WO2016188055A1

WO2016188055A1 - Interpolation filtering method and apparatus

Info

Publication number: WO2016188055A1
Application number: PCT/CN2015/094775
Authority: WO
Inventors: 游少芳; 张集文
Original assignee: 深圳市中兴微电子技术有限公司
Priority date: 2015-05-22
Filing date: 2015-11-17
Publication date: 2016-12-01
Also published as: CN106301664A

Abstract

An interpolation filtering method is disclosed in the present invention, including: obtaining a phase deviation of received data; determining an interpolation coefficient list corresponding to the phase deviation of the data according to pre-stored relationship between phase deviations and interpolation coefficient lists in a storage area; and generating adjusting data corresponding to the data by performing a calculation of the data with the interpolation coefficient list as the tap coefficient of the data.

Description

Method and device for interpolation filtering

Technical field

The present invention relates to the field of communications, and in particular, to a method and apparatus for interpolation filtering.

Background technique

In the process of implementing the technical solutions of the embodiments of the present application, at least the following technical problems exist in the related technologies:

In the development of transmission technology, optical fiber has proved to be an indispensable medium. How to transmit the most abundant information with the least amount of optical fiber. For this exploration, the development of optical transmission has basically gone through the following stages: space division multiplexing (SDM), time division multiplexing (TDM) and wavelength division. Use phase (WDM).

Today, cable transmission is still dominated by the WDM system. With the development of communication technology, the commercial 40G wavelength division transmission has gradually evolved to 100G and 400G transmission. At the same time, the data transmission distance is also constantly expanding. In this way, the wavelength division system will bring about many problems such as chromatic dispersion, polarization film dispersion, and strong filtering effect in the transmission process. These problems need to be solved by digital signal processing, which is referred to as 100G DSP processing.

For the non-homologous clocks at the two ends of the general communication system, there is always a certain deviation in the clock frequencies at both ends. Moreover, the deviation keeps shaking with the problem of heat dissipation in the surrounding environment and its own board level. In this case, we need to interpolate between the samples based on the phase deviation detected in real time.

According to the above description, there are two main problems when doing the interpolation filter:

1. The interpolation coefficient requires real-time calculation;

2. The coefficient (or range of values) is uncertain when the jitter at both ends of the transceiver is uncertain.

3. When the system needs to be upgraded, the hardware needs to be redesigned.

Based on the above analysis, how to implement flexible configuration, good timing synchronization, simple and scalable interpolation filtering, and lack of corresponding solution mechanism in the prior art.

Summary of the invention

In view of this, the embodiment of the present invention is to provide a method and an apparatus for interpolation filtering, which solves at least the problems existing in the prior art, and can implement flexible interpolation, good timing synchronization, and simple and scalable interpolation filtering.

The technical solution of the embodiment of the present invention is implemented as follows:

A method for interpolation filtering according to an embodiment of the present invention, the method includes:

Obtaining the phase deviation of the received data;

Determining, according to the relationship between the phase deviation pre-stored in the storage area and the interpolation coefficient list, an interpolation parameter list corresponding to the phase deviation of the data;

The adjustment coefficient corresponding to the data is generated by using the interpolation coefficient list as a tap coefficient of the data and the data.

In the above solution, the method further includes:

Traverse the values of all phase deviations;

Calculating a list of interpolation coefficients for each phase deviation by calculating each phase deviation by each formula in the interpolation algorithm;

The interpolation coefficient list corresponding to each phase deviation is stored for each phase deviation to obtain a relationship between the phase deviation and the interpolation coefficient list.

In the above solution, the interpolation coefficient list corresponding to each phase deviation is stored corresponding to each phase deviation, and the relationship between the phase deviation and the interpolation coefficient list is included:

Each phase deviation is a storage address, and the interpolation coefficient list corresponding to each phase deviation is stored in a storage address indicated by each phase deviation.

In the above solution, determining the interpolation parameter list corresponding to the phase deviation of the data according to the relationship between the phase deviation pre-stored in the storage area and the interpolation coefficient list includes:

The interpolation coefficient list corresponding to the phase interpolation of the data is acquired from the address of the storage area with the phase deviation of the data as an address.

In the above solution, the method further includes:

The depth of the storage area is determined according to the bit width of the value of the phase deviation.

In the above solution, the method further includes:

The width of the storage area is determined based on the sum of the bit widths of the interpolation coefficients corresponding to all phase deviations.

An apparatus for interpolation filtering according to an embodiment of the present invention, where the apparatus includes: an obtaining module, a searching module, and a calculating module;

The acquiring module is configured to acquire a phase deviation of the received data;

The searching module is configured to determine, according to a relationship between a phase deviation pre-stored in the storage area and a list of interpolation coefficients, an interpolation parameter list corresponding to a phase deviation of the data;

The calculating module is configured to calculate, by using the interpolation coefficient list as a tap coefficient of the data, the adjustment data corresponding to the data to generate the data.

In the above solution, the device further includes: a configuration module configured to traverse the values of all phase deviations; and each phase deviation is respectively calculated by each formula in the interpolation algorithm to obtain a list of interpolation coefficients of each phase deviation; corresponding phases The deviation stores the interpolation coefficient list corresponding to each phase deviation to obtain a relationship between the phase deviation and the interpolation coefficient list.

In the above solution, the configuration module is specifically configured to traverse the values of all phase deviations; each phase deviation is respectively calculated by each formula in the interpolation algorithm to obtain a list of interpolation coefficients of each phase deviation; and each phase deviation is stored The address stores the interpolation coefficient list corresponding to each phase deviation in the storage address indicated by each phase deviation.

In the above solution, the searching module is configured to acquire, from the address of the storage area, an interpolation coefficient list corresponding to the phase interpolation of the data by using the phase deviation of the data as an address.

In the above solution, the apparatus further includes: a depth configuration module configured to determine a depth of the storage area according to a bit width of a value of the phase deviation.

In the above solution, the apparatus further includes: a width configuration module configured to determine a width of the storage area according to a sum of bit widths of interpolation coefficients corresponding to all phase deviations.

The acquisition module, the search module, the calculation module, the configuration module, the depth configuration module, and the width configuration module may use a central processing unit (CPU) and a digital number when performing processing. Implemented by a DSP (Digital Singnal Processor) or a Field-Programmable Gate Array (FPGA).

A method and apparatus for interpolation filtering according to an embodiment of the present invention acquires a phase deviation of received data; and determines a list of interpolation parameters corresponding to a phase deviation of the data according to a relationship between a phase deviation pre-stored in the storage region and a list of interpolation coefficients And calculating, by using the interpolation coefficient list as a tap coefficient of the data and the data to generate adjustment data corresponding to the data. The embodiments of the present invention solve at least the problems in the prior art, and can implement flexible configuration of interpolation coefficients and reduce delays by adding a small amount of overhead, thereby facilitating expansion of subsequent versions.

DRAWINGS

1 is a schematic flowchart of a method for interpolation filtering according to Embodiment 1 of the present invention;

2 is a real-time calculation formula of interpolation coefficients of a 100G wavelength division transmission system according to an embodiment of the present invention;

3 is a circuit diagram of a conventional interpolation coefficient calculation method;

4 is a schematic diagram of a block diagram of a clock recovery subsystem applying an interpolation filtering method according to Embodiment 2 of the present invention;

FIG. 5 is a schematic structural diagram of the interpolation filter shown in FIG. 4; FIG.

FIG. 6 is a schematic structural diagram of an apparatus for interpolation and filtering according to Embodiment 3 of the present invention;

FIG. 7 is a schematic structural diagram of another apparatus for interpolation filtering according to Embodiment 3 of the present invention.

detailed description

The implementation of the technical solution will be further described in detail below with reference to the accompanying drawings.

Embodiment 1

Embodiment 1 of the present invention provides a method for interpolation filtering, as shown in FIG. 1 , the method includes:

S101: Acquire a phase deviation of the received data.

S102: Determine, according to a relationship between a phase deviation pre-stored in the storage area and a list of interpolation coefficients, an interpolation parameter list corresponding to a phase deviation of the data;

S103: Perform, by using the interpolation coefficient list as a tap coefficient of the data, the adjustment data corresponding to the data to generate the data.

Presetting a storage area for presetting all the relationship between the phase deviation and the interpolation coefficient list in the system at the receiving end, wherein the depth of the storage area is determined according to the bit width of the value of the phase deviation, corresponding to all phase deviations The sum of the bit widths of the interpolation coefficients determines the width of the storage area, and after determining the depth and width of the storage area, the relationship of all the phase deviations to the interpolation coefficient list is stored in the storage area.

In the embodiment of the present invention, the method further includes: traversing the values of all the phase deviations; and calculating the interpolation coefficients of each phase deviation by calculating each phase deviation by each formula in the interpolation algorithm; The list of interpolation coefficients corresponding to the phase deviation is stored to obtain a relationship between the phase deviation and the interpolation coefficient list.

At the receiving end, there are various possible values for the phase deviation of the received data. Taking the 100G wavelength division transmission system as an example, the phase deviation of the received data is 8 bits, and any one of 000000000 to 11111111 is encoded. , that is, any one from 0 to 255; the interpolation algorithm of the 100G wavelength division transmission system includes five formulas, as shown in FIG. 2, including h ₀ (u), h ₁ (u), h ₂ (u), h ₃ (u), h ₄ (u), h ₅ (u), the formula of the first, second, third, fourth, and fifth interpolation coefficients when the phase deviation is u, and the u value is passed through The calculation of several formulas yields a list of interpolation coefficients corresponding to the u value. Iterate through the values of all phase deviations from 0 to 255, and obtain the interpolation coefficient list of each phase deviation by calculating each phase deviation by each formula in the interpolation algorithm. Specifically, 0 is passed through h ₀ (u), h ₁ (u), h ₂ (u), h ₃ (u), h ₄ (u), h ₅ (u) are calculated by obtaining a list of interpolation coefficients when u is 0, and passing 1 through h ₀ (u), h ₁ (u), h ₂ (u), h ₃ (u), h ₄ (u), h ₅ (u) are calculated by obtaining a list of interpolation coefficients when u is 1, until 255 is passed through h ₀ (u), h ₁ (u), h ₂ (u), h ₃ (u), h ₄ (u), h ₅ (u) are calculated as a list of interpolation coefficients when u is 255, and 0 to 255 and their corresponding interpolation coefficients The list is stored in the storage area to obtain a relationship between the phase deviation and the list of interpolation coefficients.

Here, the interpolation coefficient list corresponding to each phase deviation is stored for each phase deviation, and the relationship between the phase deviation and the interpolation coefficient list includes: storing each phase deviation as a storage address, and storing the interpolation coefficient list corresponding to each phase deviation in each phase. The storage address represented by the deviation. That is, in the storage area, the interpolation coefficient list corresponding to 0 is stored at address 00000000, the interpolation coefficient list corresponding to 1 is stored in 00000001, and the interpolation coefficient list corresponding to 2 is stored at address 00000010.

In S101, after receiving the data, the receiving end of the communication system obtains the phase deviation of the received data by: clock pre-filtering, phase-collecting, and loop filtering, wherein the clock pre-filtering is the code caused by the dispersion. Inter-band interference filtering, phase-checking completes the clock phase deviation check; loop filtering completes the clock phase deviation filtering, and obtains the phase deviation of the received interpolation filtering data. Here, the acquisition of the phase deviation of the data is prior art and will not be described in detail herein.

In S102, the phase deviation of the received data acquired in step S101 is matched with the relationship between the phase deviation stored in the storage area and the interpolation coefficient list, and the interpolation coefficient list of the phase deviation of the data is searched for. When the phase deviation of the data is 00000000, the list of interpolation coefficients corresponding to 0 is searched, and when the phase deviation of the data is 00001001, the list of interpolation coefficients corresponding to 9 is searched.

Here, the interpolation coefficient list corresponding to each phase deviation is stored for each phase deviation, and the relationship between the phase deviation and the interpolation coefficient list includes: storing each phase deviation as a storage address, and storing the interpolation coefficient list corresponding to each phase deviation in each phase. The storage address represented by the deviation. Taking the phase deviations of 00000000 and 00001001 as an example, when the phase deviation of the data is 00000000, the list of interpolation coefficients stored in the storage area at the address of 00000000 is searched, and when the phase deviation of the data is 00001001, the address in the storage area is located at 00001001. A list of interpolation coefficients stored at.

In S103, the interpolation coefficient list searched in S102 is used as the tap coefficient of the data, and the received data is adjusted by the tap coefficient to obtain the adjusted data for transmission, that is, the data of the receiving end is adjusted, so that the receiving end of the system It is consistent with the clock frequency of the data at the transmitting end to synchronize the clock source of the transceiver.

It should be noted that, taking the 100G wavelength division transmission system as an example, the interpolation coefficient calculation method in the conventional interpolation filtering method is as shown in FIG. 3. When the interpolation filter receives the phase deviation of the data, the phase deviation is passed through multiple multiplication methods. To calculate the formulas h ₀ (u), h ₁ (u), h ₂ (u), h ₃ (u), h ₄ (u), h ₅ (u), the calculation of h ₀ (u) needs to be u, the coefficient one obtained by multiplying the third-order multiplier and multiplying by 0.32376, the coefficient two obtained by multiplying by the second-order multiplier and multiplying by 0.04416, and the coefficient three obtained by multiplying the first-order multiplier by 0.07683, After the coefficient one, the coefficient two and the coefficient three-phase add three coefficients, and the saturation truncation process is performed, the interpolation coefficient corresponding to u through h ₀ (u) is obtained, so when it is necessary to obtain the list of interpolation coefficients corresponding to u The interpolation coefficients corresponding to h ₀ (u), h ₁ (u), h ₂ (u), h ₃ (u), h ₄ (u), and h ₅ (u) need to be calculated separately to obtain a list of interpolation coefficients. In the existing interpolation filtering method, the interpolation coefficient is calculated in real time, the system is complicated, and the interpolation is continuously shaken with the problem of the surrounding environment and the heat dissipation of the board level, and the interpolation is performed when the jitter of the transmitting and receiving ends is uncertain. The coefficient (or range of values) is uncertain; further, when the system needs to be upgraded, the hardware needs to be redesigned, which consumes a lot of human resources.

In the interpolation filtering method provided by the embodiment of the present invention, the interpolation coefficient is pre-configured in the system. In the system, all the possible results of the interpolation coefficient calculation formula of the interpolation algorithm are traversed and stored in the storage area, thereby realizing the look-up table inside the chip instead of the real-time calculation, that is, the speed is increased, the power consumption is reduced, and the power is greatly enhanced. Flexibility. Here, the storage area may be a random access memory (RAM).

4 is a schematic diagram of a block diagram of a clock recovery subsystem applying an interpolation filtering method according to Embodiment 4 of the present invention; the clock recovery subsystem occupies a very important position in 100G, as shown in FIG. 4, including four main parts, respectively : Clock pre-filtering, phase discrimination, loop filtering, and interpolation filters. Among them, the clock pre-filtering is to filter the inter-symbol interference caused by the dispersion; the phase-checking completes the check of the clock phase deviation; the loop filtering completes the filtering of the clock phase deviation; and the interpolation filter performs the interpolation based on the phase deviation. A specific implementation of the interpolation filtering method provided by the embodiment of the present invention can be implemented in an interpolation filter in the system.

The structure of the interpolation filter is as shown in FIG. 5, which can explain the calculation method of the interpolation filter provided by the embodiment of the present invention; when the interpolation deviation u of the data is obtained, u is input into the RAM (256x26), and the table is checked. The mode obtains the calculated interpolation coefficient of u through h ₀ (u), h ₁ (u), h ₂ (u), h ₃ (u), h ₄ (u), h ₅ (u), that is, obtains u corresponding Interpolation list coefficient,

Embodiment 2

In an actual application, the interpolation filter method provided by the embodiment of the present invention in a 100G optical transmission system is applicable.

The implementation of the method can utilize the mature RAM IP core of the Application Specific Integrated Circuit (ASIC), and traverse all possible results of the interpolation coefficient calculation formula of the interpolation coefficient algorithm into the storage area RAM, thereby An internal implementation look-up table replaces the existing real-time calculations, which increases speed and power consumption while greatly enhancing flexibility.

The specific description is as follows:

First, the bit width of the interpolation pointer (u in Figure 2) is determined according to the algorithm requirement of the interpolation coefficient. For example, 8 bit in 100G, this bit width determines the depth of RAM that needs to be customized;

Secondly, the bit width of the interpolation coefficient is determined according to an algorithm of the interpolation coefficient, and the sum of the bit widths of the tap coefficients of each tap coefficient (ie, the interpolation coefficient in the interpolation coefficient list) (if the symmetric coefficient is used, the bit width can be reduced by about half), The bit width determines the width of the RAM;

Finally, the phase deviation u value is traversed into the algorithm formula (as shown in FIG. 3), and the interpolation coefficient list (the head coefficient of the tap coefficients) corresponding to each u value is combined as the RAM storage content. Among them, the result of calculating all the u values will be configured as the storage contents of the entire RAM at the time of power-on. When the system works, the interpolation filter can find the corresponding interpolation coefficient list (tap coefficient) according to the u value as the address;

Of course, the depth and width of the RAM can be appropriately reserved to facilitate more difficult situations in the future, such as the transceiver clock jitter exceeds the algorithm expectations, the interpolation pointer needs to be widened, the tap needs to be increased, etc., which may increase the RAM. The depth and width, of course, can also be replaced in later versions such as 100G upgrade to 400G or even 1T.

By adding the above processing flow and device, and adding a small amount of overhead, the following effects are achieved:

1. The interpolation coefficient can be flexibly configured according to different working conditions;

2, system delay is reduced, for large data systems such as 100G, delay reduction can significantly improve system performance;

3, through the use of RAM can also make the system power consumption reduction, because the control of the RAM is more convenient than controlling a bunch of registers and combinations as shown in Figure 3, the clock is very convenient to turn off;

4. If the subsequent version wants to increase or decrease the RAM, it can be easily replaced without redesigning the circuit.

Embodiment 3

In an embodiment of the present invention, an apparatus for interpolating filtering is provided. As shown in FIG. 6, the apparatus includes: an obtaining module 301, and a searching module 302. And a calculation module 303; wherein the obtaining module 301 is configured to acquire a phase deviation of the received data;

The searching module 302 is configured to determine, according to the relationship between the phase deviation pre-stored in the storage area and the interpolation coefficient list, an interpolation parameter list corresponding to the phase deviation of the data;

The calculating module 303 is configured to calculate, by using the interpolation coefficient list as a tap coefficient of the data, the adjustment data corresponding to the data to generate the data.

As shown in FIG. 7, the apparatus further includes: a configuration module 304 configured to traverse the values of all phase deviations; and each phase deviation is respectively calculated by each formula in the interpolation algorithm to obtain a list of interpolation coefficients of each phase deviation; Each phase deviation stores a list of interpolation coefficients corresponding to each phase deviation to obtain a relationship between the phase deviation and the interpolation coefficient list.

The configuration module 304 is configured to traverse the values of all the phase deviations; each phase deviation is respectively calculated by each formula in the interpolation algorithm to obtain a list of interpolation coefficients of each phase deviation; each phase deviation is a storage address, and each phase is The list of interpolation coefficients corresponding to the deviation is stored in the storage address indicated by each phase deviation.

Here, the searching module 302 is specifically configured to acquire the interpolation coefficient list corresponding to the phase interpolation of the data from the address of the storage area by using the phase deviation of the data as an address.

The apparatus also includes a depth configuration module 305 configured to determine a depth of the storage region based on a bit width of a value of the phase deviation.

The apparatus also includes a width configuration module 306 configured to determine a width of the storage region based on a sum of bit widths of interpolation coefficients corresponding to all phase offsets.

The integrated modules described in the embodiments of the present invention may also be stored in a computer readable storage medium if they are implemented in the form of software functional modules and sold or used as separate products. Based on such understanding, the technical solution of the embodiments of the present invention may be embodied in the form of a software product in essence or in the form of a software product stored in a storage medium, including a plurality of instructions. Make a computer device (can be a personal computing The machine, server, or network device, etc.) performs all or part of the methods described in various embodiments of the present invention. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. . Thus, embodiments of the invention are not limited to any specific combination of hardware and software.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention.

Industrial applicability

Claims

A method of interpolation filtering, the method comprising:

Obtaining the phase deviation of the received data;

Determining, according to the relationship between the phase deviation pre-stored in the storage area and the interpolation coefficient list, an interpolation parameter list corresponding to the phase deviation of the data;

The adjustment coefficient corresponding to the data is generated by using the interpolation coefficient list as a tap coefficient of the data and the data.
The method of claim 1 wherein the method further comprises:

Traverse the values of all phase deviations;

Calculating a list of interpolation coefficients for each phase deviation by calculating each phase deviation by each formula in the interpolation algorithm;

The interpolation coefficient list corresponding to each phase deviation is stored for each phase deviation to obtain a relationship between the phase deviation and the interpolation coefficient list.
The method according to claim 2, wherein storing the interpolation coefficient list corresponding to each phase deviation corresponding to each phase deviation to obtain a relationship between the phase deviation and the interpolation coefficient list comprises:

Each phase deviation is a storage address, and the interpolation coefficient list corresponding to each phase deviation is stored in a storage address indicated by each phase deviation.
The method according to claim 3, wherein determining the interpolation parameter list corresponding to the phase deviation of the data according to the relationship between the phase deviation pre-stored in the storage region and the interpolation coefficient list comprises:

The interpolation coefficient list corresponding to the phase interpolation of the data is acquired from the address of the storage area with the phase deviation of the data as an address.
The method of claim 1 wherein the method further comprises:

The depth of the storage area is determined according to the bit width of the value of the phase deviation.
The method of claim 1 wherein the method further comprises:

The width of the storage area is determined based on the sum of the bit widths of the interpolation coefficients corresponding to all phase deviations.
An apparatus for interpolation filtering, the apparatus comprising: an acquisition module, a lookup module, and a calculation module; wherein

The acquiring module is configured to acquire a phase deviation of the received data;

The searching module is configured to determine, according to a relationship between a phase deviation pre-stored in the storage area and a list of interpolation coefficients, an interpolation parameter list corresponding to a phase deviation of the data;

The calculating module is configured to calculate, by using the interpolation coefficient list as a tap coefficient of the data, the adjustment data corresponding to the data to generate the data.
The apparatus according to claim 7, wherein said apparatus further comprises: a configuration module configured to traverse the values of all phase deviations; and each phase deviation is calculated by each formula in the interpolation algorithm to obtain each phase deviation A list of interpolation coefficients; storing the interpolation coefficient list corresponding to each phase deviation for each phase deviation to obtain a relationship between the phase deviation and the interpolation coefficient list.
The apparatus according to claim 8, wherein the configuration module is configured to traverse the values of all phase deviations; and obtain a permutation coefficient list of each phase deviation by calculating each phase deviation by each formula in the interpolation algorithm; Each phase deviation is a storage address, and the interpolation coefficient list corresponding to each phase deviation is stored in a storage address indicated by each phase deviation.
The apparatus of claim 9, wherein the lookup module is configured to acquire a list of interpolation coefficients corresponding to phase interpolation of the data from the address of the storage area with a phase deviation of the data as an address.
The apparatus of claim 7, wherein the apparatus further comprises: a depth configuration module configured to determine a depth of the storage area based on a bit width of a value of the phase deviation.
The apparatus of claim 7, wherein the apparatus further comprises: a width configuration module configured to determine a width of the storage area based on a sum of bit widths of interpolation coefficients corresponding to all phase deviations.