[go: up one dir, main page]

TWI374609B - Device and method for adaptive adjustment of length of finite impulse response filter - Google Patents

Device and method for adaptive adjustment of length of finite impulse response filter Download PDF

Info

Publication number
TWI374609B
TWI374609B TW97135008A TW97135008A TWI374609B TW I374609 B TWI374609 B TW I374609B TW 97135008 A TW97135008 A TW 97135008A TW 97135008 A TW97135008 A TW 97135008A TW I374609 B TWI374609 B TW I374609B
Authority
TW
Taiwan
Prior art keywords
finite impulse
impulse response
circuit
signal
specific
Prior art date
Application number
TW97135008A
Other languages
Chinese (zh)
Other versions
TW200935736A (en
Inventor
Jung Jen Chang
Chi Shun Weng
Ming Je Li
Meng Han Hseih
Original Assignee
Realtek Semiconductor Corp
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 Realtek Semiconductor Corp filed Critical Realtek Semiconductor Corp
Priority to US12/322,503 priority Critical patent/US8285772B2/en
Publication of TW200935736A publication Critical patent/TW200935736A/en
Application granted granted Critical
Publication of TWI374609B publication Critical patent/TWI374609B/en

Links

Landscapes

  • Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)

Description

1374609 【修正無劃線版] 九、發明說明: 【發明所屬之技術領域】 本發明係關於分配數位濾波器資源之裝置及方法,特 別是關於依據與決定數位濾波器階數(tap數)相關之參數 而適應性地分配數位濾波器資源之裝置及方法。 【先前技術】 數位滤波器的構成元件包含加法器、乘法器及/或延遲 元件等!有別於類比濃波器的構成元件如電阻、電容、電 感等。由於數位濾波器具有溫度的恆定性,如加法器、乘 法器及延遲元件等對於溫度改變的靈敏度不高,故只要適 S地決疋數位遽波|§之參數’即可以達到一定的準破度盘 穩定度。。 一般而言,數位濾波器包含有限脈衝響應(fmite impulse response)濾波器與無限脈衝響應(infmiteimpulse response)濾波器,以N階的有限脈衝響應濾波器為例,其 輸入訊號與輸出訊號之間的關係’可以由下述的差分等式(j) 及其轉換函數來表示: y[n] = Er=〇bix[n-i] (1) H[z] = Zn=〇bnz~n (2) 其中,x[n-i]是輸入訊號,y[n]為輸出訊號,n為濾波器的 階數(tap數)或稱為長度,匕為濾波器的係數。此外,如 等式(1)右方所示’N階的濾波器共具有(N+i)項。h[z]為y[n] 的·Ζ轉換函數。 如同此技術領域中所熟知,舉例說明的Ν階有限脈衝 響應濾波器可以以圖1所示的方塊圖來表示。 5 【修正無劃線版】 一般而言’有限脈衝響應濾波器的階數愈多,則濾波 效果愈佳’但是,階數愈多,則成本愈南。 圖2是方塊圖,顯示適應性有限脈衝響應濾波器的架 構200。如同所知,適應性濾波器使用回饋以自行調整濾波 器的係數。如圖2所示,校正處理器202根據有限脈衝響 應濾波器201的輪出ύ(η)與期望訊號u(n)的誤差回饋與有限 脈衝響應渡波器的輸入x(n)(= u(n)+i⑻),來調整有限脈衝 響應滤波器2〇1的係數以使輸出ύ(η)最佳化,其中,χ(η) = u(n)+i(n),i(n)為干擾雜訊。因此,適應性有限脈衝響應遽 波器常被用於無法事先預知干擾之訊號處理。 有限脈衝響應濾波器的應用非常廣泛。舉例而言,在 乙太網路連線應用中,會使用有限脈衝響應濾波器,以消 •除各種訊號干擾。在乙太網路連線中,在使用多組無屏蔽 的絞線對作為多通道連線之情形中,在一通道中,訊號傳 送時會於通道本身產生回聲(echo)干擾並對相鄰的通道造 成近端串音(near-end cross talk,簡稱NEXT),且通道本 身會受到來自相鄰通道的近端串音。在此情形中,可以使 用有限脈衝響應濾波器來消除通道中的這些干擾源或雜 訊。由於有限脈衝響應濾波器的階數在電路設計階段即視 應用情形而被設計為一固定階數,所以,當有限脈衝響應 濾波器實際用於不同的應用時,訊號品質可能會因例如通 訊線品質、線長或通訊環境等變化因素而無法得到良好的 濾波效果,亦即,階數預先固定的有限脈衝響應濾波器無 法適應性地因應環境變化下的干擾,亦或為了因應最糟情 【修正無劃線版】 形而將;慮波器階數設計過多而造成無謂的浪費。舉例而 言,固定使用Μ階有限脈衝響應濾波器資源來處理回聲干 擾,另外固定使用Μ階有限脈衝響應濾波器資源來處理近 端串音時,當回聲干擾輕微而只需少於厘階資源來進行濾 波,但近端串音嚴重而需要多於“階資源來進行濾波時, 在回聲干擾處理方面即產生浪費,但在近端串音處理方面 卻出現遽波效果不足之情形。 因此,需要能夠在有限的濾波器資源下,即在濾波器 總階數固定不變情形下,視應用環境而適應性地調整分配 有限濾波器的資源之裝置及方法,藉以取得符合所需的優 良訊號品質。 【發明内容】 慮及上述,本發明提供可以依據與決定濾波器階數相 關之參數而適應性分配有限脈衝響應濾波器資源的裝置及 方法。 根據本發明的一態樣,提供可適應性調整特定用途的 有限脈衝響麟波㈣數的裝置,肋使傳輸訊號品質最 佳化,其包括:複數個特定有限脈衝響應濾波器,各個特 疋有限脈衝響應濾波器具有個別的固定階數,輸入端接收 造成個別特定干擾之個別特定訊號’纟依據侧之據波器 係數以提供賴於個卿定干擾的脈衝補償,肋消除個 別特定訊號在該傳輸訊號中造成的個別特定干擾;至少一 個非特定修脈衝響麟波H,具有@定_數,可適應 性地串接該複數個特定有限脈衝響應濾波器中的任一特定 1374609 [修正無劃線版】 有限脈衝響應遽波器;控制單元,用以控制該至少一 特定有限脈衝響應滤波器與該多個特定有限 器的不同串接組合;及評估單元,評估不同組合所對2 訊號品質,找出-對應最佳訊號品質之組合,並促使 制單元選取該串接組合以用於消除該傳輸訊號中的干擾Λ。工 根2發明的另-態樣’提供—種有限脈衝響應遽波 裝置,包3:—第一有限脈衝響應濾波電路,具有一 固定階數,用來濾波-第-訊號;—第二有限脈衝響應遽 波電路,具有-第二固定階數,用來遽波一第二訊號,其 —開關電路,分別輕接 =第-與第二植脈衝響顧波電路;—非特定有限脈衝 響麟波電路,具有一第三固定階數;一評估電路,用來 依據-或多個參數來评估濾波該第一訊號之效果以及評估 渡波該第二訊號之效果’以產生一或多個評估結果;以及 一控制電路,用來依據該-或多個評估結果將該非特定有 限脈衝響麟波電路之Μ階經由該開關電路祕至該第一 有限脈衝響應遽波電路,並將該非特定有限脈衝響應滤波 電路之Ν ρ總由該開關電路輕接至該第二有限脈衝響應濾 波電路,其中該Μ與該Ν為零或正整數。 根據本發明的再-態樣,提供—種乙太網路通訊裝 置’具有有限脈衝響應濾波資源可分配之功能,該乙太網 路通訊裝置包含一收發器.,該收發器包含:-第-有限脈 衝響應濾波電路miu定階數,用來滤波一第一 訊號;一第二有限脈衝響應濾波電路,具有一第二固定階 【修正無劃線版】 數,用來濾波-第二訊號,其巾該第二訊號不同於該第一 訊號;-開關電路,分_接該第H有限脈衝響應 遽,電路卜非特定有限脈衝響應舰電路,具有一第三 固定階數;-評估電路,用來依據—或多個參數來評估滤 波該第-魏之絲以及評傾顧第二訊號之效果以 產生-或多個評估結果;以及_控制電路,用來依據該一 或多個評估結果將該非特定有限脈衝響應遽波電路之Μ階 經由該開關電路耦接至該第一有限脈衝響應濾波電路,並 將該非特疋有限脈衝響應〉慮波電路之Ν階經由該開關電路 耦接至該第二有限脈衝響應濾波電路,其中該厘與該 零或正整數。 根據本發明的另一態樣,提供一種乙太網路通訊裝 置,具有有限脈衝響應濾波資源可分配之功能,該乙太網 路通訊裝置包含:一第一收發器,該第一收發器包含:一 第一有限脈衝響應濾波電路,具有一第一固定階數,用來 濾波一第一訊號;一第二收發器,該第二收發器包含:一 第二有限脈衝響應濾波電路,具有一第二固定階數,用來 滤波一第二訊號,其中該第二訊號不同於該第一訊號;一 開關電路,分別耦接該第一與第二有限脈衝響應濾波電 路;一非特定有限脈衝響應濾波電路,具有一第三固定階 數,一s平估電路,用來依據一或多個參數來評估滤波該第 一訊號之效果以及評估濾波該第二訊號之效果,以產生一 或多個評估結果;以及一控制電路,用來依據該一或多個 評估結果將該非特定有限脈衝響應濾波電路之“階經由該 【修正無劃線版】 開關電路耦接至該第一有限脈衝響應濾波電路,並將該非 特定有限脈衝響應濾波電路階經由該開關電路耦接至 該第二有限脈衝響應濾波電路,其中該Μ與該N為零或正 整數。 根據本發明的再一態樣,提供一種乙太網路通訊裝 置,具有有限脈衝響應濾波資源可分配之功能,該乙太網 路通訊裝置包含:ϋ發||,絲接收__第一訊號, 該第-收發器包含:H限脈衝響應濾波電路,用來 濾波該第一訊號;以及一第一非特定有限脈衝響應濾波電 路’具有-第-岐階數;-第二收發器,用來接收一第 一訊號’該第二收發器包含:一第二有限脈衝響應濾波電 路’用來濾波該第二訊號;以及一第二非特定有限脈衝響 應遽波電路,具有U定階數;—開關電路,分別輕 接該第一與第二非特定有限脈衝響應濾波電路;一評估電 路,用來依據一或多個參數來評估濾波該第一訊號之效果 以及評估濾波該第二訊號之效果,以產生一或多個評估結 果;以及-控制電路,用來依據該__或多個評估結果將^ 第一及第二非特定有限脈衝響應濾波電路之階數總和中之 Μ階經由該開關電路祕线第—有限脈衝響應滤波電 路’並將該第-及第二雜定有限脈衝響應濾波電路之階 數總和中之Ν階經由該開關電路耦接至該第二有限脈衝響 應濾波電路,其中該Μ與該Ν為零或正整數。.. 根據本發明的又-態樣’提供可適應性調整特定用途 的有限脈衝響應驗器贿的方法,肋使傳輸訊號品質 【修正無劃線版】 最?化,包括下述步驟:設置複數個特定有限脈衝響應瀘 波益各個特疋有限脈衝響應遽波器具有個別的固定階 數輸入端接枚造成個別特定干擾之個別特定訊號,並依 據個別之係數以提供對應於個職定干擾的脈衝補 ,用以2除個別特定訊號在該傳輸訊號中造成的個別特 疋干擾’設置至少-個非特定有限脈衝響減波器,該至 少-個非狀紐麵響絲波^具有固錢數,以及, 可適應性地串接該複數個特定有限脈衝響應·器中的任 特疋有限脈衝響應遽波^;串接組合產生步驟,將該至 少-個非特定魏脈鱗麟波Ε無多娜定有限脈衝 響應滤波器形成不同串接組合;及評估步驟,評估不同組 合所對應之喊品質’找出__對絲佳峨品質之組合, 並將該串接組合用於消除該傳輸訊號中的干擾。 依據本發明之實施例,濾波H之Φ接組合所對應之訊 號品質之相關參數可以是該組合中該至少一個非特定有限 脈衝響應遽波器的個別係數的絕對值的平方的總和、個別 係數的絕對值總和、或訊號雜訊比,然上述實施例並非本 發明之限制,本技術領域具有通常知識可_任何習知與 訊號品質有關之數值及運算方式來得到該相關參數。 另外’本發明之裝置及方法可用於但不限制定於網 路交換器、網路集線器、網路路由器、網路閘道器、及 服器·、或資料宁心。 . σ 5 從下述參考附圖之本發明的實施例說明中,當可更容 易瞭解本發明之目的、技術内容、特點及其所達成之功效。 【修正無劃線版】 【實施方式】 本發明提供可適應性調整特定用 波器(以下簡稱FIR)的階數之裝置及方土有限脈衝響應滤1374609 [Fixed Unlined Edition] IX. Description of the Invention: [Technical Field] The present invention relates to an apparatus and method for allocating digital filter resources, and more particularly to correlating with determining a digital filter order (tap number) Apparatus and method for adaptively allocating digital filter resources with parameters. [Prior Art] The constituent elements of the digital filter include adders, multipliers, and/or delay elements, etc.! It is different from the components of the analog mixer such as resistors, capacitors, and inductors. Since the digital filter has a constant temperature, such as adders, multipliers, and delay elements, the sensitivity to temperature changes is not high, so as long as the appropriate number of chopping waves can be achieved by a certain number of parameters. Dial stability. . In general, a digital filter includes a finite impulse response filter and an infmite impulse response filter, and an N-order finite impulse response filter is used as an example between the input signal and the output signal. The relationship ' can be expressed by the differential equation (j) and its transfer function as follows: y[n] = Er=〇bix[ni] (1) H[z] = Zn=〇bnz~n (2) where , x[ni] is the input signal, y[n] is the output signal, n is the order of the filter (tap number) or called the length, and 匕 is the coefficient of the filter. Further, the filters of the 'N order' shown as shown on the right side of the equation (1) have a total of (N + i) terms. h[z] is the Ζ[Ζ] conversion function. As is well known in the art, the illustrated finite order impulse response filter can be represented by the block diagram shown in FIG. 5 [Correction-free version] Generally speaking, the more the order of the finite impulse response filter, the better the filtering effect. However, the more the order, the more cost. Figure 2 is a block diagram showing the architecture 200 of an adaptive finite impulse response filter. As is known, the adaptive filter uses feedback to adjust the coefficients of the filter by itself. As shown in FIG. 2, the correction processor 202 is based on the error feedback of the enthalpy (η) and the desired signal u(n) of the finite impulse response filter 201 and the input x(n) of the finite impulse response ferrite (= u( n) +i(8)) to adjust the coefficient of the finite impulse response filter 2〇1 to optimize the output ύ(η), where χ(η) = u(n)+i(n), i(n) To interfere with noise. Therefore, adaptive finite impulse response choppers are often used for signal processing that does not predict interference in advance. The application of finite impulse response filters is very extensive. For example, in an Ethernet connection application, a finite impulse response filter is used to eliminate various signal interferences. In the case of an Ethernet connection, when multiple sets of unshielded twisted pairs are used as multi-channel connections, in one channel, the signal transmission will cause echo interference and adjacent to the channel itself. The channel causes a near-end cross talk (NEXT), and the channel itself is subject to near-end crosstalk from adjacent channels. In this case, a finite impulse response filter can be used to eliminate these sources or noise in the channel. Since the order of the finite impulse response filter is designed to be a fixed order in the circuit design phase depending on the application, when the finite impulse response filter is actually used in different applications, the signal quality may be due to, for example, the communication line. Quality, line length or communication environment and other factors can not get good filtering effect, that is, the finite impulse response filter with pre-fixed order can not adapt to the interference under environmental changes, or to cope with the worst situation [ Corrected the unlined version] The shape is too much; the design of the wave filter is too much and causes unnecessary waste. For example, the finite-order finite impulse response filter resource is fixedly used to process the echo interference, and the finite-order finite impulse response filter resource is fixedly used to process the near-end crosstalk. When the echo interference is slight, only less than the centimeter resource is needed. To filter, but the near-end crosstalk is serious and requires more than "order resources for filtering, it is wasteful in the echo interference processing, but there is a lack of chopping effect in the near-end crosstalk processing. Therefore, It is necessary to be able to adaptively adjust the resources and equipment of the finite filter according to the application environment under the condition of limited filter resources, that is, when the total order of the filter is fixed, so as to obtain the excellent signal that meets the requirements. SUMMARY OF THE INVENTION In view of the above, the present invention provides an apparatus and method for adaptively distributing finite impulse response filter resources in accordance with parameters associated with determining filter orders. According to one aspect of the present invention, an adaptable Strictly adjust the finite impulse ringing wave (four) number of devices for specific purposes, the rib optimizes the quality of the transmitted signal, including a plurality of specific finite impulse response filters, each characteristic finite impulse response filter having an individual fixed order, and the input receiving the individual specific signal causing the specific specific interference '纟 according to the side of the wave coefficient to provide a The pulse compensation of the interference, the rib eliminates the individual specific interference caused by the individual specific signals in the transmission signal; at least one non-specific repair pulse lining H, with @定_number, adaptively concatenating the plurality of specific limited Any one of the impulse response filters 1374609 [corrected unlined version] finite impulse response chopper; a control unit for controlling different concatenation of the at least one specific finite impulse response filter and the plurality of specific finite devices The combination; and the evaluation unit evaluates the quality of the 2 signals of the different combinations, finds the combination of the best signal quality, and causes the unit to select the serial combination for eliminating interference in the transmitted signal. Another aspect of the invention provides a finite impulse response chopper device, package 3: - a first finite impulse response filter circuit having a fixed The order is used to filter the -first signal; the second finite impulse response chopper circuit has a second fixed order for chopping a second signal, and the switching circuit is respectively lightly connected to the first - and a second implant pulse response circuit; - a non-specific finite impulse ring oscillating circuit having a third fixed order; an evaluation circuit for evaluating the effect of filtering the first signal and evaluating based on - or a plurality of parameters Passing the effect of the second signal to generate one or more evaluation results; and a control circuit for using the switching circuit to determine the order of the non-specific finite impulse ringing circuit based on the one or more evaluation results The first finite impulse responds to the chopper circuit, and the Νρ of the non-specific finite impulse response filter circuit is lightly connected to the second finite impulse response filter circuit by the switch circuit, wherein the Μ and the Ν are zero or positive integer According to a re-state of the present invention, there is provided an Ethernet communication device having a function of finite impulse response filtering resource assignability, the Ethernet communication device comprising a transceiver. The transceiver package Included: - a finite impulse response filter circuit miu order number is used to filter a first signal; a second finite impulse response filter circuit has a second fixed order [corrected unlined version] number for filtering a second signal, wherein the second signal is different from the first signal; the switching circuit is connected to the H-th finite impulse response, and the circuit is non-specific finite impulse response ship circuit having a third fixed order An evaluation circuit for evaluating the effect of filtering the first-dimensional filament and evaluating the second signal to generate - or a plurality of evaluation results based on - or a plurality of parameters; and a control circuit for using the one or more The evaluation result couples the order of the non-specific finite impulse response chopper circuit to the first finite impulse response filter circuit via the switch circuit, and passes the non-characteristic finite impulse response>wave circuit to the circuit through the switch circuit The second finite impulse response filter circuit is coupled to the zero or positive integer. According to another aspect of the present invention, an Ethernet communication device is provided, which has a function of finite impulse response filtering resource assignability, and the Ethernet communication device includes: a first transceiver, the first transceiver includes a first finite impulse response filter circuit having a first fixed order for filtering a first signal; a second transceiver comprising: a second finite impulse response filter circuit having a a second fixed order for filtering a second signal, wherein the second signal is different from the first signal; a switching circuit coupled to the first and second finite impulse response filtering circuits respectively; and a non-specific finite pulse The responsive filter circuit has a third fixed order, and an s averaging circuit for evaluating the effect of filtering the first signal and evaluating the effect of filtering the second signal according to one or more parameters to generate one or more And a control circuit for using the non-specific finite impulse response filter circuit according to the one or more evaluation results The switch circuit is coupled to the first finite impulse response filter circuit, and the non-specific finite impulse response filter circuit step is coupled to the second finite impulse response filter circuit via the switch circuit, wherein the Μ and the N are zero or positive According to still another aspect of the present invention, an Ethernet communication device is provided, which has a function of finite impulse response filtering resource assignability, and the Ethernet communication device includes: bursting ||, silk receiving __ a signal, the first transceiver includes: a H-limit impulse response filter circuit for filtering the first signal; and a first non-specific finite impulse response filter circuit having a -th-th order; - a second transceiver For receiving a first signal, the second transceiver includes: a second finite impulse response filter circuit for filtering the second signal; and a second non-specific finite impulse response chopper circuit having a U-level a switching circuit for respectively connecting the first and second non-specific finite impulse response filter circuits; an evaluation circuit for evaluating the first signal according to one or more parameters And evaluating an effect of filtering the second signal to generate one or more evaluation results; and - a control circuit for filtering the first and second non-specific finite impulse responses according to the __ or the plurality of evaluation results The first order of the order of the circuits is passed through the switch circuit, the finite impulse response filter circuit of the first line and the second and second finite impulse response filter circuits are passed through the switch circuit And coupled to the second finite impulse response filter circuit, wherein the Μ and the Ν are zero or a positive integer.. according to the present invention, the trait of the invention provides an adaptable adjustment for a specific use of the finite impulse response tester The method, the rib makes the transmission signal quality [corrected non-line version] the most, including the following steps: setting a plurality of specific finite impulse responses, each characteristic finite impulse response chopper has an individual fixed order input Individually specific signals that cause individual specific interferences, and provide pulse supplements corresponding to individual job disturbances according to individual coefficients, for dividing 2 individual signals in the transmission signal Individual characteristic disturbances 'set at least one non-specific finite impulse noise reducer, the at least one non-shaped nucleus wave has a solid number, and adaptively cascades the plurality of specific finite任 疋 finite impulse response 遽 ^ ; ; ; ; 串 串 串 串 串 串 串 串 串 串 串 串 串 串 串 ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; And the evaluation step, evaluating the combination of the shouting quality corresponding to the different combinations 'finding __ to the quality of the silk, and using the serial combination to eliminate the interference in the transmitted signal. According to an embodiment of the present invention, the correlation parameter of the signal quality corresponding to the Φ connection combination of the filter H may be the sum of the squares of the absolute values of the individual coefficients of the at least one non-specific finite impulse response chopper in the combination, and the individual coefficients. The sum of the absolute values, or the signal-to-noise ratio, is not limited by the present invention. The prior art has the general knowledge that any of the values and operations related to the signal quality can be used to obtain the relevant parameters. Further, the apparatus and method of the present invention can be used, but not limited to, in network switches, network hubs, network routers, network gateways, and servers, or information. σ 5 From the following description of the embodiments of the present invention with reference to the accompanying drawings, the object of the invention, the technical contents, the features, and the effects thereof can be more easily understood. [Correction-free version] [Embodiment] The present invention provides an apparatus for adapting the order of a specific wave device (hereinafter referred to as FIR) and a square earth finite impulse response filter.

的實施例t,設細㈣ 適應性配置的非特定FIR,以及,依據特^ ^及用於 的不同組合的訊號品質,決定能夠取 號之特定與非特定FIR的串接組合。一 _ 、S ,3係方翻,顯示根據本發明的可適應性調整观 P白數的裝置獅。如圖3所示,根據本發明的裝置綱包含 評估單元302、控制單元3〇4、k個特定m細至通、 J個開關電路(Switching Circuit,例如多工器)蓮至3〇8j、 及J ^固非蚊FIR道至31Qj。特定FIR遍至通係 分別根據輸人至其的訊號以產生對應於各干擾之補償,非 特定FIR3101至31〇j用以適應性地串接在特定fir之後, 以加長特疋FIR的階數’藉以進—步消除特定干擾。控制 單^304控制非特定FIR與特定HR的不同串接組合評 估单兀302根據不同的串接組合的響應強度以決定可以良 好地消除干擾之特定FIR與非特定FIR触佳組合。 了般而5,在不同的應用中,訊號在被傳輸或處理時, 會遭受到多種不同的特定干擾。這些特定的訊號干擾具有 :識別的不同特徵,而當某些特定的干擾對訊號品質有顯 著的衫響.時’其會成為要消除的干擾。因此,根據本發明 的實施例之裝置·,可哺著不同顧職有階數固定之 特定FIR·至遍,以產生響應來消除所傳輸訊號中的 12 【修正無劃線版】 特定干擾。特定FIR3061至306k,可以視應用的情形以用 於消除至少一種以上的不同特定干擾。舉例而言,FIR3〇61 及3062用於干擾A,而其餘的特定FIR3〇63用於干擾B, 其中,FIR3061及3062的階數固定且二者的階數寸視應用 而為相同或不同,用於干擾B的特定FIR之階數亦為固定 且彼此之階數可為相同或不同。或者,全部FIR為特定用 於同一特定干擾的FIR且階數均固定。換言之,各個特定 FIR的階數可以依設計需求而為彼此相同或不同。 在裝置300中,非特定的FIR31〇1至FIR31〇j之階數 亦為固定的且彼此之階數可視需求而為相同或不同的。當 如上所述用於特定干擾的特^ FIR的階數不足以消除特定 干擾而無法取得良好的訊號品質時,非特定的nR31〇1至 310j中至少之一會被串接至所述特定nR之後以產生足 以消除干擾的補償〇 接著,將以裝置300應用於乙太網路為例,以進 說明其操作。 一般而έ,裝置300在用於不同的應用中時,會先進 ^測試操作,在此測試操作中,可以決定裝置·中的特 定及非特定FIR於所述應用環境中的最佳組合。 在乙太網路連線中,若使用多組無屏蔽的絞線對作為 多個通道的連線媒介時,在—通道巾發出喊時,會於通 道本身中造成回聲干擾,賴相_道造成近端串立 之干擾。換言之,—通射的―端發送的訊號^ 在通道巾產生―干擾源並對其它通道產生另-干擾源。 【修正無劃線版】 舉例而言,具有四通道A、B、C及D的乙太網路中,通道 A會遭受本身產生_聲顿,同時也受到來自通道6、= 或D的近端串音干擾。如同—般所知般在以 線對作為喊通道郎太網路連射,回聲干擾及近^ 曰干擾分別為具有不同的可辨識特徵之顯著干擾。 在應用於四通道連接(通道A、B、c、D)的乙太網 路連接(未顯示)中的通道辑,.裝置·中的特定观 數目的為4個(k=4)分別為回聲消除FIR3〇01及近端串音消 除FIR3G62、3063、3G64,非特定FIR31G1和纖的數目 為2ϋ=2)。回聲消除FIR3061用於消除通道A本身發出的 訊號所造成的回聲干擾且階數預設$ Lech。,近端串音消除 FIR 3062、3063、3064分別用於消徐來自相鄰通道B、c、 D對通道A造成的近端串音干擾且階數預設為。兩個 適應性配置的非特定FIR纖和31G2之階數分別為 及LneD。如同一般所知,由於回聲干擾所造成的通道響應 階數遠大於近端串音干擾所造成的通道響應階數,所以, 回聲消除FIR的階數較佳地設定成遠大於近端串音消除 FIR的數’亦即’ Lech。遠大於lnext。此外,此處所使用 的特定及非特定FIR均為適應性FIR。 圖4係顯示具有根據本發明的可適應性調整FIR階數 的裝置之通訊裝置400。如圖4所示,通訊裝置4〇〇包含媒 體存取控_(:)、實體魏子層(PCS)、誠十億位元 媒體獨立介面層(RGMI)、及實體層(ΡΗγ^。實體層包含發 訊器402及收訊器403’根據本發明實施例的裝置3〇〇設於 【修正無劃線版】 收訊器403中。收訊器403又包含加法器4〇5、及等化器 406 〇 當裝置300設置於通訊裝置4〇〇中,而裝置4〇〇經由 乙太網路連線中的通道A與另一對應的通訊裝置R(未顯示) 連接時,在正常操作之前,會進行預先測試。.通訊裝置4〇〇 及通訊裝置R可為個人電腦、網路交換器、集線器、路由 器、閘道器、伺服器、資料中心、或類似之裝置。 在預先測試期間,通訊裝置4〇〇經由通道a發出訊號 SA,同時’對應的通訊裝置R會將原始訊號%經由通道a 傳送至通訊裝置400。對通訊裝置4〇〇而言,所收到的訊號 SR包含原始訊號SQ、訊號SA造成的回聲干擾Seeh〇A、通 道B發出的訊號SB對通道a所造成的近端串音干擾 snextB、通道c發出的訊號sc所造成的近端串音干擾Snextc、 通道D發出的訊號SD所造成的近端串音干擾SnextD、以及 其它雜訊Sn()ise。亦即’ SR可以表示為: SR - S〇+ Sech〇A+SnextB+ Snextc+ SnextD+Snoise 此處’ Sn()ise造成的響應不顯著,並非要處理的對象, 故在往後的說明中,將予以忽略。 因此,如果將訊號SR減掉可能的最大干擾響應 SeCh〇A+SnextB+ snextc+ SnextD,即可得到最佳的訊號傳輸訊品 質。 根據上述,在裝置300中’只要從非特定FIK接於回 聲消除FIR 3061或近端串音消除fiR 3062、3063、3064之 不同組合中,找出提供最大脈衝響應強度的組合,即可得 1374609 c修正無劃線版] 到最佳的通訊品質。注意,由於FIR之係數代表響應的振 幅,所以,下述中會以FIR的係數來表示所產生的響應強 度。In the embodiment t, the non-specific FIR of the (4) adaptive configuration is set, and the serial combination of the specific and non-specific FIRs that can be taken is determined according to the signal quality of the different combinations used. A _, S, and 3 series are turned over to show the device lion that adjusts the P white number according to the present invention. As shown in FIG. 3, the apparatus according to the present invention includes an evaluation unit 302, a control unit 3〇4, k specific m-to-pass, and J switching circuits (for example, multiplexers) to 3〇8j, And J-solid non-mosquito FIR road to 31Qj. The specific FIR traversal system respectively generates compensation corresponding to each interference according to the signal input to it, and the non-specific FIR3101 to 31〇j are used to adaptively cascade after the specific fir to lengthen the order of the special FIR. 'By the way to eliminate specific interference. The control unit 304 controls the different series combination evaluations of the non-specific FIR and the specific HR based on the response strength of the different series combinations to determine the specific FIR and non-specific FIR touch combinations that can well eliminate the interference. In general, in different applications, signals are subject to a variety of different specific interferences when transmitted or processed. These specific signal interferences have different characteristics of recognition, and when certain interferences have a significant shirting sound quality, they become interference to be eliminated. Therefore, the apparatus according to the embodiment of the present invention can feed a specific FIR-to-period with a fixed order of different jobs to generate a response to eliminate 12 [corrected unlined version] specific interference in the transmitted signal. The particular FIRs 3061 through 306k may be used to eliminate at least one of the different specific disturbances depending on the application. For example, FIR3〇61 and 3062 are used for interference A, and the remaining specific FIR3〇63 is used for interference B, where the order of FIR 3061 and 3062 is fixed and the order of the two is the same or different depending on the application. The order of the particular FIR used to interfere with B is also fixed and the order of each other may be the same or different. Alternatively, all FIRs are FIRs that are specific to the same particular interference and the order is fixed. In other words, the order of each particular FIR can be the same or different from each other depending on the design requirements. In apparatus 300, the order of the non-specific FIRs 31〇1 to FIR31〇j is also fixed and the order of each other may be the same or different depending on the requirements. When the order of the special FIR for the specific interference as described above is insufficient to eliminate the specific interference and the good signal quality cannot be obtained, at least one of the non-specific nRs 31〇1 to 310j is concatenated to the specific nR. Thereafter, to generate a compensation sufficient to eliminate interference, the device 300 is applied to the Ethernet network as an example to illustrate its operation. In general, device 300, when used in a different application, will advance the test operation, in which the optimal combination of specific and non-specific FIRs in the device can be determined. In the Ethernet connection, if multiple sets of unshielded twisted pairs are used as the connection medium for multiple channels, when the channel towel is shouted, it will cause echo interference in the channel itself. Causes interference from the near end. In other words, the signal transmitted by the end of the pass-through generates an "interference source" in the channel towel and generates another source of interference on the other channel. [Corrected no-dash version] For example, in an Ethernet network with four channels A, B, C, and D, channel A will suffer from its own _ sound, and also from near channel 6, = or D. Side crosstalk interference. As is generally known, in the line pair as the shouting channel Langtai network, echo interference and near interference are significant interference with different identifiable features. In the channel set in the Ethernet connection (not shown) applied to the four-channel connection (channels A, B, c, D), the number of specific views in the device is 4 (k=4) respectively. Echo cancellation FIR3〇01 and near-end crosstalk cancellation FIR3G62, 3063, 3G64, the number of non-specific FIR31G1 and fiber is 2ϋ=2). The echo cancellation FIR3061 is used to eliminate the echo interference caused by the signal from channel A itself and the order is preset to $ Lech. Near-end crosstalk cancellation FIRs 3062, 3063, and 3064 are used to eliminate near-end crosstalk interference from adjacent channels B, c, and D to channel A, respectively, and the order is preset. The order of the non-specific FIR fiber and the 31G2 of the two adaptive configurations are respectively L and D. As is generally known, the channel response order due to echo interference is much larger than the channel response order caused by near-end crosstalk interference, so the order of the echo cancellation FIR is preferably set to be much larger than the near-end crosstalk cancellation. The number of FIR 'is 'Lech. Far greater than lnext. In addition, the specific and non-specific FIRs used herein are adaptive FIR. Figure 4 is a communication device 400 showing a device having an adaptive adjustment of the FIR order in accordance with the present invention. As shown in FIG. 4, the communication device 4 includes a media access control_(:), an entity Weizi layer (PCS), a Chengyi billion media independent interface layer (RGMI), and a physical layer (ΡΗγ^. entity The layer includes a transmitter 402 and a receiver 403'. The device 3 according to the embodiment of the present invention is disposed in the [corrected unlined version] receiver 403. The receiver 403 further includes an adder 4〇5, and The equalizer 406 is disposed in the communication device 4, and the device 4 is normally connected to another corresponding communication device R (not shown) via the channel A in the Ethernet connection. Before the operation, pre-test will be carried out. The communication device 4 and the communication device R can be a personal computer, a network switch, a hub, a router, a gateway, a server, a data center, or the like. During the communication device 4, the signal SA is sent via the channel a, and the corresponding communication device R transmits the original signal % to the communication device 400 via the channel a. For the communication device 4, the received signal SR Contains the original signal SQ, the signal SA caused by the echo interference Seeh〇A, The near-end crosstalk caused by the signal SB from the channel B to the near-end crosstalk caused by the channel a, the near-end crosstalk caused by the signal sc from the channel c, and the near-end crosstalk caused by the signal SD from the channel D, SnextD And other noise Sn()ise. That is, 'SR can be expressed as: SR - S〇+ Sech〇A+SnextB+ Snextc+ SnextD+Snoise where the response caused by 'Sn()ise is not significant, not the object to be processed Therefore, in the following description, it will be ignored. Therefore, if the signal SR is subtracted from the possible maximum interference response SeCh〇A+SnextB+ snextc+ SnextD, the best signal transmission quality can be obtained. In 300, 'as long as the non-specific FIK is connected to the different combinations of echo cancellation FIR 3061 or near-end crosstalk cancellation fiR 3062, 3063, 3064, find the combination that provides the maximum impulse response strength, then you can get 1374609 c correction without line Version] to the best communication quality. Note that since the coefficient of FIR represents the amplitude of the response, the resulting response strength is expressed by the coefficient of FIR.

在本實施例中,以串接於特定FIR之後的非特定FIR 的個別係數的絕對值平方總和來評估此種串接方式之訊號 品質,藉由評估不同串接方式所對應之訊號品質以決定最 佳的FIR組合。當然,上述利用個別係數的絕對值平方總 和之方式並非本發明之限制,也可以以例如各係數的絕對 值總和、雜訊比(SNR)、或預定的臨界值來作為評估訊號 品質之依據。本技術:領域具有通常知識可應用任何已知的 與訊號品質有關之數值及運算方式來得到該訊號品質相關 參數。 現在,再參考圖3,舉例說明決定響應強度的評估操作。 當裝置300設於通訊裝置4〇〇中以用於通道a時,回 聲消除FIR 3061的輸入端接收通訊裝置4〇〇發出的訊號 SA(輸入1)、近端串音消除FIR 3〇62、3〇63、3〇64的輸入 端分別接收通道B、C、及D發出的訊號SB(輸入2)、sc(輸 入3)、及SD(輸入4)。 首先,在啟始時,評估單元3〇2依據例如SNR等索引 值來判斷目前的FIR ,组態是否足以提供良好的通訊品質。 若否,則進行各種組態的測試,由評估單元3〇2根據回聲 消除FIR 3〇61接收訊號SA時所產生的脈衝響應強度以決 疋FIR 3061的階數Lech。是否足以消除回聲干擾造成的響 應。若為否’則在控制單元304的控制下,將非特定FZRwi 1374609 [修芷無劃線版] 接於回聲消除FIR3061,並將FIR3101此時的個別係數的 絕對值的平方相加作為訊號品質之估測。舉例而言,若FIR 3101的階數為Lnec,在此情形中’其個別係數為 bi[〇]、...MLnec-I]時,則訊號品質之估測可以表示為 |bi[〇]| +、。接著,類似地,將 fir 3102 串接 於FIR 3061之後並執行類似的處理,以取得其訊號品質之 估測。然後,將FIR 3101及3102串接於FIR 3061之後, 並執行類似的處理以取得對應的訊號品質之估測。 評估單元302根據如此得到的訊號品質之估測,將對 應於回音消除最大估測值之FIR 3061與非特定FIR3101、 或3102的組合判定為最佳組合,並據以使控制單元3〇4操 作以選取此組合作為用於消除干擾的FIR組合❶舉例而言, 若所有組合中,FIR 3061與FIR 3101及3102相串接的組 合,其非特定FIR3101和31〇2的響應強度為最大者,則此 組合便被選為消除此四通道乙太網路應用中的干擾之hr 組合。 如果,在啟始時,評估單元3〇2依據例如SNR等索引 值判^前的FIR組態不足以提供良好的通訊品質時,、且 評估單元302判^ FIR 3061的階數足以消除回聲干擾時, 則在控制單7C 304的控制下,對近端串音消除观3〇62執 行類似於上述的處理,將非特定FIR 31〇1、31〇2、或幻⑴ 及3102接於近端串音消除FIR 3G62之後:,以取得對應的塑 應強度。同樣地,對近端宰音消除FIR 3〇63及3〇6 似的操作及處理。 仃頸 17 1374609 【修正無劃線版】 類似地,評估單元302根據如此取得的響應強度,決 定最佳的近端串音FIR與非特定FIR的組合。在上述說明 中,為了簡明起見,依序地對個別的近端串音消除Hr執 行串接非特定FIR之操作,但是,應瞭解,也可以對各個 近端串音干擾FIR同時串接分別的非特^ FIR以找出最佳 組合。舉例而言,可以將FIR 31〇1接於FIR 3〇62之後,同 時’將FIRMO2接於FIR3〇64之後,以取得對應的響應強 度,並據以取得此應用中較佳的組合。 在上述中,評估單元3〇2以來自各FIR的脈衝響應作 f各種FIR組態的判斷依據。但是,本發明不侷限於此。 牛例而言’在U SNR為判斷依據時,评估單元3〇2會接收 來自等化器406的输出以決定FIr組合。 在上述說财,雖然在個別的通道中均包含特定 評估單元、控制單元、等等,但是,各通道也可 匕3各自的特定及非特定的FIR,但共崎 in二各通道的資特定或非特定打心^ 特定fir或非特 在而要時王我部份_於其它 因ί:根縣跡狀她找FIR、械的ς估通丄 等早疋可⑽需要而作最佳的分配, = 化及訊號品質$佳化。 運成胃&配最佳 本發明可用:於網路交換^太例來說明本發明’亦即 路間道器等,但本發明:應乙=由器、網 个丨民於乙太網路。舉例而 【修正無劃線版】 言,本發明可以用於無線通訊裝置。此外,上述說明雖然 以消除不同的干擾源為例說明,但是,本發明不限於此, 而是可以用於例如多通道訊號裝置中fir的適應性配置。 舉例而言,在配備有爪個分別的FIR以用於m個通道的通 訊裝置中,當m個通道中僅有k個通道被使用而其它的m_k 個通道未使用,則可將未使用之m_k個通道的FIR用於使 用中的k個通道,以增進其訊號處理品質。只要是可以應 用FIR資源分配者皆為本發明之實施範圍。 除此之外,尚須說明:本發明的應用情況至少包括: (1)在單一通訊裝置(例如前述網路交換器、網路集 線器、網路路由器、網路閘道器、無線通訊裝置等)之其 中-收發H巾,具有複數轉定有限脈衝響麟波器,各 個特定有赚衝響職波器具有_ _定階數,而適應 性地將其中一個有限脈衝響應濾波器與位於該收發器内或 該收發科的其他雜定魏脈衝響麟波^ φ接 其階數; (2)在單-通訊裝置之複數收發器中,各具有至少一 限脈衝響顏波^ ’各個特定有限脈衝響應遽波 定階數’而她地將其中-個有限脈衝 波讀位於該複數收發器内或該複數收發器外的且 他非特疋有限脈衝響應驗ϋφ接㈣加其階數; ⑶在複數通訊裝置個別所屬之收發財,各至 限脈衝響應遽波器,各靖定有限脈衝響應 皮器具有__定階數,㈣應性地將其中-個有限 1374609 ί修正無劃線版] 脈衝響應濾波器與和其對應之收發器内或收發器外的其他 非特定有限脈衝響應濾波器串接而增加其階數。 以上二種狀況申,非特定有限脈衝響應;慮波器可以與 欲藉串接增加階數的待定有限脈衝響應濾波器位在同一個 收發器中、或不同收發器中、或不在收發器中而設置在電 路其他部分。 本發明可視應用環境而適應性地調整有限脈衝響應濾 波器的階數,可以克服先前技術之有限脈衝響應濾波器的 階數為預先m定且無法視環境變化或干擾訊號的狀態改變 之缺點,因而具有在有限的成本之内取優良訊號品質的 優點。本發明可雜軒度或雜暫_寬(transiti〇n bandwidth)等與決定濾波器之階數相關之參數來以決定如 何分配濾波器之資源,使訊號品質最佳化 一 定HR所負責處理之訊號暫態頻寬愈宽,本二;;分: 較多之非特定FIR之階數予該特定FIR。 根據本發明之可適應性調整FIR階數的裝置可以在 有限的電路成本及雜資源下,消除通道巾的各種干擾源的 響應’以確保訊號的傳輸品質。 雖然在上述說财’以實施例來說明本發明,但是,本 發明不限於所揭示的特定轉。在不悖離本發_精神及後 附申請專现其均等範圍之下,f•於此技藝者可以產生 不同的變化、修改、替代、等等。 【圖式簡單說明】 圖1係方塊圖’顯示N階有限脈衝響應遽波器的-實 20 1374609 【修正無劃線版】 例; 圖2係方塊圖’顯示適紐有限脈衝響應紐器架構的 一實例; 圖3顯示根據本發明的可適應性調整有限脈衝響應濾 波器階數之装置;及 曰一 圖4顯示通訊裝置實施例,其具有根據本發明的可適應 性調整有限脈衝響應濾波器階數之裝置。 【主要元件符號說明】 2〇〇適應性有限脈衝響應濾波器 201有限脈衝響應濾波器 202校正處理器 3〇1可適應性調整有限脈衝響應濾波器階數的裝置 302評估單元 304控制單元 3061-306k 特定 fir 308 l-308j 多 号In this embodiment, the signal quality of the serial connection mode is evaluated by summing the absolute values of the individual values of the individual coefficients of the non-specific FIR after the specific FIR, and determining the signal quality corresponding to the different serial connection methods to determine the signal quality. The best FIR combination. Of course, the above method of utilizing the sum of the squares of the absolute values of the individual coefficients is not a limitation of the present invention, and the absolute value sum, the noise ratio (SNR) of each coefficient, or a predetermined threshold value may be used as the basis for evaluating the signal quality. This technique: The field has the usual knowledge to apply any known value and operation method related to signal quality to obtain the signal quality related parameters. Now, referring again to FIG. 3, an evaluation operation for determining the response strength will be exemplified. When the device 300 is disposed in the communication device 4A for the channel a, the input end of the echo cancellation FIR 3061 receives the signal SA (input 1) and the near-end crosstalk cancellation FIR 3〇62 from the communication device 4, The input terminals of 3〇63 and 3〇64 receive signals SB (input 2), sc (input 3), and SD (input 4) from channels B, C, and D, respectively. First, at the beginning, the evaluation unit 〇2 judges the current FIR based on an index value such as SNR, and whether the configuration is sufficient to provide good communication quality. If not, a test of various configurations is performed, and the evaluation unit 3〇2 cancels the impulse response intensity generated when the FIR 3〇61 receives the signal SA based on the echo to determine the order Lech of the FIR 3061. Is it sufficient to eliminate the response caused by echo interference? If no, then under the control of the control unit 304, the non-specific FZRwi 1374609 [repair unlined version] is connected to the echo cancellation FIR3061, and the square of the absolute value of the individual coefficients of the FIR3101 at this time is added as the signal quality. Estimate. For example, if the order of the FIR 3101 is Lnec, in this case, when the individual coefficients are bi[〇], ... MLnec-I], the estimation of the signal quality can be expressed as |bi[〇] | +,. Next, similarly, fir 3102 is serially connected to FIR 3061 and a similar process is performed to obtain an estimate of its signal quality. Then, FIR 3101 and 3102 are connected in series after FIR 3061, and similar processing is performed to obtain an estimate of the corresponding signal quality. The evaluation unit 302 determines the combination of the FIR 3061 and the non-specific FIR 3101, or 3102 corresponding to the echo cancellation maximum estimated value as the optimal combination based on the estimation of the signal quality thus obtained, and accordingly causes the control unit 3〇4 to operate. In order to select the combination as the FIR combination for eliminating interference, for example, if the combination of FIR 3061 and FIR 3101 and 3102 is in series, the response strength of the non-specific FIR 3101 and 31 〇 2 is the largest. This combination is then chosen to eliminate the hr combination of interference in this four-channel Ethernet application. If, at the beginning, the evaluation unit 3〇2 determines that the previous FIR configuration is insufficient to provide good communication quality according to an index value such as SNR, and the evaluation unit 302 determines that the order of the FIR 3061 is sufficient to eliminate the echo interference. At the same time, under the control of the control sheet 7C 304, a process similar to the above is performed on the near-end crosstalk cancellation view 3, 62, and the non-specific FIR 31〇1, 31〇2, or illusion (1) and 3102 are connected to the near end. Crosstalk is removed after FIR 3G62: to obtain the corresponding plastic strength. Similarly, the near-end slaughter eliminates the operations and processing of FIR 3〇63 and 3〇6. Cervical neck 17 1374609 [Corrected unlined version] Similarly, the evaluation unit 302 determines the optimum combination of the near-end crosstalk FIR and the non-specific FIR based on the response strength thus obtained. In the above description, for the sake of brevity, the operation of serially connecting the non-specific FIR is performed on the individual near-end crosstalk cancellation Hr sequentially, but it should be understood that the respective near-end crosstalk interference FIR can also be serially connected simultaneously. Non-special FIR to find the best combination. For example, the FIR 31〇1 can be connected to the FIR 3〇62, and the FIRMO2 can be connected to the FIR3〇64 to obtain the corresponding response strength, and the preferred combination in this application can be obtained accordingly. In the above, the evaluation unit 3〇2 makes the judgment basis of various FIR configurations with the impulse response from each FIR. However, the invention is not limited thereto. In the case of a cow, when the U SNR is the basis for judgment, the evaluation unit 3〇2 receives the output from the equalizer 406 to determine the FIr combination. In the above-mentioned financial calculations, although specific evaluation units, control units, and the like are included in individual channels, each channel may also have 3 specific and non-specific FIRs, but the commonality of each channel is specific. Or non-specific heart ^ specific fir or non-specific when the king part I _ other reasons ί: root county traces her look for FIR, the evaluation of the equipment, etc., the early can be (10) need for the best distribution , = and signal quality is better. Yuncheng stomach & best with the invention can be used: in the network exchange ^ too to illustrate the invention 'that is, inter-channel devices, etc., but the invention: should be B = by the device, the network of the people in the Ethernet road. For example, [corrected unlined version] The present invention can be applied to a wireless communication device. Furthermore, although the above description has been made by taking as an example to eliminate different interference sources, the present invention is not limited thereto, but can be applied to, for example, an adaptive configuration of fir in a multi-channel signal device. For example, in a communication device equipped with a claw FIR for m channels, when only k channels of m channels are used and other m_k channels are unused, unused The FIR of m_k channels is used for k channels in use to improve the signal processing quality. As long as it is possible to apply the FIR resource allocator, it is the scope of implementation of the present invention. In addition, it should be noted that the application of the present invention includes at least: (1) in a single communication device (such as the aforementioned network switch, network hub, network router, network gateway, wireless communication device, etc.) Among them - send and receive H towel, with a complex number of finite impulse ringer, each specific occupant has a _ _ order, and adaptively put one of the finite impulse response filters The other miscellaneous Wei pulse ring φ φ in the transceiver or the transceiver block is connected to the order thereof; (2) in the multi-transceiver of the single-communication device, each has at least one pulse-limited radiant wave ^ 'specific The finite impulse response chopping order ' and her finite pulse wave is read in the complex transceiver or outside the complex transceiver and the non-special finite impulse response test φ ( (4) plus its order; (3) In the complex communication device, the individual transmission and reception, each of the limit impulse response chopper, each Jingding finite impulse response skin device has __ fixed order, (4) should be modified one of the limited 1374609 ί unlined version Impulse response filter Order to increase the number and other non-specific finite impulse which corresponds to the inside and outside of the transceiver or transceivers response filter connected in series. The above two conditions apply, non-specific finite impulse response; the filter can be in the same transceiver, or different transceivers, or not in the transceiver, with the pending finite impulse response filter bits that are to be cascaded to increase the order. And set it in other parts of the circuit. The invention can adaptively adjust the order of the finite impulse response filter according to the application environment, and can overcome the shortcomings of the prior art finite impulse response filter whose order is pre-determined and cannot change depending on the environment or the state of the interference signal. Therefore, it has the advantage of taking good signal quality within a limited cost. The invention can determine the parameters of the filter according to the parameters of the order of the filter, such as the complexity or the width of the filter, to determine how to allocate the resources of the filter, so that the signal quality is optimized and the HR is responsible for processing. The wider the signal transient bandwidth, the second;; points: more non-specific FIR orders to the specific FIR. The apparatus for adapting the FIR order according to the present invention can eliminate the response of various interference sources of the channel towel with limited circuit cost and resources to ensure the transmission quality of the signal. Although the present invention has been described by way of example, the invention is not limited to the specifics disclosed. Without departing from the scope of this issue, the spirit of the applicant and the scope of the application can be varied, modified, replaced, and so on. [Simple diagram of the diagram] Figure 1 is a block diagram of 'N-order finite impulse response chopper-real 20 1374609 [corrected non-marked version] example; Figure 2 is a block diagram 'displays suitable finite impulse response controller architecture 1 shows an apparatus for adapting the order of finite impulse response filters according to the present invention; and FIG. 4 shows an embodiment of a communication device having adaptively adjusted finite impulse response filtering according to the present invention. Device of the order. [Major component symbol description] 2〇〇Adaptive finite impulse response filter 201 finite impulse response filter 202 correction processor 〇1 adaptability adjustment device of finite impulse response filter order 302 evaluation unit 304 control unit 3061- 306k specific fir 308 l-308j multiple number

3101-310j 非特定 FIR 400通訊裝置 402發訊器 403收訊器 405加法器 406等化器 213101-310j Non-specific FIR 400 communication device 402 transmitter 403 receiver 405 adder 406 equalizer 21

Claims (1)

【修正無劃線版】 十、申請專利範圍: L 一種可適應性調整特定用途的有限脈衝響應濾波器 階數的裝置,用以使傳輸訊號品質最佳化,包括: 複數個特定有限脈衝響應濾波器,各個特定有限脈衝 響應遽波器具有個綱定的階數,以及接收造成個別特定 干擾之個職定訊雜人,並依據_之遽波器係數以提 供對應於個別特定干擾的補償,用以消除個別特定訊號在 該傳輸訊號中造成的個別特定干擾; 至少一個非特定有限脈衝響應濾波器,具有固定的階 數’可適應性地串接該複數個特定有限脈衝響應遽波器中 的任一特定有限脈衝響應濾波器; ,制單元,肋控繼至少—個非特定有限脈衝響應 遽波器與該多個射有限脈衝響麟波器的不同串接电 合;及 ' 評估單元,評估不同組合所對應之訊號品質,找出一 對應最佳纖品質之串接組合,並促使該控解元選取該 串接組合以用於消除該傳輸訊號中的干擾。 2·如申請專利範圍第i項所述之裝置,其中,該多個有 限脈衝響顧m該至少-個麵定有限脈衝響應遽 波器均為適紐有·崎應驗ϋ。 〜 3 人如申請專利範圍第!項所述之裝置,其中,由串接組 a所對應之訊號品質之依據為在触合巾該至少 特定有限脈衝響顧波!!的侧係數的絕對值的平方的 總和。 1374609 【修正無劃線版】 4. 如申請專利範圍第1項所述之裝置,其中,由串接組 合所對應之訊號品質之依據為在該組合中該至少一個非 特定有限脈衝響應瀘、波器的個別係數的絕對值總和。 5. 如申請專利範圍第i項所述之裝置,其中,由串接組 合所對應之喊^質之依據為在該組合巾駐少一個非 特定有限脈衝響應濾波器的訊號雜訊比。 6. —種可適紐機敎錢的有限脈衝響應滤波器 階數的方法’用以使傳輸訊號品質最佳化,包括下述步驟: 設錢數個特定有限脈衝響應滤波器,各個特定有限 脈衝,縣波器具有個別的固定強度,輸人端接收造成個 f特疋干狀侧較域,並雜_之驗器係數以 β供對應於個·定干擾的補償,扣消除_特定訊號 在該傳輸訊號中造成的個別特定干擾;. β又置至個非特定有限脈衝響賴、波H,該至少-個非特定有限脈衝響顧波器具有固定階數,以及,可適 應f生地串接該複數個特定有限脈衝響應滤波器中的任一 特定有限脈衝響應濾波器; 廄清t接組°產生步驟,將該至少一個非特定有限脈衝響 ff波器與該多個敕有限脈衝響麟波器形成不同串 接組合;及 才估步驟’汗估不同組合所對應之訊一 對應最佳訊號品質之组入#监貝找出 輸訊號相干擾 絲該串接組合驗消除該傳 7.如申凊專利範圍第6項所述之方法其中,該多個有 23 【修正無劃線版】 波器均為嫌脈衝響應遽 絕雜纽麟魏歧__係數的 牛驟φ Γ專纖财6項所述之方法,其中,在該評估 二驟中’由串接組合所對應之訊號品質之依據為在該組合 該至)一個非特定有限脈衝響應滤波器的個別係數的 絕對值總和》 10.如申明專利範圍第6項所述之方法,其中,在該評估 步驟中,由串接組合所對應之訊號品質之依據為在該組合 中該至少-個非特定有限脈衝響顧波獅訊號雜訊比。 11· -種通訊設備’使用如申請專利範圍第丨至5項中任 一項之裝置,用以提供傳輸訊號品質的最佳化。 12. 如申請專利範圍第11項之通訊設備,其中,該通訊 設備係選自網路錢n、網路集線器、網路路由器、網路 閘道器、及祠服器、資料中心組成的群組之一。 13. 如申請專利範圍第12項之通訊設備,其中,該網路 為乙太網路。 Η. —種有限脈衝響應濾波裝置,包含: 一第一有限脈衝響應濾波電路,具有一第一固定階數,用 來濾波一第一訊號; 一第一有限脈衝響應濾殊電路,具有一第二固定階數,用 24 無劃線版】 來濾波一第二訊號,其中該第二訊號不同於該第一訊 號;. 一開關電路,分別耦接該第一與第二有限脈衝響應濾波電 路; 〜非特定有限脈衝響應遽波電路,具有一第三固定階數· '-評估電路,用來依據一或多個參數來評估濾波該第一訊 號之效果以及評估濾波該第二訊號之效果,以產生一或 多個評估結果;以及 一 〜控制電路,用來依據該一或多個評估結果將該非特定有 限脈衝響應濾波電路之Μ階經由該開關電路耦接至該 第一有限脈衝響應濾波電路,並將該非特定有限脈衝響 應濾波電路之Ν階經由該開關電路耦接至該第二有^ 脈衝響應濾波電路,其中該河與該]^為零或正整數。 15.如申請專利範圍第14項所述之裝置,其中該Μ與該 Ν之和等於該第三固定階數。 /、Λ 如申請專利範圍第14項所述之裝置,其中該Μ與該 第一固定階數之和傾於該Ν與該第二固定階數之和。 ^如申μ專利IdgJ第14項所述之裝置,其中該控制電路 該非特定有限脈衝響應濾波電路之階數依序由少至多 ,由多▲至少經由該開關電路㈣至該第—脈衝滤波電 楚i該评估電鹤著財蚊有限脈衝響麟波電路與該 ,衝’緑電接.關之變化來依據該_或多個參 =評估驗該第—職之效果,並產生該-或多個評估 链果。 25 1374609 【修正無劃線版】 18·如申請專利範圍第17項所述之裝置,其中該控制電路 將該非特定有限脈衝響應濾波電路之階數依序遞減或遞· 增地經由該開關電路耦接至該第二脈衝濾波電路,該評估 電路隨著該非特定有限脈衝響應濾波電路與該第二脈衝 濾波電路耦接關係之變化來依據該一或多個參數以評估 濾波該第二訊號之效果,並產生該一或多個評估結果。 19. 如申請專利範圍第14項所述之裝置,其中該一或多個 參數為訊雜比、該第一有限脈衝響應濾波器之各階係數之 運算值、該第二有限脈衝響應濾波器之各階係數之運算值 以及位元錯誤率之至少其中之一。 20. —種乙太網路通訊裝置,具有有限脈衝響應濾波資源 可分配之功能,該乙太網路通訊裝置包含一收發器,該收 發器包含: 一第一有限脈衝響應慮波電路,具有一第一固定階數,用 來濾波一第一訊號; 一第二有限脈衝響應濾波電路,具有一第二固定階數,用 來濾波一第二訊號,其中該第二訊號不同於該第一訊 號; 一開關電路,分別輕接該第一與第二有限脈衝響應遽波電 路; 一非特定有限脈衝響應濾波電路,具有一第三固定階數; 一評估電路’用來依據一或多個參數來評估濾波該第一訊 號之效果以及評估濾波該第二訊號之效果,以產生一或 多個評估結果;以及 26 1374609 【修正無劃線版】 一控制電路’用來依據該一或多個評估結果將該非特定有 限脈衝響應濾波電路之Μ階經由該開關電路輕接至該 第一有限脈衝響應濾波電路,並將該非特定有限脈衝響 應濾波電路之Ν階經由該開關電路耦接至該第二有限 脈衝響應濾波電路,其中該Μ與該Ν為零或正整數。 21.如申請專利範圍第20項所述之裝置,其中該馗盥 Ν之和等於該第三固定階數。 22. 如申請專利範圍第20項所述之裴置,其中該Μ與該 第一固定階數之和不同於該Ν與該第二固定階數之和。 23. 如申請專利範圍第2〇項所述之裝置,其中該控制電路 將該非特定有限脈衝響應濾波電路之階數依序由少至多 或由多至少經由該開關電路耦接至該第一脈衝濾波電 路該砰估電路隨著該非特定有限脈衝響應濾波電路與該 第脈衝濾波電路麵接關係之變化來依據該一或多個參 =評估濾波該第一訊號之效果,並產生該一或多個評估 24. 如申請專利顧第23項所述之裝置,其中該控制電路 定有限脈衝響應遽波電路之階數依序遞減或遞 二及Μ開關電路祕至該第二脈衝濾波電路,該評估 :::著該非特定有限脈衝響應遽波電路與該第 =„係之變化來依據該一或多個參數以評估 叙效果,並產线—或錄評储果。 參數專纖_2G項所収裝置,其愧—或多個 4雜比、該第-有限脈衝響應濾〉皮器之各階係數之 27 1374609 【修正無劃線版】 運算值、該第二有限脈衝響應據波器之各階係數之運算值 以及位元錯誤率之至少其中之一。 26. -種乙太網路通訊裝置,具有有限脈衝響應濾波資源 可分配之功此,該乙太網路通訊裝置包含: 一第一收發器,該第一收發器包含: -第-有限脈衝響麟波電路,具有_第一固定階數, 用來濾波一第一訊號; 一第二收發器,該第二收發器包含: 一第二有限脈衝響應濾波電路,具有一第二固定階數, 用來濾波一第二訊號,其中該第二訊號不同於該第一 訊號; 一開關電路,分別耦接該第一與第二有限脈衝響應濾波 路; ' 一非特定有限脈衝響應濾波電路,具有一第三固定階數; 一評估電路,用來依據一或多個參數來評估濾波該第一訊 唬之效果以及評估濾波該第二訊號之效果,以產生一戋 多個評估結果;以及 一控制電路,用來依據該一或多個評估結果將該非特定有 限脈衝響應濾破電路之Μ階經由該開關電路耦接至該 第一有限脈衝響應濾波電路,並將該非特定有限脈衝響 應遽波電路之Ν階經由該開關電路輕接至該第二有阳 脈衝響應濾波電路,其中該Μ與該Ν為零或正整數。 27·如申請專利範圍第26項所述之裝置,其中該“與該 Ν之和等於該第三固定階數。 28 【修正無劃線版】 28. 如申請專利範圍帛26項所述之裝置,其中該m與該 第-固定階數之和不同於該N與該第二固讀數之和。 29. 如申請專利範圍第26項所述之裳置,其中該控制電路 將該非特定有限脈衝響麟波電路之階數依序由少至多 或由多至少經由該開關電路輕接至該第—脈衝渡波電 路’該評估電路隨著該非特定有限脈衝響應遽波電路盘該 第-脈衝紐電路_ _之變化來依據該—或多個參 數以評估·該第-訊號之效果,並產生該—或多個評估 結果。 3〇·如申请專利範圍第29項所述之裝置,其中該控制電路 將該非特定有限脈衝響應濾波電路之階數依序遞減或遞 增地經由+該開關電路輕接至該第二脈衝滤波電路 ,該評估 電路隨著該非特定有限脈衝響應舰電路與該第二脈衝 濾波電路祕_之變化來依據該—或多個參數以評估 遽波該第二訊號之效果,並產生該—或辣評估結果。 31.如申請專利範圍第26項所述之装置其中該一或多個 參,為訊雜比、該第—有限脈衝響應濾、波器之錢係數之 運算值、該第二有限脈衝響應濾波器之各階係數之運算值 以及位元錯誤率之至少其中之一。 32’如申靖專利範圍第跖項所述之裝置其中該非特定有 限脈衝響應濾波電路射設置於該第—收發器巾或該第 一收發器中、或分佈於該第-及第二收發器中、或獨立設 置於該第H收發器之外。 33.—種乙太網路通訊裝置,具有有限脈衝響應濾波資源 29 1374609 【修正無劃線版】 可分配之功能,該乙太網路通訊裝置包含: -第-收發器’用來接收一第一訊號,該第一收發器包含: 一第一有限脈衝響應濾波電路’用來濾波該第一訊號; 以及 一第一非特定有限脈衝響應濾波電路,具有一第一固定 階數; 一第二收發器’用來接收一第二訊號,該第二收發器包含: -第-有限脈衝響應驗電路m皮該第二訊號; 以及 一第二非特定有限脈衝響應濾波電路,具有一第二固定 階數 -開關電路’分她接該第—與第二非特定有限脈衝響應 濾波電路; -評估電路,用來依據—或多個參絲評倾波該第一訊 號之效果以及評估渡波該第二訊號之效果,以產生一或 多個評估結果;以及 一控制電路,用來依據該一或多個評估結果將該第一及第 二非特定有限脈衝響應濾波電路之階數總和中之M階 經由該開關電路耦接至該第一有限脈衝響應濾波電 路’並將該第-及第二非特定有限脈衝響應遽波電路之 P皆數總和巾之N PW!由該關電路_至該第二有限 脈衝響應遽波電路,其中該Μ與該N為零或正整數。 34.如申請專利範㈣33項所述之裝置,其中該m與該 N之和等於該第一與第二固定階數之和。 ^ 30 1374609 【修正無劃線版】 35.如申請專利範圍第33項所述之裝置,其中該一或多個 參數為訊雜比、該第一有限脈衝響應濾波器之各階係數之 運算值、該第二有限脈衝響應濾波器之各階係數之運算值 以及位元錯誤率之至少其中之一。[Correction-free version] X. Patent application scope: L A device that can adjust the order of the finite impulse response filter for specific purposes to optimize the transmission signal quality, including: a plurality of specific finite impulse responses The filter, each specific finite impulse response chopper has an outline order, and receives the individual fixed-talkers causing the specific interference, and according to the chopper coefficient of _ to provide compensation corresponding to the specific interference For eliminating individual specific interference caused by individual specific signals in the transmission signal; at least one non-specific finite impulse response filter having a fixed order 'adaptively connecting the plurality of specific finite impulse response choppers Any of the specific finite impulse response filters; , the unit, the rib control followed by at least one non-specific finite impulse response chopper and the plurality of shot finite impulse ringers; and 'evaluation Unit, evaluate the signal quality corresponding to different combinations, find a concatenation combination corresponding to the best fiber quality, and prompt the control element selection The series combination of the transmission signal for canceling interference in. 2. The device of claim i, wherein the plurality of finite impulses m are at least one of the finite impulse response choppers. ~ 3 people like the scope of patent application! The device according to the item, wherein the signal quality corresponding to the serial group a is based on the at least a specific limited pulse response wave in the contact towel! The sum of the squares of the absolute values of the side coefficients of !. 1. The apparatus of claim 1, wherein the signal quality corresponding to the serial combination is based on the at least one non-specific finite impulse response in the combination, The sum of the absolute values of the individual coefficients of the wave. 5. The apparatus of claim i, wherein the shunting quality of the tandem combination is based on a signal to noise ratio of a non-specific finite impulse response filter in the composite towel. 6. A method of finite impulse response filter order that can be used to save money. To optimize the quality of the transmitted signal, including the following steps: Set a number of specific finite impulse response filters, each specific limited Pulse, the county waver has an individual fixed intensity, the input end of the input end causes a f-trending dry side compared to the domain, and the detector coefficient of the miscellaneous _ is supplied by β for the compensation of the fixed interference, and the deduction is _specific signal The individual specific interference caused in the transmission signal; β is further set to a non-specific finite impulse response, the wave H, the at least one non-specific finite impulse response waver has a fixed order, and, can adapt to the f Cascading any one of the plurality of specific finite impulse response filters; 廄 t t ° 产生 generating step, the at least one non-specific finite impulse sigma and the plurality of 敕 finite pulses The sounding wave machine forms different series combination; and the evaluation step 'Khan evaluates the corresponding combination of the best signal quality corresponding to the different combinations of the sweat. #监贝 Find the signal phase interference wire The serial connection test eliminates the transmission7. The method according to claim 6 of the patent scope, wherein the plurality of 23 [corrected non-marked version] are all stunned impulse response 遽 纽 纽 纽 纽 纽 纽 纽 纽 纽 _ _ _ 系数 系数 系数 Γ Γ Γ The method described in the sixth item, wherein in the evaluation step 2, the signal quality corresponding to the serial combination is based on the absolute coefficient of the individual coefficient of the non-specific finite impulse response filter. The method of claim 6, wherein, in the evaluating step, the signal quality corresponding to the serial combination is based on the at least one non-specific finite impulse in the combination Gu Boshi signal noise ratio. 11. A type of communication device' uses a device as claimed in any one of claims 5 to 5 to provide an optimization of the quality of the transmitted signal. 12. The communication device of claim 11, wherein the communication device is selected from the group consisting of a network money, a network hub, a network router, a network gateway, a server, and a data center. One of the groups. 13. The communication device of claim 12, wherein the network is an Ethernet network. A finite impulse response filtering device comprising: a first finite impulse response filter circuit having a first fixed order for filtering a first signal; a first finite impulse response filter circuit having a first a second fixed signal is used to filter a second signal, wherein the second signal is different from the first signal; a switching circuit coupled to the first and second finite impulse response filtering circuits respectively a non-specific finite impulse response chopper circuit having a third fixed order · '- evaluation circuit for evaluating the effect of filtering the first signal and evaluating the effect of filtering the second signal based on one or more parameters And generating a one or more evaluation results; and a control circuit for coupling the step of the non-specific finite impulse response filter circuit to the first finite impulse response via the switch circuit according to the one or more evaluation results a filter circuit, and coupling the step of the non-specific finite impulse response filter circuit to the second pulse response filter circuit via the switch circuit, wherein the river With this]^ is zero or a positive integer. 15. The device of claim 14, wherein the sum of the Μ and the 等于 is equal to the third fixed order. The device of claim 14, wherein the sum of the Μ and the first fixed order is greater than the sum of the 固定 and the second fixed order. The device of claim 14, wherein the control circuit has a sequence of the non-specific finite impulse response filter circuit in a sequence of as few as at least ▲ at least via the switch circuit (4) to the first pulse filter Chui. The evaluation of the electric crane and the finite-pulse lining wave circuit with the mosquito, and the change of the green-light connection, according to the _ or more parameters = evaluation of the effect of the first job, and generate the - or Multiple evaluation chains. The apparatus of claim 17, wherein the control circuit sequentially decrements or increments the order of the non-specific finite impulse response filter circuit via the switch circuit. The second circuit is coupled to the second pulse filter circuit, and the evaluation circuit adjusts the relationship between the non-specific finite impulse response filter circuit and the second pulse filter circuit to evaluate the second signal according to the one or more parameters. The effect and the one or more evaluation results are generated. 19. The device of claim 14, wherein the one or more parameters are a signal to noise ratio, an operational value of each order coefficient of the first finite impulse response filter, and the second finite impulse response filter At least one of an operation value of each order coefficient and a bit error rate. 20. An Ethernet communication device having a finite impulse response filtering resource assignable function, the Ethernet communication device comprising a transceiver, the transceiver comprising: a first finite impulse response wave circuit having a first fixed order for filtering a first signal; a second finite impulse response filter circuit having a second fixed order for filtering a second signal, wherein the second signal is different from the first a switching circuit for respectively connecting the first and second finite impulse response chopper circuits; a non-specific finite impulse response filtering circuit having a third fixed order; an evaluation circuit 'for one or more Parameter to evaluate the effect of filtering the first signal and evaluate the effect of filtering the second signal to generate one or more evaluation results; and 26 1374609 [corrected non-linear version] a control circuit is used to determine one or more The evaluation result is that the step of the non-specific finite impulse response filter circuit is lightly connected to the first finite impulse response filter circuit via the switch circuit, and the Ν specific finite impulse response filter stage circuits through the switching circuit coupled to said second finite impulse response filter circuit, wherein the Μ is zero or a positive integer with the Ν. 21. The device of claim 20, wherein the sum of the 馗盥 is equal to the third fixed order. 22. The device of claim 20, wherein the sum of the Μ and the first fixed order is different from the sum of the 固定 and the second fixed order. 23. The device of claim 2, wherein the control circuit couples the order of the non-specific finite impulse response filter circuit to at least one or at least via the switch circuit to the first pulse. The filter circuit estimates the effect of filtering the first signal according to the one or more parameters along with the change of the interface relationship between the non-specific finite impulse response filter circuit and the first pulse filter circuit, and generates the one or more The apparatus of claim 23, wherein the control circuit determines that the order of the finite impulse response chopper circuit is sequentially decreasing or the second and the switching circuit are secret to the second pulse filtering circuit. Evaluation::: The non-specific finite impulse response chopper circuit and the change of the system are based on the one or more parameters to evaluate the effect, and the production line - or record the fruit. Parameter special fiber _2G item The received device, the 愧- or a plurality of 4-to-multiple ratios, the first-order finite impulse response filter, the various order coefficients of the 27 1374609 [corrected non-linear version] operation value, the second finite impulse response data device At least one of the operational value of the order coefficient and the bit error rate. 26. An Ethernet communication device having a finite impulse response filtering resource assignable, the Ethernet communication device includes: a transceiver, the first transceiver comprising: a first-finite impulse ring-splitting circuit having a first fixed order for filtering a first signal and a second transceiver comprising: a second finite impulse response filter circuit having a second fixed order for filtering a second signal, wherein the second signal is different from the first signal; a switching circuit coupled to the first and second a finite impulse response filter circuit; 'a non-specific finite impulse response filter circuit having a third fixed order; an evaluation circuit for evaluating the effect of filtering the first signal and evaluating the filter based on one or more parameters The effect of the second signal to generate a plurality of evaluation results; and a control circuit for filtering the non-specific finite impulse response to the level of the circuit based on the one or more evaluation results The first finite impulse response filter circuit is coupled to the first finite impulse response filter circuit, and the step of the non-specific finite impulse response chopper circuit is lightly connected to the second positive pulse response filter circuit via the switch circuit, wherein the Μ The device is a zero or a positive integer. The device of claim 26, wherein the sum of the sum is equal to the third fixed order. [Amendment to the unlined version] 28. The device of claim 26, wherein the sum of the m and the first fixed order is different from the sum of the N and the second solid reading. 29. The device of claim 26, wherein the control circuit sequentially steps the order of the non-specific finite impulse ringing circuit to the first step by at least or at least via the switching circuit. a pulse crossing circuit 'the evaluation circuit according to the change of the non-specific finite impulse response chopper circuit board the first-pulse circuit __ according to the - or a plurality of parameters to evaluate the effect of the first-signal, and generate the - Or multiple evaluation results. 3. The device of claim 29, wherein the control circuit lightly or sequentially increments the order of the non-specific finite impulse response filter circuit via the + switch circuit to the second pulse filter circuit The evaluation circuit is responsive to the change of the non-specific finite impulse response ship circuit and the second pulse filter circuit to evaluate the effect of chopping the second signal and generate the - or spicy evaluation result. 31. The device of claim 26, wherein the one or more parameters are a signal to noise ratio, the first finite impulse response filter, an operation value of a money coefficient of the wave device, and the second finite impulse response filter. At least one of an operation value of each order coefficient and a bit error rate. 32. The device of claim 3, wherein the non-specific finite impulse response filter circuit is disposed in the first transceiver or the first transceiver, or distributed in the first and second transceivers Medium, or independently set outside the H-th transceiver. 33.-Ethernet communication device with finite impulse response filtering resources 29 1374609 [Corrected no-dash version] Assignable function, the Ethernet communication device includes: - the first transceiver is used to receive one a first signal, the first transceiver includes: a first finite impulse response filter circuit 'for filtering the first signal; and a first non-specific finite impulse response filter circuit having a first fixed order; The second transceiver is configured to receive a second signal, the second transceiver includes: - a first-finite impulse response circuit, the second signal; and a second non-specific finite impulse response filter circuit having a second a fixed order-switch circuit 'dividing the first-and second non-specific finite impulse response filter circuit; - an evaluation circuit for evaluating the effect of the first signal and evaluating the wave according to - or a plurality of reference wires An effect of the second signal to generate one or more evaluation results; and a control circuit for reacting the first and second non-specific finite impulse responses based on the one or more evaluation results The Mth order of the sum of the order of the filter circuit is coupled to the first finite impulse response filter circuit via the switch circuit and combines the P sum of the first and second non-specific finite impulse response chopper circuits PW! from the off circuit _ to the second finite impulse response chopper circuit, wherein the Μ and the N are zero or a positive integer. 34. The apparatus of claim 23, wherein the sum of m and the N is equal to a sum of the first and second fixed orders. The apparatus of claim 33, wherein the one or more parameters are a signal-to-noise ratio, and an operation value of each order coefficient of the first finite impulse response filter. And an operation value of each order coefficient of the second finite impulse response filter and at least one of a bit error rate. 31 1374609 【修正無劃線版】 七、 指定代表圖: (一) 、本案代表圖為:第3圖 (二) 、本案代表圖之元件代表符號簡單說明: 300可適應性調整有限脈衝響應濾波器階數的裝置 302 評估單元 304控制單元 3061-306k 特定 FIR 3081-308j 多工器 3101-310j 非特定 FIR 八、 ;^案若有化學式時,請揭示最能顯示發明特徵的化 學式: 無031 1374609 [Revised unlined version] VII. Designated representative map: (1) The representative figure of this case is: Figure 3 (2), the representative symbol of the representative figure in this case is a simple description: 300 adaptability adjustment finite impulse response filtering Device 302 evaluation unit 304 control unit 3061-306k specific FIR 3081-308j multiplexer 3101-310j non-specific FIR VIII, ^ ^ If there is a chemical formula, please reveal the chemical formula that best shows the characteristics of the invention: no 0
TW97135008A 2008-02-04 2008-09-12 Device and method for adaptive adjustment of length of finite impulse response filter TWI374609B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US12/322,503 US8285772B2 (en) 2008-02-04 2009-02-03 Order adaptive finite impulse response filter and operating method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US2609908P 2008-02-04 2008-02-04

Publications (2)

Publication Number Publication Date
TW200935736A TW200935736A (en) 2009-08-16
TWI374609B true TWI374609B (en) 2012-10-11

Family

ID=40977359

Family Applications (1)

Application Number Title Priority Date Filing Date
TW97135008A TWI374609B (en) 2008-02-04 2008-09-12 Device and method for adaptive adjustment of length of finite impulse response filter

Country Status (2)

Country Link
CN (1) CN101505287B (en)
TW (1) TWI374609B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12088325B2 (en) 2021-12-29 2024-09-10 Dynami Vision Ltd. Filter system and operation method thereof

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI472151B (en) * 2011-08-30 2015-02-01 Richwave Technology Corp Filter system capable of automatically adjusting bandwidth and method of automatically adjusting bandwidth of a filter
US9525453B2 (en) * 2014-06-10 2016-12-20 Apple Inc. Intermodulation cancellation of third-order distortion in an FDD receiver
CN114928349B (en) * 2022-06-27 2024-02-27 奉加微电子(昆山)有限公司 Continuous-time pipelined analog-to-digital converter and its digital reconstruction filter
CN115425998B (en) * 2022-11-08 2023-02-10 长沙驰芯半导体科技有限公司 Multi-order multi-frequency-band anti-interference intelligent ultra-wideband antenna self-adaption method

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1655468B (en) * 2004-02-09 2010-05-05 瑞昱半导体股份有限公司 Adjustable echo eliminator for full duplex communication system
US8014516B2 (en) * 2006-01-27 2011-09-06 Mediatek Inc. Method and apparatus for echo cancellation
JPWO2007102611A1 (en) * 2006-03-07 2009-07-23 Nsc株式会社 Interpolation function generation circuit

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12088325B2 (en) 2021-12-29 2024-09-10 Dynami Vision Ltd. Filter system and operation method thereof

Also Published As

Publication number Publication date
CN101505287A (en) 2009-08-12
TW200935736A (en) 2009-08-16
CN101505287B (en) 2012-03-21

Similar Documents

Publication Publication Date Title
US7164764B2 (en) Method and apparatus for precode crosstalk mitigation
JP5064877B2 (en) Methods and apparatus for equalization and crosstalk mitigation
US8296347B2 (en) Use of line characterization to configure physical layered devices
US6160790A (en) Crosstalk canceller system and method
TWI374609B (en) Device and method for adaptive adjustment of length of finite impulse response filter
KR101251897B1 (en) Joint signal processing across a plurality of line termination cards
US8804794B2 (en) Adjustable latency transceiver processing
US6912208B2 (en) Method and apparatus for equalization and crosstalk mitigation
JP4866979B2 (en) Separation of transmitted and received signals
TW201032572A (en) Communication apparatus with echo cancellations and method thereof
JP4614829B2 (en) Communication apparatus and communication method
KR101687929B1 (en) Gain adaptation for downstream vectoring systems
US8995504B2 (en) Methods and apparatuses for data transmission
US8285772B2 (en) Order adaptive finite impulse response filter and operating method thereof
JP2011199895A (en) Multiple channel interference cancellation
US8345583B2 (en) Method for enhancing ethernet channel impairment and apparatus using the same
KR100475771B1 (en) Device and method for echo compensation in a two-wire full duplex channel transmission method
WO2017070951A1 (en) Echo cancellation-related circuit and method
JP4796051B2 (en) Sub-block domain transform multiplex signal processing
CN109644019B (en) Method and Apparatus for Data Transmission Through Metal Wire Pairs
JP3653045B2 (en) Receiving method and receiver for full-duplex transmission system
JP4463747B2 (en) Interference signal reduction circuit
JPH04239825A (en) Echo canceller