  • the plurality of clocks enter a fine adjustment state from the coarse adjustment state; wherein, the plurality of clocks include: a reference clock and a non-reference clock;
  • the non-reference clock When there is a correlation between the non-reference clock and the reference clock, the non-reference clock performs fine adjustment of the non-reference clock based on the compensation difference between the non-reference clock and the reference clock, and the adjustment range of the reference clock.
  • the reference clock is a local first-level clock
  • the non-reference clock is a local second-level clock.
  • the method further includes:
  • the phase discrimination module respectively calculates a phase discrimination value between the input reference and the local first-level clock, and between the input reference and the local second-level clock, according to an input reference;
  • the clock adjustment module calculates a first adjustment value according to the phase-detection value of the local first-level clock output by the phase-detection module, and calculates a second adjustment value according to the phase-detection value of the local second-level clock output by the phase-detection module; An adjustment value and a second adjustment value respectively perform coarse adjustment on the local first-level clock and the local second-level clock.
  • the coarse adjustment state is entered into the fine adjustment state.
  • the condition that the correlation between the non-reference clock and the reference clock is specific is: the phase-detection value of the local first-level clock is less than the duration of the first threshold, and the phase-detection value of the local second-level clock is less than The duration of the first threshold is greater than the second threshold.
  • the fine tuning of the non-reference clock includes: The clock adjustment module calculates a compensation difference between the two clocks according to the adjustment value of the local first-level clock and the adjustment value of the local second-level clock; according to the compensation difference and the adjustment range of the local first-level clock, The local second-level clock is cycled.
  • a plurality of clocks enter a fine adjustment state from a coarse adjustment state; when there is a correlation between a non-reference clock and a reference clock, the non-reference clock is based on a compensation difference between the non-reference clock and the reference clock, and an adjustment range of the reference clock, Perform fine-tuning of the non-reference clock.
  • the correlation between the local multi-clocks is utilized, and the clock frequency is accurately and quickly fine-tuned, thereby obtaining a good performance index, and the above-mentioned clock reference technology is abnormally sensitive to the clock frequency jitter. In the system, it can also be widely used.
  • FIG. 1 is a schematic structural diagram of a system embodiment of the present invention.
  • DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The basic idea of the present invention is: multiple clocks enter a fine adjustment state from a coarse adjustment state; when there is a correlation between a non-reference clock and a reference clock, the non-reference clock is based on a compensation difference between the non-reference clock and the reference clock. , and the adjustment range of the reference clock, fine-tuning the non-reference clock.
  • a clock frequency adjustment system includes: an adjustment unit, configured to enter a fine adjustment state from a coarse adjustment state of a plurality of clocks; a non-reference clock according to a non-reference clock and a reference when a correlation between a non-reference clock and a reference clock The compensation difference between the clocks, and the adjustment range of the reference clock, are fine-tuned for the non-reference clock.
  • multiple clocks include: reference clock and non-reference clock.
  • the reference clock is usually the local first-level clock
  • the non-reference clock is the other-level clock.
  • the adjustment unit further includes: a phase discrimination module and a clock adjustment module.
  • the phase detecting module is configured to be a local first-level clock in the reference clock
  • the non-reference clock is a local second-level clock
  • the phase discrimination value between the input reference and the local first-level clock, and the input reference and the local second-level clock are respectively calculated according to the input reference.
  • the clock adjustment module is configured to: when the reference clock is the local first-level clock, and the non-reference clock is the local second-level clock, in the coarse adjustment state, according to the phase-detection value of the local first-level clock output by the phase-detecting module Calculating a first adjustment value, and calculating a second adjustment value according to the phase detection value of the local second-level clock output by the phase-detecting module; respectively, according to the first adjustment value and the second adjustment value, respectively, respectively, the local first-level clock and the local second-level The clock is coarsely adjusted.
  • condition for the correlation between the non-reference clock and the reference clock is specifically: the phase-detection value of the local first-level clock is less than the duration of the first threshold, and the phase-detection value of the local second-level clock is less than the first threshold. The duration is greater than the second threshold.
  • the clock adjustment module is further configured to calculate a compensation difference between the two clocks according to the adjustment value of the local first-level clock and the adjustment value of the local second-level clock; according to the compensation difference value and the local first-level clock Adjust the range and cycle the local second-level clock.
  • a method of clock frequency adjustment comprising the following:
  • multiple clocks enter the fine-tuning state from the coarse adjustment state; among them, multiple clocks include: reference clock and non-reference clock.
  • the non-reference clock When there is correlation between the non-reference clock and the reference clock in multiple clocks, the non-reference clock performs fine adjustment of the non-reference clock according to the compensation difference between the non-reference clock and the reference clock, and the adjustment range of the reference clock. .
  • the reference clock is the local first-level clock
  • the non-reference clock is the local second-level clock
  • the method further comprises: the phase discrimination module respectively calculating the input reference and the local first-level clock, and the reference between the input reference and the local second-level clock according to the input reference
  • the phase adjustment module calculates a first adjustment value according to the phase-detection value of the local first-level clock output by the phase-detection module, according to the local second-level clock output by the phase-detection module
  • the phase adjustment value calculates a second adjustment value
  • the clock adjustment module performs coarse adjustment on the local first-level clock and the local second-level clock respectively according to the first adjustment value and the second adjustment value.
  • the fine-tuning state enters the fine-tuning state.
  • condition for the correlation between the non-reference clock and the reference clock is specifically: the phase-detection value of the local first-level clock is less than the duration of the first threshold, and the phase-of-phase detection of the local second-level clock The duration of the value less than the first threshold is greater than the second threshold.
  • the fine adjustment of the non-reference clock includes: the clock adjustment module calculates the compensation difference between the two clocks according to the adjustment value of the local first-level clock and the adjustment value of the local second-level clock; The adjustment range of the primary clock is performed on the local second-level clock.
  • the present invention is a solution for accurately and quickly adjusting the clock frequency, and is suitable for occasions requiring high clock frequency accuracy, which can effectively improve the performance index of the output clock and shorten the lock time of the clock.
  • the local first-level clock performs a coarse adjustment of the step size K1
  • the local second-level clock performs a coarse adjustment of the step size ⁇ 2
  • Wl, Kl, and ⁇ 2 are pre-settable values.
  • the filtering operation is an optional operation, and the present invention can also calculate the clock adjustment value directly from the phase-detection value without filtering, and will not be described again.
  • phase-detection values to be filtered are all smaller than the threshold ⁇ , then the local first-level clock and the second-level clock are locked, and the coarse adjustment is fine-tuned.
  • the adjustment step size is changed from Kl, ⁇ 2 to Ll, L2; Ll, L2 are pre-settable values.
  • the local first-level clock and the second-level clock lock hold time are both greater than the threshold N, and the root According to the adjustment value, the compensation difference P of the two clocks is calculated and saved, and is not changed.
  • the statistical time window length W1 of the local first-level clock is changed to a larger W2, and the statistical time window length W1 of the local second-level clock is changed to a smaller W3, and the adjustment value Y of the local second-level clock is limited to a certain value.
  • this range is calculated by the formula ⁇ [ ⁇ + ⁇ - ⁇ /2, ⁇ + ⁇ + ⁇ /2].
  • is the adjustment value of the local second-level clock
  • X is the adjustment value of the local first-level clock
  • is the compensation difference between the two clocks
  • Q is the value of the dynamic adjustment range.
  • the core of the invention is: using multiple clocks, such as the dual clock in the following embodiments, using the long-term statistics and tuning method to perform frequency correction on the first clock module, and acquiring the first The difference between the two clock modules and the first clock module in frequency alignment; when adjusting the second clock module, the short-time statistics are used, and the current adjustment value and the compensation difference of the first clock module are referenced. Fine-tuning in a small area.
  • the coarse adjustment of each clock can be performed; the correlation between multiple clocks can be established on the basis of the coarse adjustment, and the fine adjustment can be performed by means of the reference between multiple clocks, thereby effectively improving the clock Performance metrics, reduced lock-in time, and can be used in a variety of clock frequency recovery applications, with a strong versatility and scope of application.
  • the invention is illustrated by way of example below.
  • the clock frequency adjustment system includes: a phase detector, a clock generator 1, a clock generator 2, a statistical operator, a configuration controller, and a modifier.
  • the system includes the adjustment unit for realizing the clock frequency adjustment, and the logical division of the adjustment unit is various.
  • One specific implementation may be: Phase module and clock adjustment module.
  • the phase detector is a specific implementation of the phase discrimination module; a specific implementation of the clock adjustment module includes a clock generator 1, a clock generator 2, a statistical operator, a configuration controller, and a modifier.
  • the phase detector is configured to calculate the reference and the local first-level clock according to the input PPIS reference, And the phase discrimination value with the local second-level clock.
  • Clock Generator 1 is used to generate the frequency of the local first-level clock.
  • the clock generator 2 is used to generate the frequency of the local second stage clock.
  • the statistical operator is used to filter the phase discrimination value of the input, and obtain the reference deviation value by counting the phase identification value.
  • the configuration controller is used to adjust the input reference deviation value according to the configuration policy.
  • the corrector is used to correct the adjustment value Y based on the adjustment value X and the compensation difference value P.
  • the method for adjusting the clock frequency includes the following steps:
  • Step 301 Set the statistical time window of the statistical operator to 32 seconds.
  • Step 302 The phase detector performs phase identification on the input PP1S reference and the first-stage and second-level clock frequencies, and the statistical operator counts the phase-detection values, and periodically outputs the reference deviation value according to the set statistical time window length.
  • Step 303 The configuration controller determines whether the reference offset values of the first-level and second-level clocks are respectively less than a threshold of 0.2 Hz.
  • the first-level clock and the second-stage clock are performed in step 304.
  • the first-stage clock and the second-stage clock are all performed in step 306.
  • the reference deviation value of the first-stage clock and the second-stage clock is less than the threshold 0.2 Hz, and the other is not less than the threshold 0.2 Hz, the reference deviation value is not less than the threshold 0, and the 2 Hz clock proceeds to step 304, which is less than the threshold of 0.2 Hz.
  • the clock proceeds to step 306.
  • Step 304 Check the set value, and set the clock whose reference deviation value in the first and second stage clocks of the configuration controller is greater than or equal to the threshold of 0.2 Hz, and the adjustment step is 0.2 Hz.
  • Step 305 Configure the adjustment value to the clock generator 1 or the clock generator according to the reference deviation value and the step size, and return to step 302.
  • Step 306 Check the set value, and set the first and second level clocks in the configuration controller.
  • the test deviation value is less than the threshold 0.2Hz clock, and the adjustment step is 0.01Hz.
  • step 306 if the condition greater than the threshold of 64 seconds is satisfied, the first-level and second-level clocks are not only fine-tuned, but also have correlation between the two, and may be based on correlation, by the second
  • the stage clock refers to the first stage clock, and in the adjustment range of the first stage clock, the process proceeds to step 307: the configuration controller determines whether the duration of the first stage and the second stage clock reference deviation value are less than 0.2 Hz is greater than a threshold of 64 seconds. When it is judged that the threshold is less than 64 seconds, step 308 is performed. When it is judged that the threshold is greater than 64 seconds, that is, there is correlation between the two clocks of the week, and step 309 is performed.
  • Step 308 Configure the adjustment value to the clock generator 1 or the clock generator 2 according to the reference deviation value and the step size, and return to step 302.
  • Step 309 The configuration controller calculates the compensation difference between the two clocks according to the adjustment values configured to the clock generator 1 and the clock generator 2, and configures them to the corrector.
  • Step 310 The modifier adjusts the configuration of the controller output to the clock generator 2, and corrects according to the compensation difference and the adjustment range of the first-stage clock.
  • Step 311 The configuration controller configures the adjustment value to the clock generator 1 according to the reference deviation value and the step size, and the corrector configures the corrected adjustment value to the clock generator 2, and returns to step 302.

Description

一种时钟频率调整的系统及方法 技术领域
本发明涉及频率调整技术, 尤其涉及一种实现高精度快速时钟频率调 整的系统及方法。 背景技术
时钟在几乎所有数字产品中都存在, 需求和应用场景的不同, 对时钟 的精确度和稳定度性能指标也完全不同。 在大多数应用中, 设备本地的时 钟并不需要由外界作用就能正常工作, 时钟的性能指标也能满足需求。 相 对而言, 工业测量和通信领域对时钟性能指标要求较高, 在多数情况下还 需要采用辅助手段对时钟进行校准, 例如使用铷原子钟或者卫星授时作为 参考, 来调整时钟频率。
同步技术及以太网的迅速发展, 使得一种新的提供时钟参考的技术逐 渐普遍起来, 主要用于低成本解决卫星授时信号无法覆盖室内的问题。 该 技术的优势是可以让局域网内所有从属时钟都使用同一个时钟参考, 从而 使这些从属时钟完全保持一致。
从属时钟通常使用从以太网报文中恢复出来的每秒一次脉沖 ( PP1S, Pulse Per one Second )信号作为时钟频率调整的参考。 由于该技术本身固有 的缺陷, 在目前的技术条件下, 恢复出来的 PP1S信号会有几百个十亿分之 一秒(ns, nanosecond ) 的抖动。 同时, 恢复出来的 PP1S信号与卫星授时 发送的 PP1S信号最大的区别在于: 抖动分布完全不同。 卫星授时 PP1S信 号的抖动分布接近于正态分布, 而恢复 PP1S信号的抖动分布与同步算法、 应用场景有极大的关系, 抖动分布几乎接近于随机。 为了确保恢复出来的 PP1S信号与卫星授时发送的 PP1S信号一致, 就必须进行时钟频率的校正。 然而, 目前的校正技术只提供一个本地时钟进行调整, 会带来入锁时 间长和调整精度不高的问题, 无法利用本地多时钟之间建立的相关性进行 精确快速的微调, 因此, 使用现有的校正技术并不能获得很好的性能指标, 限制了上述提供时钟参考的技术在对时钟频率抖动异常敏感的系统中的应 用。 发明内容
有鉴于此, 本发明的主要目的在于提供一种时钟频率调整的系统及方 法, 实现了对时钟频率精确快速的微调。
为达到上述目的, 本发明的技术方案是这样实现的:
一种时钟频率调整的系统, 该系统包括: 调整单元, 用于多个时钟由 粗调状态进入微调状态, 且非参考时钟与参考时钟之间具备相关性时, 非 参考时钟根据非参考时钟与参考时钟之间的补偿差值, 和参考时钟的调整 范围, 进行非参考时钟的微调;
其中, 所述多个时钟包括: 所述参考时钟和所述非参考时钟。
其中, 所述调整单元, 进一步包括: 鉴相模块和时钟调整模块; 其中, 鉴相模块, 用于在所述参考时钟为本地第一级时钟, 所述非参考时钟 为本地第二级时钟的情况下, 在所述粗调状态时, 根据输入参考, 分别计 算所述输入参考与所述本地第一级时钟、 以及所述输入参考与所述本地第 二级时钟之间的鉴相值;
时钟调整模块, 用于在所述参考时钟为本地第一级时钟, 所述非参考 时钟为本地第二级时钟的情况下, 在所述粗调状态时, 根据鉴相模块输出 的本地第一级时钟的鉴相值计算第一调整值, 根据鉴相模块输出的本地第 二级时钟的鉴相值计算第二调整值; 根据第一调整值和第二调整值, 分别 对本地第一级时钟和本地第二级时钟进行粗调。
其中, 非参考时钟与参考时钟之间具备相关性的条件具体为: 所述本 地第一级时钟的鉴相值小于第一门限的持续时间, 和所述本地第二级时钟 的鉴相值小于第一门限的持续时间, 均大于第二门限。
其中, 所述时钟调整模块, 进一步用于根据本地第一级时钟的调整值, 和本地第二级时钟的调整值计算出两时钟之间的补偿差值; 根据所述补偿 差值和本地第一级时钟的调整范围, 对本地第二级时钟进行 周。
一种时钟频率调整的方法, 该方法包括:
多个时钟由粗调状态进入微调状态; 其中, 所述多个时钟包括: 参考 时钟和非参考时钟;
非参考时钟与参考时钟之间具备相关性时, 非参考时钟根据非参考时 钟与参考时钟之间的补偿差值, 和参考时钟的调整范围, 进行非参考时钟 的微调。
其中, 所述参考时钟为本地第一级时钟, 所述非参考时钟为本地第二 级时钟; 在所述粗调状态时, 该方法进一步包括:
鉴相模块根据输入参考, 分别计算所述输入参考与所述本地第一级时 钟、 以及所述输入参考与所述本地第二级时钟之间的鉴相值;
时钟调整模块根据鉴相模块输出的本地第一级时钟的鉴相值计算第一 调整值, 根据鉴相模块输出的本地第二级时钟的鉴相值计算第二调整值; 时钟调整模块根据第一调整值和第二调整值, 分别对本地第一级时钟和本 地第二级时钟进行粗调。
其中, 所述本地第一级时钟的鉴相值, 和所述本地第二级时钟的鉴相 值均小于第一门限时, 由所述粗调状态进入微调状态。
其中, 非参考时钟与参考时钟之间具备相关性的条件具体为: 所述本 地第一级时钟的鉴相值小于第一门限的持续时间, 和所述本地第二级时钟 的鉴相值小于第一门限的持续时间, 均大于第二门限。
其中, 所述进行非参考时钟的微调, 具体包括: 时钟调整模块根据本地第一级时钟的调整值, 和本地第二级时钟的调 整值计算出两时钟之间的补偿差值; 根据所述补偿差值和本地第一级时钟 的调整范围, 对本地第二级时钟进行 周。
本发明多个时钟由粗调状态进入微调状态; 非参考时钟与参考时钟之 间具备相关性时, 非参考时钟根据非参考时钟与参考时钟之间的补偿差值, 和参考时钟的调整范围, 进行非参考时钟的微调。
采用本发明, 利用在本地多时钟之间建立的相关性, 实现了对时钟频 率精确快速的微调, 从而获得了很好的性能指标, 使上述提供时钟参考的 技术在对时钟频率抖动异常敏感的系统中, 也能得到广泛的应用。 附图说明
图 1为本发明系统实施例的组成结构示意图。 具体实施方式 本发明的基本思想是: 多个时钟由粗调状态进入微调状态; 非参考时 钟与参考时钟之间具备相关性时, 非参考时钟根据非参考时钟与参考时钟 之间的补偿差值, 和参考时钟的调整范围, 进行非参考时钟的微调。
下面结合附图对技术方案的实施作进一步的详细描述。
一种时钟频率调整的系统, 该系统包括: 调整单元, 用于多个时钟由 粗调状态进入微调状态; 非参考时钟与参考时钟之间具备相关性时, 非参 考时钟根据非参考时钟与参考时钟之间的补偿差值, 和参考时钟的调整范 围, 进行非参考时钟的微调。
其中, 多个时钟包括: 参考时钟和非参考时钟。 参考时钟通常为本地 第一级时钟, 非参考时钟为其他各级时钟。
这里, 调整单元进一步包括: 鉴相模块和时钟调整模块。 其中, 鉴相 模块, 用于在参考时钟为本地第一级时钟, 非参考时钟为本地第二级时钟 的情况下, 在粗调状态时, 根据输入参考, 分别计算输入参考与本地第一 级时钟、 以及输入参考与本地第二级时钟之间的鉴相值。 时钟调整模块, 用于在参考时钟为本地第一级时钟, 非参考时钟为本地第二级时钟的情况 下, 在粗调状态时, 根据鉴相模块输出的本地第一级时钟的鉴相值计算第 一调整值, 根据鉴相模块输出的本地第二级时钟的鉴相值计算第二调整值; 根据第一调整值和第二调整值, 分别对本地第一级时钟和本地第二级时钟 进行粗调。
这里, 非参考时钟与参考时钟之间具备相关性的条件具体为: 本地第 一级时钟的鉴相值小于第一门限的持续时间, 和本地第二级时钟的鉴相值 小于第一门限的持续时间, 均大于第二门限。
这里, 时钟调整模块, 进一步用于根据本地第一级时钟的调整值, 和 本地第二级时钟的调整值计算出两时钟之间的补偿差值; 根据补偿差值和 本地第一级时钟的调整范围, 对本地第二级时钟进行 周。
一种时钟频率调整的方法, 该方法包括以下内容:
一: 多个时钟由粗调状态进入微调状态; 其中, 多个时钟包括: 参考 时钟和非参考时钟。
二: 多个时钟中, 非参考时钟与参考时钟之间具备相关性时, 非参考 时钟根据非参考时钟与参考时钟之间的补偿差值, 和参考时钟的调整范围, 进行非参考时钟的微调。
以下针对双时钟的情况进行阐述, 双时钟中, 参考时钟为本地第一级 时钟, 非参考时钟为本地第二级时钟。
就以上第一方面而言, 在粗调状态时, 该方法进一步包括: 鉴相模块 根据输入参考, 分别计算输入参考与本地第一级时钟、 以及输入参考与本 地第二级时钟之间的鉴相值; 时钟调整模块根据鉴相模块输出的本地第一 级时钟的鉴相值计算第一调整值, 根据鉴相模块输出的本地第二级时钟的 鉴相值计算第二调整值; 时钟调整模块根据第一调整值和第二调整值, 分 别对本地第一级时钟和本地第二级时钟进行粗调。
本地第一级时钟的鉴相值, 和本地第二级时钟的鉴相值均小于第一门 限时, 由粗调状态进入微调状态。
就以上第二方面而言, 非参考时钟与参考时钟之间具备相关性的条件 具体为: 本地第一级时钟的鉴相值小于第一门限的持续时间, 和本地第二 级时钟的鉴相值小于第一门限的持续时间, 均大于第二门限。
进行非参考时钟的微调, 具体包括: 时钟调整模块根据本地第一级时 钟的调整值, 和本地第二级时钟的调整值计算出两时钟之间的补偿差值; 根据补偿差值和本地第一级时钟的调整范围, 对本地第二级时钟进行 周。
综上所述, 本发明是一种实现时钟频率精确快速调整的方案, 适用于 对时钟频率精度要求较高的场合, 可以有效地提高输出时钟的性能指标, 缩短时钟的入锁时间。
本发明的技术方案主要包括以下内容:
一: 统计一段时间窗长 W1 内输入参考与本地第一级时钟, 以及与本 地第二级时钟的鉴相值, 对鉴相值分别进行滤波, 计算出本地第一级时钟 的调整值 X和本地第二级时钟的调整值 Y。
其中, 本地第一级时钟进行步长为 K1的粗调, 本地第二级时钟进行步 长为 Κ2的粗调, Wl、 Kl、 Κ2均为预先可设定值。 需要指出的是: 滤波操 作是可选操作, 本发明也可以不经过滤波, 直接由鉴相值计算得到时钟的 调整值, 不做赘述。
二: 待滤波后的鉴相值均小于门限 Μ, 则认为本地第一级时钟和第二 级时钟已入锁, 从粗调变为精调。
其中, 调整步长由 Kl、 Κ2改为 Ll、 L2; Ll、 L2为预先可设定值。 三: 待本地第一级时钟和第二级时钟入锁保持时间均大于门限 N, 根 据调整值计算出两时钟的补偿差值 P, 并保存, 不再变更。
其中, 本地第一级时钟的统计时间窗长 W1改为更大的 W2, 本地第二 级时钟的统计时间窗长 W1改为更小的 W3,本地第二级时钟的调整值 Y限 制在一定范围内, 这个范围由公式 ί^ [Χ + Ρ - β/2, χ + Ρ + β/2]计算获得。 该 公式中, Υ为本地第二级时钟的调整值; X为本地第一级时钟的调整值; Ρ 为两时钟之间的补偿差值, Q为动态调整范围的值。
本发明与现有的校正技术相比, 其核心在于: 使用多个时钟, 比如以 下实施例中的双时钟, 采用长时间统计及 调的方式对第一个时钟模块进 行频率校正, 同时获取第二个时钟模块与第一个时钟模块在频率调准时的 补偿差值; 在对第二个时钟模块进行调整时, 采用短时间统计以及参考第 一个时钟模块的当前调整值以及补偿差值, 在小范围内进行微调。 采用本 发明, 既可以进行各个时钟的粗调; 又可以在粗调的基础上, 建立多时钟 之间的相关性, 并借助于多时钟之间的参照进行精调, 从而可以有效地提 高时钟的性能指标, 减少入锁时间, 并可以用于各种时钟频率恢复应用场 合, 具有^艮强的通用性和适用范围。
以下对本发明进行举例阐述。
系统实施例:
如图 1 所示, 本实施例中, 时钟频率调整的系统包括: 鉴相器、 时钟 产生器 1、 时钟产生器 2、 统计运算器、 配置控制器和修正器。 这里需要指 出的是: 由以上系统原理的组成可知: 系统中, 包括用于实现时钟频率调 整的调整单元, 而调整单元的逻辑划分多种多样, 一种具体实现可以为: 调整单元具体包括鉴相模块和时钟调整模块。 本实施例中, 鉴相器即为鉴 相模块的一种具体实现; 时钟调整模块的一种具体实现包括时钟产生器 1、 时钟产生器 2、 统计运算器、 配置控制器和修正器。
其中,鉴相器用于根据输入 PPIS参考,计算该参考与本地第一级时钟、 以及与本地第二级时钟之间的鉴相值。 时钟产生器 1 用于生成本地第一级 时钟的频率。 时钟产生器 2用于生成本地第二级时钟的频率。 统计运算器 用于对输入的鉴相值进行滤波, 并通过对鉴相值的统计获得参考偏差值。 配置控制器用于将输入的参考偏差值按照配置策略进行调整。 修正器用于 根据调整值 X和补偿差值 P, 对调整值 Y进行修正。
方法实施例:
本实施例中, 时钟频率调整的方法流程包括以下步骤:
步骤 301 : 设定统计运算器的统计时间窗长为 32秒。
步骤 302: 鉴相器对输入 PP1S参考及第一级和第二级时钟频率进行鉴 相, 统计运算器对鉴相值进行统计, 并根据设定的统计时间窗长周期地输 出参考偏差值。
步骤 303:配置控制器判断第一级和第二级时钟的参考偏差值是否分别 小于门限 0.2Hz。
其中, 当第一级时钟和第二级时钟的参考偏差值均未小于门限 0.2Hz 时, 第一级时钟和第二级时钟都进行步骤 304。
当第一级时钟和第二级时钟的参考偏差值均小于门限 0.2Hz 时, 第一 级时钟和第二级时钟都进行步骤 306。
当第一级时钟和第二级时钟的参考偏差值中一个小于门限 0.2Hz,另一 个未小于门限 0.2Hz时,参考偏差值未小于门限 0,2Hz的时钟进行步骤 304, 小于门限 0.2Hz的时钟进行步骤 306。
步骤 304: 检查设定值, 并设定配置控制器的第一级和第二级时钟中参 考偏差值大于等于门限 0.2Hz的时钟, 调整步长为 0.2Hz。
步骤 305: 根据参考偏差值和步长, 配置调整值至时钟产生器 1或时钟 产生器, 返回步骤 302。
步骤 306: 检查设定值, 并设定配置控制器的第一级和第二级时钟中参 考偏差值小于门限 0.2Hz的时钟, 调整步长为 0.01Hz。
这里需要指出的是: 在步骤 306之前是第一级和第二级时钟各自进行 的粗调; 在步骤 306之后是第一级和第二级时钟进行的微调, 也可以称为 精调。 而且, 在步骤 307中, 如果满足大于门限 64秒的条件后, 第一级和 第二级时钟不仅进行微调, 而且二者之间具备相关性, 可以在具备相关性 的基础上, 由第二级时钟参照第一级时钟, 在第一级时钟的调整范围内进 步骤 307 : 配置控制器判断第一级和第二级时钟参考偏差值均小于 0.2Hz的持续时间是否大于门限 64秒, 当判断出小于门限 64秒时, 进行步 骤 308, 当判断出大于门限 64秒时, 即 周的两时钟之间具备相关性, 进 行步骤 309。
步骤 308: 根据参考偏差值和步长, 配置调整值至时钟产生器 1或时钟 产生器 2, 返回步骤 302。
步骤 309:配置控制器根据配置到时钟产生器 1和时钟产生器 2的调整 值, 计算出两时钟的补偿差值, 并配置到修正器。
步骤 310: 修正器对配置控制器输出的配置到时钟产生器 2的调整值, 根据补偿差值和第一级时钟的调整范围进行修正。
步骤 311 : 配置控制器根据参考偏差值和步长, 配置调整值至时钟产生 器 1 , 修正器将修正后的调整值配置至时钟产生器 2, 返回步骤 302。
以上所述, 仅为本发明的较佳实施例而已, 并非用于限定本发明的保 护范围。

Claims

权利要求书
1、 一种时钟频率调整的系统, 其特征在于, 该系统包括: 调整单元, 用于多个时钟由粗调状态进入微调状态, 且非参考时钟与参考时钟之间具 备相关性时, 非参考时钟根据非参考时钟与参考时钟之间的补偿差值, 和 参考时钟的调整范围, 进行非参考时钟的微调;
其中, 所述多个时钟包括: 所述参考时钟和所述非参考时钟。
2、 根据权利要求 1所述的系统, 其特征在于, 所述调整单元, 进一步 包括: 鉴相模块和时钟调整模块; 其中,
鉴相模块, 用于在所述参考时钟为本地第一级时钟, 所述非参考时钟 为本地第二级时钟的情况下, 在所述粗调状态时, 根据输入参考, 分别计 算所述输入参考与所述本地第一级时钟、 以及所述输入参考与所述本地第 二级时钟之间的鉴相值;
时钟调整模块, 用于在所述参考时钟为本地第一级时钟, 所述非参考 时钟为本地第二级时钟的情况下, 在所述粗调状态时, 根据鉴相模块输出 的本地第一级时钟的鉴相值计算第一调整值, 根据鉴相模块输出的本地第 二级时钟的鉴相值计算第二调整值; 根据第一调整值和第二调整值, 分别 对本地第一级时钟和本地第二级时钟进行粗调。
3、 根据权利要求 2所述的系统, 其特征在于, 非参考时钟与参考时钟 之间具备相关性的条件具体为: 所述本地第一级时钟的鉴相值小于第一门 限的持续时间, 和所述本地第二级时钟的鉴相值小于第一门限的持续时间, 均大于第二门限。
4、 根据权利要求 3所述的系统, 其特征在于, 所述时钟调整模块, 进 一步用于根据本地第一级时钟的调整值, 和本地第二级时钟的调整值计算 出两时钟之间的补偿差值; 根据所述补偿差值和本地第一级时钟的调整范 围, 对本地第二级时钟进行 周。
5、 一种时钟频率调整的方法, 其特征在于, 该方法包括: 多个时钟由粗调状态进入微调状态; 其中, 所述多个时钟包括: 参考 时钟和非参考时钟;
非参考时钟与参考时钟之间具备相关性时, 非参考时钟根据非参考时 钟与参考时钟之间的补偿差值, 和参考时钟的调整范围, 进行非参考时钟 的微调。
6、 根据权利要求 5所述的方法, 其特征在于, 所述参考时钟为本地第 一级时钟, 所述非参考时钟为本地第二级时钟; 在所述粗调状态时, 该方 法进一步包括:
鉴相模块根据输入参考, 分别计算所述输入参考与所述本地第一级时 钟、 以及所述输入参考与所述本地第二级时钟之间的鉴相值;
时钟调整模块根据鉴相模块输出的本地第一级时钟的鉴相值计算第一 调整值, 根据鉴相模块输出的本地第二级时钟的鉴相值计算第二调整值; 时钟调整模块根据第一调整值和第二调整值, 分别对本地第一级时钟和本 地第二级时钟进行粗调。
7、 根据权利要求 6所述的方法, 其特征在于, 所述本地第一级时钟的 鉴相值, 和所述本地第二级时钟的鉴相值均小于第一门限时, 由所述粗调 状态进入微调状态。
8、 根据权利要求 7所述的方法, 其特征在于, 非参考时钟与参考时钟 之间具备相关性的条件具体为: 所述本地第一级时钟的鉴相值小于第一门 限的持续时间, 和所述本地第二级时钟的鉴相值小于第一门限的持续时间, 均大于第二门限。
9、 根据权利要求 8所述的方法, 其特征在于, 所述进行非参考时钟的 微调, 具体包括:
时钟调整模块根据本地第一级时钟的调整值, 和本地第二级时钟的调 整值计算出两时钟之间的补偿差值; 根据所述补偿差值和本地第一级时钟 的调整范围, 对本地第二级时钟进行 周。
PCT/CN2009/076157 2009-12-28 2009-12-28 一种时钟频率调整的系统及方法 WO2011079430A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2009/076157 WO2011079430A1 (zh) 2009-12-28 2009-12-28 一种时钟频率调整的系统及方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2009/076157 WO2011079430A1 (zh) 2009-12-28 2009-12-28 一种时钟频率调整的系统及方法

Publications (1)

Publication Number Publication Date
WO2011079430A1 true WO2011079430A1 (zh) 2011-07-07

Family

ID=44226110

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/076157 WO2011079430A1 (zh) 2009-12-28 2009-12-28 一种时钟频率调整的系统及方法

Country Status (1)

Country Link
WO (1) WO2011079430A1 (zh)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101227246A (zh) * 2008-01-28 2008-07-23 中兴通讯股份有限公司 一种主从时钟同步的方法及装置
US20090168943A1 (en) * 2007-12-28 2009-07-02 Mediatek Inc. Clock generation devices and methods
CN101604182A (zh) * 2008-06-13 2009-12-16 原相科技股份有限公司 自动调整时钟频率的方法以及时钟频率调整电路

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090168943A1 (en) * 2007-12-28 2009-07-02 Mediatek Inc. Clock generation devices and methods
CN101227246A (zh) * 2008-01-28 2008-07-23 中兴通讯股份有限公司 一种主从时钟同步的方法及装置
CN101604182A (zh) * 2008-06-13 2009-12-16 原相科技股份有限公司 自动调整时钟频率的方法以及时钟频率调整电路

Similar Documents

Publication Publication Date Title
US10797686B1 (en) Phase predictor and associated method of use
US9763207B2 (en) Timing synchronization device and timing synchronization control method
EP2528255B1 (en) A method and a device for controlling frequency synchronization
WO2018137548A1 (zh) 一种时钟同步装置及方法
CN107147379A (zh) 基于fpga的边沿检测方法、系统及时钟数据恢复电路
US9537446B2 (en) System for producing a system clock and temperature gradient detection system
TWI357721B (en) Oscillation tuning circuit and method
JP2007033447A (ja) 周波数誤差を補正するスペクトルアナライザ及びその方法
WO2007062577A1 (fr) Boucle a phase asservie et methode d'amelioration de precision d'horloge
JP5153766B2 (ja) データ受信回路それを利用した試験装置
WO2014161360A1 (zh) 一种同步时钟信号补偿方法、装置及计算机存储介质
US20110026572A1 (en) Baud rate error detection circuit and baud rate error detection method
JP2013528011A (ja) 注入同期を補償する装置および方法
JP2009199139A (ja) パケット監視装置
CN104218946B (zh) 一种多参考无缝切换锁相装置
WO2017148240A1 (zh) 锁相环失锁的检测系统及检测方法
CN109188889B (zh) 一种原子钟1pps时间同步方法和系统
WO2011079430A1 (zh) 一种时钟频率调整的系统及方法
CN113541915B (zh) 一种宽动态范围的快速时钟恢复实现方法及装置
US20230077975A1 (en) Time synchronization between ieds of different substations
EP3160077A1 (en) Clock recovery apparatus and clock recovery method
US8140881B2 (en) Circuitry and method for detection of network node aging in communication networks
TWI400596B (zh) 同步接收電路及方法
US6680991B1 (en) Detection of frequency differences between signals
CN106304316B (zh) 一种频率同步性能检测方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09852715

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09852715

Country of ref document: EP

Kind code of ref document: A1