TWI816518B - Optimization system for composite equalizer and method thereof - Google Patents
Optimization system for composite equalizer and method thereof Download PDFInfo
- Publication number
- TWI816518B TWI816518B TW111131239A TW111131239A TWI816518B TW I816518 B TWI816518 B TW I816518B TW 111131239 A TW111131239 A TW 111131239A TW 111131239 A TW111131239 A TW 111131239A TW I816518 B TWI816518 B TW I816518B
- Authority
- TW
- Taiwan
- Prior art keywords
- equalizer
- eye
- mark
- inter
- height
- Prior art date
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 33
- 239000002131 composite material Substances 0.000 title claims abstract description 14
- 230000004044 response Effects 0.000 claims abstract description 52
- 238000004458 analytical method Methods 0.000 claims abstract description 21
- 230000000694 effects Effects 0.000 claims abstract description 17
- 238000010586 diagram Methods 0.000 claims description 53
- 238000012545 processing Methods 0.000 claims description 29
- 238000012216 screening Methods 0.000 claims description 19
- 238000005259 measurement Methods 0.000 claims description 11
- 239000003550 marker Substances 0.000 claims description 8
- 150000001875 compounds Chemical class 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 5
- 239000000463 material Substances 0.000 claims description 4
- 238000004806 packaging method and process Methods 0.000 claims description 4
- 230000008054 signal transmission Effects 0.000 abstract description 5
- 230000005540 biological transmission Effects 0.000 description 6
- 238000004590 computer program Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 239000000835 fiber Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002243 precursor Substances 0.000 description 2
- 230000001902 propagating effect Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Landscapes
- Filters And Equalizers (AREA)
- Tone Control, Compression And Expansion, Limiting Amplitude (AREA)
- Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
- Dc Digital Transmission (AREA)
Abstract
Description
本發明涉及一種優化系統及其方法,特別是複合等化器的優化系統及其方法。 The invention relates to an optimization system and a method thereof, in particular to an optimization system and a method for a composite equalizer.
近年來,隨著半導體技術的普及與蓬勃發展,晶片的尺寸越做越小,而運行速度則越來越快,但是高速的同時也會造成許多問題,例如:符碼間干擾(Inter-Symbol Interference,ISI)。 In recent years, with the popularization and vigorous development of semiconductor technology, the size of chips has become smaller and smaller, and the running speed has become faster and faster. However, high speed also causes many problems, such as: Inter-Symbol Interference (Inter-Symbol). Interference,ISI).
一般而言,傳統的數位通訊系統中,因通道為非理想通道,所以會造成符碼間干擾的現象,它會造成解調(Demodulation)的錯誤,進而使系統效能降低。一般通道的符碼間干擾,是一個碼的信號向後影響數個碼,但在有些數據通訊頻道,會造成很長的符碼間干擾的問題。 Generally speaking, in traditional digital communication systems, because the channel is a non-ideal channel, inter-symbol interference will occur, which will cause demodulation errors, thereby reducing system performance. Inter-symbol interference in general channels means that the signal of one code affects several codes backwards. However, in some data communication channels, it will cause long inter-symbol interference problems.
有鑑於此,便有廠商提出脈波形狀設計或等化技術。前者是藉由設計良好的碼信號的脈波形狀來減少它在接收端對後面碼的干擾;後者則是在接收端設計一個等化器(Equalizer)進行等化處理(Equalization),以補償通道的頻道效應,使其接近理想而消除符碼間干擾。然而,單純使用一個等化器所 達到效果不佳,特別是輸入輸出端(Input/Output,IO)進行高速傳輸時,仍然存在信號穩定性不佳之問題。 In view of this, some manufacturers have proposed pulse wave shape design or equalization technology. The former is to reduce the interference to subsequent codes at the receiving end by designing a good pulse shape of the code signal; the latter is to design an equalizer (Equalizer) at the receiving end for equalization processing (Equalization) to compensate for the channel The channel effect makes it close to ideal and eliminates inter-symbol interference. However, simply using an equalizer would The effect is not good, especially when the input/output (IO) performs high-speed transmission, there is still the problem of poor signal stability.
綜上所述,可知先前技術中長期以來一直存在信號高速傳輸的穩定性不佳之問題,因此實有必要提出改進的技術手段,來解決此一問題。 To sum up, it can be seen that the problem of poor stability of high-speed signal transmission has long existed in the prior art. Therefore, it is necessary to propose improved technical means to solve this problem.
本發明揭露一種複合等化器的優化系統及其方法。 The invention discloses an optimization system and method for a composite equalizer.
首先,本發明揭露一種複合等化器的優化系統,此系統包含:多個等化器、信號處理模組、執行模組、測量模組及篩選模組。其中,所述等化器包含連續時間線性等化器(Continuous Time Linear Equalizer,CTLE)、前饋等化器(Feed-Forward Equalizer,FFE)及決策回授等化器(Decision Feedback Equalizer,DFE);信號處理模組用以接收信號,並且通過濾波器對信號執行等化處理,以及對通道輸出的單脈衝響應進行濾波,並且將單脈衝響應的電壓最大值作為標記(Cursor)且計算前標記(Pre-cursor)和後標記(Post-cursor)的長度、數量及位置,其中,濾波器具有不斷改變的濾波器係數;執行模組連接信號處理模組,用以根據前標記及後標記的長度、數量及位置,分別計算前標記及後標記的符碼間干擾(Inter-Symbol Interference,ISI)以獲得符碼間干擾的總和的絕對值,並且執行峰值失真分析(Peak Distortion Analysis,PDA)將符碼間干擾與當前的標記相疊加以獲得單位間隔內的響應曲線,以及持續執行M次遍歷獲得對應的N個參數的眼圖,其中,M及N為正整數;測量模組連接執行模組及信號處理模組,用以測量眼圖的響應曲線以獲得眼高及眼寬,並且在執行等化處理後,基於眼高記錄峰值最高時所對應的濾波器係數、眼高及眼寬;以及 篩選模組連接所述等化器、執行模組及測量模組,用以在所述等化器中,依據連續時間線性等化器、前饋等化器及決策回授等化器的順序,由左至右進行多次篩選以組成多個等化器組合,其中,每一次篩選所組成的等化器組合至少包含二個等化器,並且對每一所述等化器組合執行峰值失真分析以獲得相應的眼高及眼寬,再根據記錄及獲得的眼高及眼寬選擇等化效果及穩定性皆為最佳的所述等化器組合。 First, the present invention discloses an optimization system for a composite equalizer. The system includes multiple equalizers, a signal processing module, an execution module, a measurement module and a screening module. Wherein, the equalizer includes a continuous time linear equalizer (CTLE), a feed-forward equalizer (Feed-Forward Equalizer, FFE) and a decision feedback equalizer (Decision Feedback Equalizer, DFE) ;The signal processing module is used to receive the signal, perform equalization processing on the signal through the filter, and filter the single pulse response of the channel output, and use the maximum voltage value of the single pulse response as a marker (Cursor) and calculate the pre-mark The length, quantity and position of (Pre-cursor) and post-cursor (Post-cursor), in which the filter has changing filter coefficients; the execution module connects the signal processing module to use the pre-cursor and post-cursor parameters according to the Length, quantity and position, calculate the inter-symbol interference (ISI) of the front mark and the rear mark respectively to obtain the absolute value of the sum of the inter-symbol interference, and perform peak distortion analysis (Peak Distortion Analysis, PDA) Superimpose the inter-symbol interference and the current mark to obtain the response curve within the unit interval, and continuously execute M times of traversal to obtain the corresponding eye diagram of N parameters, where M and N are positive integers; the measurement module connection is executed Module and signal processing module are used to measure the response curve of the eye diagram to obtain the eye height and eye width, and after performing equalization processing, record the filter coefficient, eye height and eye width corresponding to the highest peak value based on the eye height. wide; and The screening module is connected to the equalizer, the execution module and the measurement module, and is used in the equalizer according to the sequence of the continuous time linear equalizer, the feedforward equalizer and the decision feedback equalizer. , perform multiple screenings from left to right to form multiple equalizer combinations, where the equalizer combination formed by each screening contains at least two equalizers, and peak performance is performed on each of the equalizer combinations. Distortion analysis is performed to obtain the corresponding eye height and eye width, and then the equalizer combination with the best equalization effect and stability is selected based on the recorded and obtained eye height and eye width.
另外,本發明還揭露一種複合等化器的優化方法,其步驟包括:提供多個等化器,所述等化器包含連續時間線性等化器、前饋等化器及決策回授等化器;接收信號,並且通過濾波器對信號執行等化處理,以及對通道輸出的單脈衝響應進行濾波,並且將單脈衝響應的電壓最大值作為標記且計算前標記和後標記的長度、數量及位置,其中,濾波器具有不斷改變的濾波器係數;根據前標記及後標記的長度、數量及位置,分別計算前標記及後標記的符碼間干擾以獲得符碼間干擾的總和的絕對值,並且執行峰值失真分析將符碼間干擾與當前的標記相疊加以獲得單位間隔內的響應曲線,以及持續執行M次遍歷獲得對應的N個參數的眼圖,其中,M及N為正整數;測量眼圖的響應曲線以獲得眼高及眼寬,並且在執行等化處理後,基於眼高記錄峰值最高時所對應的濾波器係數、眼高及眼寬;以及在所述等化器中,依據連續時間線性等化器、前饋等化器及決策回授等化器的順序,由左至右進行多次篩選以組成多個等化器組合,其中,每一次篩選所組成的等化器組合至少包含二個等化器,並且對每一所述等化器組合執行峰值失真分析以獲得相應的眼高及眼寬,再根據記錄及獲得的眼高及眼寬選擇等化效果及穩定性皆為最佳的所述等化器組合。 In addition, the present invention also discloses a method for optimizing a compound equalizer, the steps of which include: providing a plurality of equalizers, the equalizers including a continuous-time linear equalizer, a feedforward equalizer and a decision feedback equalizer. receiver; receives the signal, performs equalization processing on the signal through the filter, and filters the single pulse response of the channel output, and uses the maximum voltage value of the single pulse response as a mark and calculates the length, number and position, where the filter has a constantly changing filter coefficient; according to the length, number and position of the front mark and the rear mark, the inter-symbol interference of the front mark and the rear mark is calculated respectively to obtain the absolute value of the sum of the inter-symbol interference , and perform peak distortion analysis to superimpose the inter-symbol interference with the current marker to obtain the response curve within the unit interval, and continuously perform M times of traversal to obtain the corresponding eye diagram of N parameters, where M and N are positive integers. ; Measure the response curve of the eye diagram to obtain the eye height and eye width, and after performing the equalization process, record the filter coefficient, eye height and eye width corresponding to the highest peak value based on the eye height; and in the equalizer , according to the order of continuous-time linear equalizer, feed-forward equalizer and decision-feedback equalizer, multiple screenings are performed from left to right to form multiple equalizer combinations, where each screening consists of The equalizer combination includes at least two equalizers, and performs peak distortion analysis on each equalizer combination to obtain the corresponding eye height and eye width, and then selects equalization based on the recorded and obtained eye height and eye width. The performance and stability of the above equalizer combination are the best.
本發明所揭露之系統與方法如上,與先前技術的差異在於本發明是透過提供多個不同構造的等化器,並且根據獲取的信號受到符碼間干擾影響的程度作為等化器優化的依據,其中,根據通道的單脈衝響應計算信號的符碼間干擾,並且執行峰值失真分析將ISI與當前的標記相疊加以獲得響應曲線進而獲得相應的眼圖,並且根據眼高及眼寬選擇等化效果及穩定性皆為最佳的等化器組合。 The system and method disclosed by the present invention are as above. The difference from the prior art is that the present invention provides a plurality of equalizers with different structures, and uses the degree of the acquired signal affected by inter-symbol interference as the basis for equalizer optimization. , among which, the inter-symbol interference of the signal is calculated based on the single pulse response of the channel, and peak distortion analysis is performed to superimpose the ISI and the current marker to obtain the response curve and then the corresponding eye diagram, and according to the eye height and eye width selection, etc. It is the best equalizer combination for both equalization effect and stability.
透過上述的技術手段,本發明可以達成提升信號高速傳輸的信號完整性及穩定性之技術功效。 Through the above technical means, the present invention can achieve the technical effect of improving signal integrity and stability of high-speed signal transmission.
100:等化器 100: Equalizer
100a:連續時間線性等化器 100a: Continuous-time linear equalizer
100b:前饋等化器 100b: Feedforward equalizer
100c:決策回授等化器 100c: Decision Feedback Equalizer
110:信號處理模組 110:Signal processing module
120:執行模組 120:Execute module
130:測量模組 130:Measurement module
140:篩選模組 140: Screening module
300:單脈衝響應 300:Single pulse response
310:標記 310:mark
320:前標記 320: front mark
330:後標記 330:Post mark
400,510,520,530,610,620,630:眼圖 400,510,520,530,610,620,630: Eye diagram
步驟210:提供多個等化器,所述等化器包含一連續時間線性等化器、一前饋等化器及一決策回授等化器 Step 210: Provide multiple equalizers, including a continuous-time linear equalizer, a feedforward equalizer, and a decision feedback equalizer.
步驟220:接收一信號,並且通過一濾波器對該信號執行等化處理,以及對一通道輸出的一單脈衝響應進行濾波,並且將該單脈衝響應的電壓最大值作為一標記且計算一前標記和一後標記的長度、數量及位置,其中,該濾波器具有不斷改變的一濾波器係數 Step 220: Receive a signal, perform equalization processing on the signal through a filter, filter a single pulse response output by a channel, and use the maximum voltage value of the single pulse response as a mark and calculate a previous The length, number and position of markers and a subsequent marker, where the filter has changing filter coefficients
步驟230:根據該前標記及該後標記的長度、數量及位置,分別計算該前標記及該後標記的一符碼間干擾以獲得該符碼間干擾的總和的絕對值,並且執行一峰值失真分析將該符碼間干擾與當前的該標記相疊加以獲得單位間隔內的一響應曲線,以及持續執行M次遍歷獲得對應的N個參數的一眼圖,其中,M及N為正整數 Step 230: Calculate the inter-symbol interference of the previous mark and the rear mark respectively according to the length, quantity and position of the previous mark and the rear mark to obtain the absolute value of the sum of the inter-symbol interference, and perform a peak value Distortion analysis superimposes the inter-symbol interference and the current marker to obtain a response curve within the unit interval, and continuously performs M traversals to obtain a corresponding eye view of N parameters, where M and N are positive integers.
步驟240:測量該眼圖的該響應曲線以獲得一眼高及一眼寬,並且在執行等化處理後,基於該眼高記錄峰值最高時所對應的該濾波器係數、該眼高及該眼寬 Step 240: Measure the response curve of the eye diagram to obtain one eye height and one eye width, and after performing equalization processing, record the filter coefficient, the eye height, and the eye width corresponding to the highest peak value based on the eye height.
步驟250:在所述等化器中,依序篩選該連續時間線性等化器、該前饋等化器及該決策回授等化器以組成多個等化器組合,並且對每一所述等化器組合執行該峰值失真分析以獲得相應的該眼高及該眼寬,再根據記錄及獲得的該眼高及該眼寬選擇等化效果及穩定性皆為最佳的所述等化器組合 Step 250: In the equalizer, sequentially screen the continuous-time linear equalizer, the feed-forward equalizer and the decision-feedback equalizer to form multiple equalizer combinations, and for each The equalizer combination performs the peak distortion analysis to obtain the corresponding eye height and eye width, and then selects the equalizer with the best equalization effect and stability based on the recorded and obtained eye height and eye width. Chemical combination
步驟251:根據該眼圖獲得的該眼寬及該眼高計算一判斷因子以作為選擇所述等化器組合的依據,該判斷因子的計算式為:α=λ(Wi/Wmax)+(1-λ)(Hi/Hmax),其中,α為該判斷因子、λ為一加權係數、Wi為該眼寬、Hi為該眼高、(Wi/Wmax)為一眼寬係數、(Hi/Hmax)為一眼高係數 Step 251: Calculate a judgment factor based on the eye width and eye height obtained from the eye diagram as a basis for selecting the equalizer combination. The calculation formula of the judgment factor is: α=λ(W i /W max ) +(1-λ)(H i /H max ), where α is the judgment factor, λ is a weighting coefficient, Wi is the eye width, H i is the eye height, (W i /W max ) is The width coefficient of one eye, (H i /H max ) is the height coefficient of one eye
步驟252:根據該判斷因子及至少一其它因素選擇所述等化器組合,所述其它因素包含晶片封裝、通信線路相關長度、材料及連接方式 Step 252: Select the equalizer combination according to the judgment factor and at least one other factor, including chip packaging, communication line related length, material and connection method.
第1圖為本發明複合等化器的優化系統的系統方塊圖。 Figure 1 is a system block diagram of the optimization system of the composite equalizer of the present invention.
第2A圖至第2C圖為本發明複合等化器的優化方法的方法流程圖。 Figures 2A to 2C are method flow charts of the optimization method of the composite equalizer of the present invention.
第3圖為單脈衝響應及符碼間干擾之示意圖。 Figure 3 is a schematic diagram of single pulse response and inter-symbol interference.
第4圖為本發明響應曲線疊加前後之示意圖。 Figure 4 is a schematic diagram before and after superposition of the response curves of the present invention.
第5圖為應用本發明篩選不同等化器組合的眼圖對比之示意圖。 Figure 5 is a schematic diagram of the eye diagram comparison of screening different equalizer combinations using the present invention.
第6圖為應用本發明在各種等化器優化下的眼圖對比之示意圖。 Figure 6 is a schematic diagram of eye diagram comparison under various equalizer optimizations using the present invention.
以下將配合圖式及實施例來詳細說明本發明之實施方式,藉此對本發明如何應用技術手段來解決技術問題並達成技術功效的實現過程能充分理解並據以實施。 The embodiments of the present invention will be described in detail below with reference to the drawings and examples, so that the implementation process of how to apply technical means to solve technical problems and achieve technical effects of the present invention can be fully understood and implemented accordingly.
請先參閱「第1圖」,「第1圖」為本發明複合等化器的優化系統的系統方塊圖,此系統包含:多個等化器100、信號處理模組110、執行模組120、測量模組130及篩選模組140。其中,所述等化器100包含連續時間線性等化器100a、前饋等化器100b及決策回授等化器100c。在實際實施上,每一次的等化器參數選擇過程中都需要與其餘等化器組合優化,以便在篩選時保證系統整體的優化性能。舉例來說,等化器的篩選順序為連續時間線性等化器100a、前饋等化器100b及決策回授等化器100c,對每一個組合執行峰值失真分析以獲得每個一組合的眼高及眼寬。 Please refer to "Figure 1" first. "Figure 1" is a system block diagram of the optimization system of the composite equalizer of the present invention. This system includes: multiple equalizers 100, a signal processing module 110, and an execution module 120. , measurement module 130 and screening module 140. The equalizer 100 includes a continuous-time linear equalizer 100a, a feedforward equalizer 100b, and a decision feedback equalizer 100c. In actual implementation, each equalizer parameter selection process needs to be optimized in combination with other equalizers to ensure the overall optimization performance of the system during screening. For example, the filtering order of equalizers is continuous-time linear equalizer 100a, feedforward equalizer 100b and decision feedback equalizer 100c, and peak distortion analysis is performed on each combination to obtain the eye level of each combination. Height and eye width.
信號處理模組110用以接收信號,並且通過濾波器對信號執行等化處理,以及對通道輸出的單脈衝響應進行濾波,並且將單脈衝響應的電壓最大值作為標記且計算前標記和後標記的長度、數量及位置,其中,濾波器具有不斷改變的濾波器係數。在實際實施上,可先對一個位元的單脈衝響應的信號做預處理以減少資料運算量。 The signal processing module 110 is used to receive the signal, perform equalization processing on the signal through the filter, and filter the single pulse response of the channel output, and use the maximum voltage value of the single pulse response as a mark and calculate the pre-mark and the post-mark. The length, number and position of the filter, where the filter has changing filter coefficients. In actual implementation, the signal of a single pulse response of one bit can be preprocessed first to reduce the amount of data calculation.
執行模組120連接信號處理模組110,用以根據前標記及後標記的長度、數量及位置,分別計算前標記及後標記的符碼間干擾以獲得符碼間干擾的總和的絕對值,並且執行峰值失真分析將符碼間干擾與當前的標記相疊加以獲得單位間隔內的響應曲線,以及持續執行M次遍歷獲得對應的N個參數的眼圖,其中,M及N為正整數。在實際實施上,當符碼間干擾的疊加使眼圖閉合至一個門檻值(例如:預先設置的眼高或眼寬的長度)時,比較響應曲線與邏輯零電平以獲得二個相交點作為眼圖的參數,所述二個相交點為左右各一的相交點。 The execution module 120 is connected to the signal processing module 110 to calculate the inter-symbol interference of the pre-mark and the post-mark respectively according to the length, quantity and position of the pre-mark and the post-mark to obtain the absolute value of the sum of the inter-symbol interference, And perform peak distortion analysis to superimpose the inter-symbol interference and the current mark to obtain the response curve within the unit interval, and continuously perform M times of traversal to obtain the corresponding eye diagram of N parameters, where M and N are positive integers. In actual implementation, when the superposition of inter-symbol interference causes the eye diagram to close to a threshold value (for example: a preset length of eye height or eye width), the response curve is compared with the logic zero level to obtain two intersection points As a parameter of the eye diagram, the two intersection points are one on the left and one on the left.
測量模組130連接執行模組120及信號處理模組110,用以測量眼圖的響應曲線以獲得眼高及眼寬,並且在執行等化處理後,基於眼高記錄峰值最高時所對應的濾波器係數、眼高及眼寬。在實際實施上,峰值最高時也代表電壓最大的時候。 The measurement module 130 is connected to the execution module 120 and the signal processing module 110 to measure the response curve of the eye diagram to obtain the eye height and eye width, and after performing the equalization process, record the highest peak value based on the eye height. Filter coefficients, eye height and eye width. In actual implementation, the highest peak value also represents the highest voltage value.
篩選模組140連接等化器100、執行模組120及測量模組130,用以在所述等化器100中,依序篩選連續時間線性等化器100a、前饋等化器100b及決策回授等化器100c以組成多個等化器組合,並且對每一所述等化器組合執行峰值失真分析以獲得相應的眼高及眼寬,再根據記錄及獲得的眼高及眼寬選擇等化效果及穩定性皆為最佳的所述等化器組合。在實際實施上,篩選模組140更可包含根據眼圖獲得的眼寬及眼高計算判斷因子以作為選擇所述等化器組合的依據,其判斷因子的計算式為:「α=λ(Wi/Wmax)+(1-λ)(Hi/Hmax)」,其中,「α」為判斷因子、「λ」為加權係數、「Wi」為眼寬、「Hi」為眼高、「(Wi/Wmax)」為眼寬係數、「(Hi/Hmax)」為眼高係數。另外,篩選模組140更可包含根據判斷因子及其它因素選擇所述等化器組合,所述其它因素包含晶片封裝、通信線路(Communication Line)相關長度、材料及連接方式。特別要說明的是,當判斷因子為數值0時,以眼高的最佳值為優化目標,當判斷因子為數值1時,以眼寬的最佳值為優化目標。 The screening module 140 is connected to the equalizer 100, the execution module 120 and the measurement module 130, and is used to sequentially screen the continuous-time linear equalizer 100a, the feedforward equalizer 100b and the decision-making in the equalizer 100. The equalizer 100c is fed back to form a plurality of equalizer combinations, and a peak distortion analysis is performed on each equalizer combination to obtain the corresponding eye height and eye width, and then the eye height and eye width are recorded and obtained. Select the above-mentioned equalizer combination with the best equalization effect and stability. In actual implementation, the screening module 140 may further include calculating a judgment factor based on the eye width and eye height obtained from the eye diagram as a basis for selecting the equalizer combination. The calculation formula of the judgment factor is: "α=λ( W i /W max )+(1-λ)(H i /H max )", where "α" is the judgment factor, "λ" is the weighting coefficient, "W i " is the eye width, and "H i " is Eye height, "(W i /W max )" is the eye width coefficient, and "( Hi /H max )" is the eye height coefficient. In addition, the screening module 140 may further include selecting the equalizer combination based on judgment factors and other factors, including chip packaging, communication line related length, materials and connection methods. In particular, when the judgment factor is 0, the best value of eye height is the optimization target; when the judgment factor is 1, the best value of eye width is the optimization target.
特別要說明的是,在實際實施上,本發明所述的模組皆可利用各種方式來實現,包含軟體、硬體或其任意組合,例如,在某些實施方式中,各模組可利用軟體及硬體或其中之一來實現,除此之外,本發明亦可部分地或完全地基於硬體來實現,例如,系統中的一個或多個模組可以透過積體電路晶片、系統單晶片、複雜可程式邏輯裝置(Complex Programmable Logic Device, CPLD)、現場可程式邏輯閘陣列(Field Programmable Gate Array,FPGA)等來實現。本發明可以是系統、方法及/或電腦程式。電腦程式可以包括電腦可讀儲存媒體,其上載有用於使處理器實現本發明的各個方面的電腦可讀程式指令,電腦可讀儲存媒體可以是可以保持和儲存由指令執行設備使用的指令的有形設備。電腦可讀儲存媒體可以是但不限於電儲存設備、磁儲存設備、光儲存設備、電磁儲存設備、半導體儲存設備或上述的任意合適的組合。電腦可讀儲存媒體的更具體的例子(非窮舉的列表)包括:硬碟、隨機存取記憶體、唯讀記憶體、快閃記憶體、光碟、軟碟以及上述的任意合適的組合。此處所使用的電腦可讀儲存媒體不被解釋為瞬時信號本身,諸如無線電波或者其它自由傳播的電磁波、通過波導或其它傳輸媒介傳播的電磁波(例如,通過光纖電纜的光信號)、或者通過電線傳輸的電信號。另外,此處所描述的電腦可讀程式指令可以從電腦可讀儲存媒體下載到各個計算/處理設備,或者通過網路,例如:網際網路、區域網路、廣域網路及/或無線網路下載到外部電腦設備或外部儲存設備。網路可以包括銅傳輸電纜、光纖傳輸、無線傳輸、路由器、防火牆、交換器、集線器及/或閘道器。每一個計算/處理設備中的網路卡或者網路介面從網路接收電腦可讀程式指令,並轉發此電腦可讀程式指令,以供儲存在各個計算/處理設備中的電腦可讀儲存媒體中。執行本發明操作的電腦程式指令可以是組合語言指令、指令集架構指令、機器指令、機器相關指令、微指令、韌體指令、或者以一種或多種程式語言的任意組合編寫的原始碼或目的碼(Object Code),所述程式語言包括物件導向的程式語言,如:Common Lisp、Python、C++、Objective-C、Smalltalk、Delphi、Java、Swift、C#、Perl、Ruby與PHP等,以及常規的程序式(Procedural)程式語言,如:C語言或類似的程式 語言。所述電腦程式指令可以完全地在電腦上執行、部分地在電腦上執行、作為一個獨立的軟體執行、部分在客戶端電腦上部分在遠端電腦上執行、或者完全在遠端電腦或伺服器上執行。 It should be noted that in actual implementation, the modules described in the present invention can be implemented in various ways, including software, hardware or any combination thereof. For example, in some implementations, each module can be implemented using software and hardware, or one of them. In addition, the present invention can also be implemented partially or completely based on hardware. For example, one or more modules in the system can be implemented through integrated circuit chips, system Single chip, complex programmable logic device (Complex Programmable Logic Device, CPLD), Field Programmable Gate Array (FPGA), etc. to achieve. The invention may be a system, method and/or computer program. The computer program may include a computer-readable storage medium having computer-readable program instructions for causing a processor to implement various aspects of the invention. The computer-readable storage medium may be a tangible device that can hold and store instructions for use by an instruction execution device. equipment. The computer-readable storage medium may be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: hard disks, random access memory, read-only memory, flash memory, optical disks, floppy disks, and any suitable combination of the above. As used herein, computer-readable storage media is not to be construed as a reference to transient signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., optical signals through fiber optic cables), or through electrical wires. transmitted electrical signals. In addition, the computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to various computing/processing devices, or downloaded through a network, such as the Internet, a local area network, a wide area network and/or a wireless network to an external computer device or external storage device. Networks may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, hubs and/or gateways. A network card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage on a computer-readable storage medium in each computing/processing device middle. Computer program instructions that perform operations of the present invention may be combination language instructions, instruction set architecture instructions, machine instructions, machine-related instructions, micro-instructions, firmware instructions, or source code or object code written in any combination of one or more programming languages. (Object Code), the programming language includes object-oriented programming languages, such as: Common Lisp, Python, C++, Objective-C, Smalltalk, Delphi, Java, Swift, C#, Perl, Ruby and PHP, etc., as well as conventional programs Procedural programming language, such as C language or similar programs Language. The computer program instructions may execute entirely on the computer, partly on the computer, as stand-alone software, partly on the client computer and partly on a remote computer, or entirely on the remote computer or server. execute on.
請參閱「第2A圖」至「第2C圖」,「第2A圖」至「第2C圖」為本發明複合等化器的優化方法的方法流程圖,其步驟包括:提供多個等化器100,所述等化器包含連續時間線性等化器100a、前饋等化器100b及決策回授等化器100c(步驟210);接收信號,並且通過濾波器對信號執行等化處理,以及對通道輸出的單脈衝響應進行濾波,並且將單脈衝響應的電壓最大值作為標記且計算前標記和後標記的長度、數量及位置,其中,濾波器具有不斷改變的濾波器係數(步驟220);根據前標記及後標記的長度、數量及位置,分別計算前標記及後標記的符碼間干擾以獲得符碼間干擾的總和的絕對值,並且執行峰值失真分析將符碼間干擾與當前的標記相疊加以獲得單位間隔內的響應曲線,以及持續執行M次遍歷獲得對應的N個參數的眼圖,其中,M及N為正整數(步驟230);測量眼圖的響應曲線以獲得眼高及眼寬,並且在執行等化處理後,基於眼高記錄峰值最高時所對應的濾波器係數、眼高及眼寬(步驟240);以及在所述等化器中,依序篩選連續時間線性等化器、前饋等化器及決策回授等化器以組成多個等化器組合,並且對每一所述等化器組合執行峰值失真分析以獲得相應的眼高及眼寬,再根據記錄及獲得的眼高及眼寬選擇等化效果及穩定性皆為最佳的所述等化器組合(步驟250)。如此一來,便能透過提供多個不同構造的等化器,並且根據獲取的信號受到符碼間干擾影響的程度作為等化器優化的依據,其中,根據通道的單脈衝響應計算信號的符碼間干擾,並且執行峰值失真 分析將ISI與當前的標記相疊加以獲得響應曲線進而獲得相應的眼圖,並且根據眼高及眼寬選擇等化效果及穩定性皆為最佳的等化器組合。 Please refer to "Figure 2A" to "Figure 2C". "Figure 2A" to "Figure 2C" are method flow charts of the optimization method of the compound equalizer of the present invention. The steps include: providing multiple equalizers. 100. The equalizer includes a continuous-time linear equalizer 100a, a feedforward equalizer 100b, and a decision feedback equalizer 100c (step 210); receives a signal, and performs equalization processing on the signal through a filter, and The single pulse response of the channel output is filtered, and the voltage maximum value of the single pulse response is used as a mark and the length, number and position of the front mark and the rear mark are calculated, wherein the filter has a constantly changing filter coefficient (step 220) ;According to the length, quantity and position of the front mark and the rear mark, calculate the inter-symbol interference of the front mark and the rear mark respectively to obtain the absolute value of the sum of the inter-symbol interference, and perform peak distortion analysis to compare the inter-symbol interference with the current The markers are superimposed to obtain the response curve within the unit interval, and M times of traversal are continuously performed to obtain the eye diagram corresponding to N parameters, where M and N are positive integers (step 230); the response curve of the eye diagram is measured to obtain Eye height and eye width, and after performing the equalization process, record the filter coefficient, eye height and eye width corresponding to the highest peak value based on the eye height (step 240); and in the equalizer, filter sequentially Continuous-time linear equalizers, feedforward equalizers and decision feedback equalizers are used to form multiple equalizer combinations, and peak distortion analysis is performed on each of the equalizer combinations to obtain the corresponding eye height and eye level. width, and then select the equalizer combination with the best equalization effect and stability according to the recorded and obtained eye height and eye width (step 250). In this way, multiple equalizers with different structures can be provided, and the degree to which the acquired signal is affected by inter-symbol interference can be used as the basis for equalizer optimization. Among them, the symbol of the signal is calculated based on the single impulse response of the channel. intersymbol interference, and performs peak distortion The analysis superimposes the ISI and the current marker to obtain the response curve and then obtains the corresponding eye diagram, and selects the equalizer combination with the best equalization effect and stability based on eye height and eye width.
以下配合「第3圖」至「第6圖」以實施例的方式進行如下說明,請先參閱「第3圖」,「第3圖」為單脈衝響應及符碼間干擾之示意圖。由於信號在透過通道傳輸的過程中,往往會因為通道時散(Time Dispersion)效應而產生失真,主要的原因在於,當通道頻率響應為非常數振幅與非線性相位時,信號的振幅和相位會因通道響應而失真,從而導致符碼間干擾(即:「ISI+/-」),這會讓接收端無法正確地辨識信號。如「第3圖」所示意,信號處理模組110會將單脈衝響應300(Pulse response)的高峰,即:電壓最大值,記錄為標記310,以及在標記310之前作為前標記320,之後則作為後標記330,並且計算長度、數量及位置,從圖中可以清楚看到後標記330呈現愈來愈發散。 The following description is provided in the form of an embodiment in conjunction with "Figure 3" to "Figure 6". Please refer to "Figure 3" first. "Figure 3" is a schematic diagram of single pulse response and inter-symbol interference. During the process of signal transmission through the channel, distortion is often caused by the channel time dispersion effect. The main reason is that when the channel frequency response is non-constant amplitude and nonlinear phase, the amplitude and phase of the signal will Distortion due to channel response results in inter-symbol interference (i.e. "ISI+/-"), which prevents the receiving end from correctly identifying the signal. As shown in "Figure 3", the signal processing module 110 records the peak value of the single pulse response 300 (Pulse response), that is, the maximum value of the voltage, as a mark 310, and before the mark 310 as the previous mark 320, and after that As the rear mark 330, and calculating the length, quantity and position, it can be clearly seen from the figure that the rear mark 330 appears increasingly divergent.
如「第4圖」所示意,「第4圖」為本發明響應曲線疊加前後之示意圖。在實際實施上,執行模組120會根據前標記320及後標記330的長度、數量及位置,分別計算前標記320及後標記330的符碼間干擾以獲得符碼間干擾的總和的絕對值,並且執行峰值失真分析將符碼間干擾與當前的標記310相疊加以獲得單位間隔(1UI)內的響應曲線,以及持續執行M次遍歷獲得對應的N個參數的眼圖400,其中,M及N為正整數。以「第4圖」為例,虛線為疊加前的信號,實線則為疊加符碼間干擾的響應曲線,隨著符碼間干擾的疊加,眼圖400會越來越閉合,將計算得到的曲線與邏輯零「0」電平進行比較即可得到左右相交點。接著,可根據響應曲線得到眼高及眼寬,並且以眼高為衡量標準找到等化後的峰值最高時所對應的濾波器係數、眼高及眼寬。 As shown in "Figure 4", "Figure 4" is a schematic diagram before and after superposition of the response curves of the present invention. In actual implementation, the execution module 120 will calculate the inter-symbol interference of the pre-mark 320 and the post-mark 330 respectively according to the length, quantity and position of the pre-mark 320 and the post-mark 330 to obtain the absolute value of the total inter-symbol interference. , and perform peak distortion analysis to superimpose the inter-symbol interference with the current mark 310 to obtain the response curve within the unit interval (1UI), and continuously perform M times of traversal to obtain the corresponding eye diagram 400 of N parameters, where, M and N is a positive integer. Taking "Figure 4" as an example, the dotted line is the signal before superposition, and the solid line is the response curve of superimposed inter-symbol interference. With the superposition of inter-symbol interference, the eye diagram 400 will become more and more closed, and the calculated The left and right intersection points can be obtained by comparing the curve with the logic zero "0" level. Then, the eye height and eye width can be obtained according to the response curve, and the eye height is used as a measurement standard to find the filter coefficient, eye height, and eye width corresponding to the highest equalized peak value.
如「第5圖」所示意,「第5圖」為應用本發明篩選不同等化器組合的眼圖對比之示意圖。首先,未經過優化的通道信號如眼圖510所示意,眼圖510及其眼高與眼寬皆不明顯。接著,經過連續時間線性等化器及前饋等化器優化後,眼圖520有了明顯的變化,在此基礎上,再與決策回授等化器的優化組合將大幅減少信號的符碼間干擾,使得眼圖更加清晰、眼寬變寬,如眼圖530所示意,如此一來,在傳輸線上的信號,其穩定時間將大幅提高。 As shown in "Figure 5", "Figure 5" is a schematic diagram of eye diagram comparison for screening different equalizer combinations using the present invention. First, the unoptimized channel signal is as shown in the eye diagram 510, and the eye diagram 510 and its eye height and eye width are not obvious. Then, after optimization of the continuous-time linear equalizer and feedforward equalizer, the eye diagram 520 has changed significantly. On this basis, the optimized combination with the decision feedback equalizer will greatly reduce the signal code. Interference makes the eye diagram clearer and the eye width wider, as shown in the eye diagram 530. As a result, the signal on the transmission line will have a significantly improved settling time.
如「第6圖」所示意,「第6圖」為應用本發明在各種等化器優化下的眼圖對比之示意圖。在實際實施上,可以使用不同等化器組合來進行優化,舉例來說,第一種優化方式是先優化連續時間線性等化器,使眼圖最佳化後,再固定連續時間線性等化器並優化決策回授等化器,用以進一步使眼圖最佳化,如:眼圖610所示意。第二種優化方式是先優化決策回授等化器,再優化連續時間線性等化器,如:眼圖620所示意。第三種優化方式是採用複合優化方式,同時優化連續時間線性等化器及決策回授等化器,使眼圖最佳化,如:眼圖630所示意。從圖中可以清楚看到,眼圖630最為清晰,也就是說,第三種優化方式具有最佳的優化效果。特別要說明的是,在參數選擇上,優化次數為M、優化參數為N,M取決於N的多少、優化參數的範圍、步長、優化時間等等因素。除此之外,關於等化器優化判斷因子的選擇,將取決於系統的設計標準來確定,由於不同系統的接收端對眼高或眼寬的靈敏度不同,從而導致對判斷因子的選擇(0至1)為數值0時,以眼高的最佳值為優化目標,判斷因子為數值1時,以眼寬的最佳值為優化目標。 As shown in "Figure 6", "Figure 6" is a schematic diagram of eye diagram comparison under various equalizer optimizations using the present invention. In actual implementation, different combinations of equalizers can be used for optimization. For example, the first optimization method is to first optimize the continuous-time linear equalizer to optimize the eye diagram, and then fix the continuous-time linear equalizer. and optimize the decision feedback equalizer to further optimize the eye diagram, as shown in eye diagram 610. The second optimization method is to optimize the decision feedback equalizer first, and then optimize the continuous-time linear equalizer, as shown in eye diagram 620. The third optimization method is to use a composite optimization method to simultaneously optimize the continuous-time linear equalizer and the decision feedback equalizer to optimize the eye diagram, such as eye diagram 630. It can be clearly seen from the figure that the eye diagram 630 is the clearest, which means that the third optimization method has the best optimization effect. In particular, in terms of parameter selection, the number of optimization times is M and the optimization parameters are N. M depends on the number of N, the range of optimization parameters, step size, optimization time and other factors. In addition, the selection of equalizer optimization judgment factors will depend on the design standards of the system. Since the receivers of different systems have different sensitivities to eye height or eye width, the selection of judgment factors (0 to 1) When the value is 0, the optimal value of eye height is used as the optimization target. When the judgment factor is 1, the optimal value of eye width is used as the optimization target.
綜上所述,可知本發明與先前技術之間的差異在於透過提供多個不同構造的等化器,並且根據獲取的信號受到符碼間干擾影響的程度作為等化 器優化的依據,其中,根據通道的單脈衝響應計算信號的符碼間干擾,並且執行峰值失真分析將ISI與當前的標記相疊加以獲得響應曲線進而獲得相應的眼圖,並且根據眼高及眼寬選擇等化效果及穩定性皆為最佳的等化器組合,藉由此一技術手段可以解決先前技術所存在的問題,進而達成提升信號高速傳輸的信號完整性及穩定性之技術功效。 In summary, it can be seen that the difference between the present invention and the prior art is that by providing a plurality of equalizers with different structures, and based on the degree of the acquired signal affected by inter-symbol interference as the equalizer The basis for processor optimization, in which the inter-symbol interference of the signal is calculated based on the single impulse response of the channel, and peak distortion analysis is performed to superimpose the ISI and the current marker to obtain the response curve and then the corresponding eye diagram, and according to the eye height and The eye width selection equalizer combination has the best equalization effect and stability. This technical means can solve the problems existing in the previous technology, thereby achieving the technical effect of improving the signal integrity and stability of high-speed signal transmission. .
雖然本發明以前述之實施例揭露如上,然其並非用以限定本發明,任何熟習相像技藝者,在不脫離本發明之精神和範圍內,當可作些許之更動與潤飾,因此本發明之專利保護範圍須視本說明書所附之申請專利範圍所界定者為準。 Although the present invention has been disclosed in the foregoing embodiments, they are not intended to limit the present invention. Anyone skilled in the similar art can make some modifications and modifications without departing from the spirit and scope of the present invention. Therefore, the present invention is The scope of patent protection shall be determined by the scope of the patent application attached to this specification.
100:等化器 100: Equalizer
100a:連續時間線性等化器 100a: Continuous-time linear equalizer
100b:前饋等化器 100b: Feedforward equalizer
100c:決策回授等化器 100c: Decision Feedback Equalizer
110:信號處理模組 110:Signal processing module
120:執行模組 120:Execute module
130:測量模組 130:Measurement module
140:篩選模組 140: Screening module
Claims (10)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW111131239A TWI816518B (en) | 2022-08-19 | 2022-08-19 | Optimization system for composite equalizer and method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW111131239A TWI816518B (en) | 2022-08-19 | 2022-08-19 | Optimization system for composite equalizer and method thereof |
Publications (2)
Publication Number | Publication Date |
---|---|
TWI816518B true TWI816518B (en) | 2023-09-21 |
TW202410671A TW202410671A (en) | 2024-03-01 |
Family
ID=88966210
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW111131239A TWI816518B (en) | 2022-08-19 | 2022-08-19 | Optimization system for composite equalizer and method thereof |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI816518B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130243066A1 (en) * | 2012-03-13 | 2013-09-19 | Lsi Corporation | Decision feedforward equalization |
WO2017209884A1 (en) * | 2016-06-01 | 2017-12-07 | Altera Corporation | Integrated circuit with continuously adaptive equalization circuitry |
US20210091983A1 (en) * | 2019-09-19 | 2021-03-25 | MACOM Technology Solutions Holding, Inc. | Using isi or q calculation to adapt equalizer settings |
US20210119835A1 (en) * | 2019-12-04 | 2021-04-22 | Intel Corporation | Transceiver parameter determination |
TW202201933A (en) * | 2020-02-26 | 2022-01-01 | 台灣積體電路製造股份有限公司 | Decision feedback equalizer |
-
2022
- 2022-08-19 TW TW111131239A patent/TWI816518B/en active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130243066A1 (en) * | 2012-03-13 | 2013-09-19 | Lsi Corporation | Decision feedforward equalization |
WO2017209884A1 (en) * | 2016-06-01 | 2017-12-07 | Altera Corporation | Integrated circuit with continuously adaptive equalization circuitry |
US20210091983A1 (en) * | 2019-09-19 | 2021-03-25 | MACOM Technology Solutions Holding, Inc. | Using isi or q calculation to adapt equalizer settings |
US20210119835A1 (en) * | 2019-12-04 | 2021-04-22 | Intel Corporation | Transceiver parameter determination |
TW202201933A (en) * | 2020-02-26 | 2022-01-01 | 台灣積體電路製造股份有限公司 | Decision feedback equalizer |
Also Published As
Publication number | Publication date |
---|---|
TW202410671A (en) | 2024-03-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
TWI709318B (en) | Clock data recovery system, serializer/deserializer receiver and operation method thereof | |
US11115247B2 (en) | Methods and circuits for asymmetric distribution of channel equalization between devices | |
US9014318B2 (en) | Self-calibrating continuous-time equalization | |
US8325793B2 (en) | Precursor ISI cancellation using adaptation of negative gain linear equalizer | |
US8873615B2 (en) | Method and controller for equalizing a received serial data stream | |
US8582635B2 (en) | Sparse and reconfigurable floating tap feed forward equalization | |
CN105791188A (en) | PAM data communication system with reflection cancellation | |
KR101417162B1 (en) | Method and apparatus for non-linear decision-feedback equalization in the presence of asymmetric channel | |
US8208529B2 (en) | Equalization apparatus and method of compensating distorted signal and data receiving apparatus | |
TW202037097A (en) | Decision feed-forward equalizer,serializer/deserializer and method for reciver generating data | |
US8036305B2 (en) | Precoder design for different channel lengths | |
KR20090057073A (en) | Method and apparatus for determining latch position for decision-feedback equalization using single-sided eye | |
TWI816518B (en) | Optimization system for composite equalizer and method thereof | |
US11018909B2 (en) | Receiver adaptation using stochastic gradient hill climbing with genetic mutation | |
CN116827735A (en) | Coefficient updating method for switching double modes of Ethernet decision feedback structure | |
WO2023273589A1 (en) | Signal decision equalization method and apparatus | |
JP2022183142A (en) | Communication link equalization process, computing device, and equalizer optimization method | |
CN117640305A (en) | Optimization system and method for composite equalizer | |
US11811566B2 (en) | Methods and systems for performing adaptive equalization of data | |
US11831475B1 (en) | Receiver using pseudo partial response maximum likelihood sequence detection | |
US12034576B2 (en) | Receiver and transmitter adaptation using stochastic gradient hill climbing with genetic mutation | |
US11381431B2 (en) | Receiver and transmitter adaptation using stochastic gradient hill climbing with genetic mutation | |
CN117294323A (en) | Link training method, device, receiving equipment and computer readable storage medium | |
TW202339449A (en) | Error detection and correction device and associated method | |
CN116935910A (en) | Signal processing method and device for memory system interface circuit |