TWI291817B - Method and system for residual frequency offset compensation of multi-carrier communication - Google Patents

Method and system for residual frequency offset compensation of multi-carrier communication Download PDF

Info

Publication number
TWI291817B
TWI291817B TW94124124A TW94124124A TWI291817B TW I291817 B TWI291817 B TW I291817B TW 94124124 A TW94124124 A TW 94124124A TW 94124124 A TW94124124 A TW 94124124A TW I291817 B TWI291817 B TW I291817B
Authority
TW
Taiwan
Prior art keywords
carrier
residual frequency
phase change
residual
frequency
Prior art date
Application number
TW94124124A
Other languages
Chinese (zh)
Other versions
TW200703989A (en
Inventor
Wen-Sheng Hou
Original Assignee
Silicon Integrated Sys 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 Silicon Integrated Sys Corp filed Critical Silicon Integrated Sys Corp
Priority to TW94124124A priority Critical patent/TWI291817B/en
Publication of TW200703989A publication Critical patent/TW200703989A/en
Application granted granted Critical
Publication of TWI291817B publication Critical patent/TWI291817B/en

Links

Landscapes

  • Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
  • Digital Transmission Methods That Use Modulated Carrier Waves (AREA)

Abstract

A residual frequency offset (RFO) compensation method and a compensation system of a multi-carrier communication system is disclosed. The compensation method of present invention includes two phase-computation steps. The first phase-computation step processes a phase variation of an OFDM data symbol each time to obtain an RFO estimator for compensating a demodulation carrier. The second-phase computation processes a plurality of OFDM data symbols each time to obtain an RFO estimator for compensating the demodulation carrier. The compensation system of the present invention includes: a Discrete Fourier Transformation (DFT) unit for performing DFT on receiving OFDM packages; a phase variation detector connecting to the DFT unit for detecting phase variations of a plurality of OFDM data symbols; and a residual frequency offset compensator connecting to the phase variation detector, which generates the desired RFO estimators in accordance with the phase variations of the plurality of OFDM data symbols and the compensation method of the present invention.

Description

1291817 九、發明說明: 【發明所屬技術領域】 本發明係有關於一種多載波通訊系統,特別是有關於正交分頻多工通訊 系統的殘餘頻率補償的方法與系統。 【先前技術】 由於多載波通訊提供較大傳輸頻寬,目前許多應用,如第3代行動通 訊、非對稱稱數位迴路(ADSL)等皆採用多載波通訊技術實施。其中,正交 分頻多工(Orthogonal Frequency Division Multiplexing,OFDM)係一種 熟知的通訊祕多工技術。比起單純的分頻多卫技術(FDM),麵通訊系 統具有更佳的頻寬糊率,更為第四代行動通賴採用。QFDM通訊系統應 用-組正㈣鮮財載資料,經反離散制葉轉換(InverseBACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multi-carrier communication system, and more particularly to a method and system for residual frequency compensation of an orthogonal frequency division multiplexing communication system. [Prior Art] Since multi-carrier communication provides a large transmission bandwidth, many applications such as third-generation mobile communication and asymmetric digital-to-digital (ADSL) are implemented by multi-carrier communication technology. Among them, Orthogonal Frequency Division Multiplexing (OFDM) is a well-known communication multiplex technology. Compared to pure crossover multi-technology (FDM), the face-to-face communication system has a better bandwidth and is more versed in the fourth generation of operations. QFDM communication system application - group positive (four) fresh financial data, reverse discrete leaf conversion (Inverse

DiscreteDiscrete

Fourier Transform,鹏)以及平行/串列轉換 Conversion)將頻域上的資料轉換為時域上複數個IDFT的點,再由一振盪 器產生之載波關變至所傳輸賴斜。於接收端,本地振妓產生相同 的載波對接收《觸,再經雜富利雜換(DFT)縣各辭上的資料。 由於接收端解調變載波與傳送端調變載波係以不同振盪器產生,所以 頻率不會綱,進而在接收端產生—頻率位移。—般*言,接收端藉由傳 送封包的别置序列(preamble)來估測頻率偏移量以進行頻率補償。然而, 不論是傳送端或接收端,振盪H產生的载波於前置相時若並未達到一穩 定值,會麟於前置相估朗鮮偏移量與實際上資料部分的頻率偏移 1291817 量不同’而殘留-個殘存偏移量(ResidUal Frequency 〇ffset,_。於 接收端,RFO 會引起(inter—furrier Tn+A f 、 w m:er carrier Interference,ICI)效應,而提高了 封包錯誤率。Fourier Transform (Peng) and Parallel/Parallel Conversion (Conversion) converts the data in the frequency domain into a plurality of IDFT points in the time domain, and then the carrier generated by an oscillator is turned off to the transmitted skew. At the receiving end, the local oscillator generates the same carrier pair to receive the data of the touch, and then the DFT county. Since the receiving end demodulation variable carrier and the transmitting end modulation carrier system are generated by different oscillators, the frequency does not occur, and thus the frequency shift occurs at the receiving end. As a general rule, the receiving end estimates the frequency offset for frequency compensation by transmitting a preamble of the packet. However, whether it is the transmitting end or the receiving end, if the carrier generated by the oscillation H does not reach a stable value in the pre-phase, the frequency offset between the pre-phase estimation and the actual data portion is 1291817. The amount is different, and the residual-residual offset (ResidUal Frequency 〇ffset, _. At the receiving end, the RFO causes (inter-furrier Tn+A f , wm: er carrier Interference, ICI) effect, and improves the packet error. rate.

第一圖中顯示一 _通訊系統接收端架構。F為本地缝器100產生 的載波頻率’ F1為由前置序列估測得到的頻率位移量。對於講Μ封包的 資料Ρ刀開始本地振盈益⑽的載波頻率F經頻率位移量η補償後 對OFDM資料符號(_ data symb〇1)解調變,之後經由離散富利葉轉換 單元1〇2作離散富利葉轉雜紐te F〇urier Transf〇rm,㈣以還原 資料訊號。附運算結果除了將送至後端運算單元進行資料處理運算外, 亦輸出至相位變化偵測器⑽,藉由於_資料符號中安插的導音信號 (Pilot tone),以偵測〇簡資料符號的相位變化。麵資料符號的相位 變化侧值輯送至後端運算單元伽雜的處理,另外,亦會輸入殘餘The first figure shows a communication system receiving end architecture. F is the carrier frequency generated by the local stitcher 100' F1 is the amount of frequency shift estimated from the preamble sequence. For the data of the packet, the carrier frequency F of the local vibration gain (10) is compensated by the frequency shift amount η, and then the OFDM data symbol (_data symb〇1) is demodulated, and then passed through the discrete Fourier transform unit. 2 for the discrete Fu Liye to turn the new te F Furier Transf〇rm, (d) to restore the data signal. In addition to the data processing operation sent to the back-end computing unit, the result of the operation is also output to the phase change detector (10), which detects the simplification data symbol by the Pilot tone inserted in the _ data symbol. Phase change. The phase change side value of the surface data symbol is sent to the processing of the back-end operation unit gamma, and the residual is also input.

鮮估廳薦作運算,進而得到_估測值ρ2。之後^回麵償解 調變載波F,而解調變載波由_開始的F + n變為f + fi + f2。 第二圖顯示在解調變載波保持不變下,_資料符號其相位變化的關 係。若排除雜訊的影響,基本上各點可以連成—直線(如圖式虛線)。其中, 直線的斜率㈣RFQ 計值。為方便起見,吾人以英文字母了表示兩連 續麵資料符號的時間間隔,以崎雨n個觀資料符號偵測得 到的相位變化值。若以η=〇為美,目,丨士穿 y ^ 絲雜則“ n個_資料符號的相位 變化可計算得到RF0的估測值Μ為 F2 = ^lThe Estimation Hall recommends the calculation, and then the _estimated value ρ2. Afterwards, the modulated carrier F is compensated, and the demodulated variable carrier is changed from F + n at the beginning of _ to f + fi + f2. The second graph shows the relationship of the phase change of the _ data symbol while the demodulation variable carrier remains unchanged. If the influence of noise is excluded, basically, each point can be connected into a straight line (as shown by the dotted line). Among them, the slope of the line (four) RFQ value. For the sake of convenience, we have used English letters to represent the time interval between two consecutive data symbols, and the phase change values detected by the n-view data symbols. If η = 〇 is beautiful, the eye, the gentleman wears y ^ 丝 杂 "" n _ data symbol phase change can be calculated to get the estimated value of RF0 F F2 = ^l

nxT 7 1291817 或是,第i、m個OFDM資料符號之相位變化估測值分別心⑴與θα), 則RF0的估測值F2應為 _ θ{τη)-θ(ί)nxT 7 1291817 Alternatively, the phase change estimates of the i-th and m-th OFDM data symbols are respectively (1) and θα), and the estimated value F2 of RF0 should be _ θ{τη)-θ(ί)

~{m-i)xT Π的大小(或是第i、m個⑽Mf料符號的間隔)影響_估測的準確度。 當η越大,偵測的相位變化值的誤差對_影響愈小(因為分母變大),故 而η愈大愈具抗雜訊能力。然而,η越大就越遲完成_的估測補償。例 ,如η為5,即利用第5個0FDM f料符號的相位變化值計算以獲得㈣估測 值進行載波頻率補償校正,則第卜5個(_估測值回授至前端補償最快也 是第6個OFDM資料符號才能補償)資料符號皆無法受益於_補償,其符 號錯誤率亦無法有效下降。若取n為2,則僅第卜2個觀資料符號未 麵率補償,然而,φ於其時間間酿短,第2個觀f料符號相位變 化的伽#差對於RFC)估啦影響越大。亦即,n值越小其抗雜訊能力越 低,得到的RF0估測值亦較不準確,反而可能補償的結果更糟。 • 綜合上述,如何於接收端迅速、精確的估測補償RF〇以降低Ici效應 實為OFDM通訊系統之重要議題。 【發明内容】 本發明的目的之一,在於提供一種應用於OFDM通訊系統之殘餘頻率補 償方法’使得接收端OFDM封包於前置序列之後的資料符號能儘早進行RF〇 補償,以降低ICI的影響。 本發明之另一目的在於提供一種應用於OFDM通訊系統之殘餘頻率補償 1291817 方法,使触端_«個_資卿餅迅賴舰,更進-步精確 地调整㈣估測值以補償解調變載波。 根據上述,本發明提出一種兩階段運算的殘餘頻率補償方法,分為兩 階段運算處理-_訊號封包,包括:在第—階段運算,每次只對目前_ =符號之__處理’進___聽,_償解調 夂載波’収在弟—階段運算,每次對複數個觀㈣符號作運算以得 到對應之殘細雜顺,進_償軸變載波。 【實施方式】 本發明的-些實施方式會詳細描述如下。然而,除了詳細描述的内容 外,本制還可以廣泛地在其他的實施例施行1本發_顧不受限定, 其以之後的申請專利範圍為準。 本發明揭露-種兩階段運算_餘_償方法,以進行勝的估測 •觸償。區分兩階段進行_估測補償係基於不同的需求考量,其中於第 iw又為避免〇_减封包巾,w端的㈣M資料符號目⑹效應影響而 造成封包錯誤率上升,其考量的重點在於補償的速度。而於第二階段係 為了進-步消除微小之鮮偏移量對接收端的影響,其考量重點在補償的 精確度。 假設習知技術於第k個〇_資料符號作第—次_補償,由前述得知, 第1至第k個喔資料符號細未補償喊波解調,其受⑹效應的影 響比第k個之_麵熟符練大。本發.㈣第—階段運算,在 9 1291817 # k個OFDM資料符號之前亦能做RFO的補償,且第k個麵f料符號後 得到與習知技術相同的RF0估測結果。 第三圖顯示應用本發明第-P皆段RF0估測運算,㈣M資料符號與其相 位變化值之關係。若不作RF0補償而始終維持以起始解調變載波⑻⑴解 凋’則於第j、k個OFDM貧料符號得到的相位估測值應為0 ‘⑴、0 ‘⑴。 .然而’本例中第i、j、k個0FM資料符號後各作一次_補償,使解調 魯變載波頻率改變’因而Θ⑷與η間不再是一條直線,而於每次腦補償 後斜率改變。當斜料〇時表科再有_,鱗傳送/接收端的載波頻率 相同。 於刖置序列結束至第i㈣_資料符號為止,解調變載波的頻率為 F例。第i個0FDM資料符號的相位估測值為0⑴,故而卿估測值 為 一 F2 i =道The size of ~{m-i)xT ( (or the interval between the i and m (10) Mf symbols) affects the accuracy of the estimation. When η is larger, the error of the detected phase change value has a smaller influence on _ (because the denominator becomes larger), so the larger the η, the more anti-noise ability. However, the larger η, the later the estimated compensation of _ is completed. For example, if η is 5, that is, the phase change value of the 5th 0FDM f symbol is used to obtain (4) the estimated value for carrier frequency compensation correction, then the 5th (the estimated value feedback to the front end compensation is the fastest) It is also the sixth OFDM data symbol to compensate.) Data symbols cannot benefit from _ compensation, and the symbol error rate cannot be effectively reduced. If n is 2, then only the 2nd data symbols are not face-rate compensated. However, φ is short-lived during its time, and the influence of the gamma-difference of the phase change of the second observation symbol on the RFC is estimated. Big. That is, the smaller the value of n is, the lower the anti-noise ability is, and the obtained RF0 estimation value is also less accurate, but the compensation result may be worse. • In summary, how to quickly and accurately estimate the compensation RF〇 at the receiving end to reduce the Ici effect is an important issue for OFDM communication systems. SUMMARY OF THE INVENTION One object of the present invention is to provide a residual frequency compensation method applied to an OFDM communication system, so that a data symbol after receiving a OFDM packet in a preamble sequence can perform RF〇 compensation as early as possible to reduce the influence of ICI. . Another object of the present invention is to provide a residual frequency compensation 1291817 method applied to an OFDM communication system, so that the touch end _ « _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Variable carrier. According to the above, the present invention provides a two-stage operation residual frequency compensation method, which is divided into two-stage operation processing--signal packet, including: in the first stage operation, each time only the current _ = symbol __ processing 'into _ __ listen, _ rectify 夂 carrier 'received in the brother-stage operation, each time to operate on a plurality of (four) symbols to obtain the corresponding residuals, and to compensate the axis to change the carrier. [Embodiment] Some embodiments of the present invention will be described in detail below. However, in addition to the detailed description, the present invention can also be widely practiced in other embodiments without limitation, which is subject to the scope of the subsequent patent application. The invention discloses a two-stage operation _ residual _ compensation method for performing the estimation of the win. The two-stage differentiation _ estimation compensation system is based on different demand considerations, in which the iw is also used to avoid the 〇 _ minus the packet towel, and the w (4) M data symbol (6) effect affects the packet error rate, and the focus of the consideration is compensation. speed. In the second stage, in order to eliminate the influence of the small fresh offset on the receiving end, the consideration is focused on the accuracy of the compensation. Assuming that the prior art is the first-time compensation for the kth 〇_ data symbol, it is known from the foregoing that the first to the kth 喔 data symbols are not compensated for the shuffling demodulation, which is affected by the (6) effect than the kth The _ _ face familiar with the practice. (4) The first-stage operation can also perform RFO compensation before 9 1291817 # k OFDM data symbols, and the k-th face f symbol can obtain the same RF0 estimation result as the prior art. The third figure shows the application of the first-P-segment RF0 estimation operation of the present invention, and (4) the relationship between the M data symbols and their phase change values. If the initial demodulation variable carrier (8)(1) is cancelled without RF0 compensation, the phase estimation value obtained from the jth and kth OFDM poor symbols should be 0 ‘(1), 0 ‘(1). However, in this example, the i, j, and k 0FM data symbols are each compensated once, so that the demodulated Lun carrier frequency changes 'so that Θ(4) and η are no longer a straight line, and after each brain compensation The slope changes. When the slanting material is smashed, the table has another _, and the carrier frequency of the scale transmission/reception end is the same. The frequency of the demodulated variable carrier is F in the end of the sequence until the i-th (fourth)_data symbol. The phase estimate of the i-th OFDM data symbol is 0 (1), so the estimated value is one F2 i = dao

—ixT ® 之後回授補償解調變載波為F+FHF2J。 第j個OFDM資料符號的rfo估測值F2J為 F2_j=^mAfter ixT ® , the compensation demodulation variable carrier is F+FHF2J. The rfo estimated value F2J of the jth OFDM data symbol is F2_j=^m

一 jxT 如前述,於第i個0FDM資料符號後已進行RF〇的補償,此時估測得到 的相位Θ⑴與原本解調變載波為簡時的相位估測值不會相 Θ0)與θ‘⑴的偏移係由前* _估測值F2—i影響所致,將F2」造成 的偏移加回去則可得到原本的相位偏移0‘(j)。其中,由於F2—1景缚第叫 1291817 至第j個GFDM資料符號,其影響的偏移量為(F2」)x(j-i)T,所以<9‘(j) 為 〜(j)= + (F2 一 i)x(j-i)T 而新的RFO的值F2 j為 F2 / = ^ 〇{j) + (F2 __ i)x (j ^ j)T _ 9<^j) + ^ x(<j ^i)T JXT jxT - jxT然後回射2」’補償解㈣載波為剛+F2」·。 再來,於第k個OFDM資料符號後作rf〇補償。同理,F2J影響第j+1 至第k個0FDM資料符號,加上先前F2一i影響第i+1至第j個〇FDM資料 符號所這成的偏移,故而實際取得的相位偵測值與0‘(k)的關係應 θ〇〇 + (F2一i)x(H)T+(F2J)x(k—j)T 而新的RFO的值F2—k為 F2 k =隱一kxT ^k^r =As described above, the RF〇 compensation has been performed after the i-th OFDM data symbol, and the estimated phase Θ(1) at this time is not related to the phase estimation value of the original demodulated variable carrier as a simple 0) and θ' The offset of (1) is caused by the influence of the former * _ estimated value F2 - i, and the offset caused by F2" is added back to obtain the original phase offset 0' (j). Among them, because the F2-1 is called 1291817 to the jth GFDM data symbol, the offset of the influence is (F2))x(ji)T, so <9'(j) is ~(j)= + (F2 - i)x(ji)T and the value of the new RFO F2 j is F2 / = ^ 〇{j) + (F2 __ i)x (j ^ j)T _ 9<^j) + ^ x (<j ^i)T JXT jxT - jxT then backfire 2"' compensation solution (four) carrier is just +F2"·. Then, rf〇 compensation is performed after the kth OFDM data symbol. Similarly, F2J affects the j+1th to kth 0FDM data symbols, plus the previous F2-i affects the offset of the i+1th to jth FDM data symbols, so the actual phase detection is obtained. The relationship between the value and 0'(k) should be θ 〇〇 + (F2 - i) x (H) T + (F2J) x (k - j) T and the value of the new RFO F2 - k is F2 k = hidden kxT ^k^r =

-j)T-j)T

η/ΐ\ θ{ί) Θ(/)Η / — AT7 咐)+ #心· — 〇Γ + (—— -------j 乂Tη/ΐ\ θ{ί) Θ(/)Η / — AT7 咐)+ #心· — 〇Γ + (—— -------j 乂T

)x(k-j)T)x(k-j)T

kxT 一以上得到的刪估計值F2—k與仍以載波頻率剛解調、而於料個麵 資料符號才作第一次RF〇估測的值相同。The depreciated value F2_k obtained by kxT or more is the same as the value that is still demodulated at the carrier frequency and the first RF estimator is estimated from the material data.

整理後的演算法流程如下。 1·偵測第i個OFDM資料符號的相位 F2:隨 ixT 變化0⑴, 取得RF0估測值F2The finished algorithm flow is as follows. 1. Detecting the phase of the i-th OFDM data symbol F2: changing 0(1) with ixT, obtaining the RF0 estimated value F2

Cs) 11 1291817 2.回授F2補償解調變載波 3. _第j個0_賴符號的她變切(】·),取得_估測值打 「θ'(7·) θ⑺沁-ΟΓCs) 11 1291817 2. Feedback F2 compensation demodulation variable carrier 3. _ j-th 0_赖 symbol of her chopping (]·), get _ estimated value hit “θ'(7·) θ(7)沁-ΟΓ

jxT jxTjxT jxT

[•回授F2補償解調變載波。 5. _第k個OFDM資料符號的相位變化⑽),取得_估測值打 . ’(7)+学 dr F2 = kxT 6·回授F2補償解調變載波[• Feedback F2 compensation demodulation variable carrier. 5. _ k-th OFDM data symbol phase change (10)), get _ estimated value hit . '(7) + learn dr F2 = kxT 6 · feedback F2 compensation demodulation variable carrier

Τχψ )x(k- j)TΤχψ )x(k- j)T

kxT 在此必須強調,第-p皆段運异並不以實施例中的三次補償為限。不過, 補償的次數越多,要較多記憶體_存先前複數個腦估測值、或各 個OFDM :身料符號的相位變化值,同時相關演算與_估測器會變得更複 *隹車乂佳實W例中,係取前3個OFDM資料符號作第一階段的RF〇估測 補偾亦即i,j,k分別為1,2, 3。如此一來,藉由本發明第一階段的方法, 0FM訊號封包中前幾個0簡資料符號可及時獲得RF〇補償而降低⑹效 應。 再者,雖然實施例中係利用前幾次肝〇補償時偵測的相位變化值運算 取付本次的RFO估測值,然亦可藉每次回授之RF〇值作運算而得到,本發 明並不限制。 第—階段運算完成後則進入第二階段運算直到一 〇FDM封包結束。第二 P白丰又係為消除微小之殘餘頻率偏移,故而選取的OFDM資料符號間隔較大。kxT It must be emphasized here that the first-p phase difference is not limited to the three compensations in the embodiment. However, the more the number of compensations, the more memory _ the previous multiple brain estimates, or the phase change values of the individual OFDM: body symbols, and the related calculations and _ estimators will become more complex. In the case of the car 乂佳实 W, the first three OFDM data symbols are taken as the first stage of the RF 〇 estimation, ie i, j, k are 1, 2, 3 respectively. In this way, by the method of the first stage of the present invention, the first few zero data symbols in the 0FM signal packet can obtain the RF〇 compensation in time and reduce the (6) effect. Furthermore, although the embodiment uses the phase change value detected during the previous liver hepatic compensation calculation to calculate the RFO estimate of the current time, it can also be obtained by calculating the RF threshold value of each feedback, the present invention. Not limited. After the first stage operation is completed, the second stage operation is entered until the end of a FDM packet. The second P Baifeng is also to eliminate the small residual frequency offset, so the selected OFDM data symbol interval is larger.

Cs) 12 1291817 ' 第四圖顯不第二階段運算之一實施例(假設第η個OFDM估到RFO值至n+1 個OFDM的中途才補償)。如圖式,第二階段運算每隔d個〇FM資料符號 才做-次’補償,係藉由兩她目麵資料符號之她差除以兩者 之時間間隔以取得_估測值。得到的·估計值與先前的腦值相加, 可進一步修正補償接收端的解調變載波。 若兩者之相位差為 Φ=0(η+ά)~0(η) • 故而修正補償RF0估計值為Cs) 12 1291817 'The fourth picture shows an embodiment of the second stage operation (assuming that the nth OFDM estimates the RFO value to be compensated midway through n+1 OFDM). As shown in the figure, the second-stage operation performs the -times compensation every d 〇FM data symbols, and divides the difference between the two her data symbols by the time interval between the two to obtain the _estimated value. The obtained estimated value is added to the previous brain value, and the demodulated variable carrier at the receiving end can be further corrected. If the phase difference between the two is Φ=0(η+ά)~0(η) • Therefore, the estimated compensation RF0 is

dxT 得到的RF0估計值與第-階段的RF〇值相加,可進一步修正補償接收端的 解調變載波。另外’亦可以-較大的d值如d,(d,大於d)作為分母代入上 式’以獲得較佳的抗雜訊能力。 必須注意的是,本發明對d值之大小、是否為固定之常數並不限定, * 一般而言d值愈大受雜訊影響的程度愈低,RF0估測值亦會較精確。再者, 上述相位變化差Φ更可以d個OFDM資料符號相位變化值線性回歸線為準 (例如’以最小平方法(LMS)取得的直線),以取得相位變化差。 需注意的是,由於第一階段運算已先行補償過解調變裁波,基本上第 二階段運算時相位變化值應如第四圖,其增減的幅度相對於第一階段運算 應較為平緩(0 (n+d)« 0 (n)),然而,相位計算,限制了丨0 (η) | < ,所 以即使 0(n+d)«0(n),有可能變成 0(n+d)«0(n) +2 。"1,ί e {-1,0,1}, ⑧ 13 1291817 所以必須有個方法移除2疋i。The RF0 estimate obtained by dxT is added to the RF-threshold of the first stage, and the demodulation variable carrier at the receiving end can be further corrected. Alternatively, a larger d value such as d, (d, greater than d) is substituted as a denominator to obtain a better anti-noise capability. It should be noted that the present invention is not limited to the magnitude of the d value and whether it is a fixed constant. * Generally, the larger the d value is, the lower the degree of noise is affected, and the RF0 estimation value is also more accurate. Furthermore, the phase change difference Φ may be based on a linear regression line of d OFDM data symbol phase change values (e.g., a straight line obtained by a least squares method (LMS)) to obtain a phase change difference. It should be noted that since the first stage operation has been compensated for the demodulation variable cutting wave first, basically the phase change value in the second stage operation should be as shown in the fourth picture, and the increase and decrease amplitude should be relatively flat compared to the first stage operation. (0 (n+d)« 0 (n)), however, the phase calculation limits 丨0 (η) | < , so even if 0(n+d)«0(n), it may become 0(n) +d)«0(n) +2 . "1, ί e {-1,0,1}, 8 13 1291817 So there must be a way to remove 2疋i.

於冗決定將Φ值更減去或加上2;r Φ4(ηΜ)-0(n)一 2;ri,i 當丨Θ (n+d)- 0 (η) I大於π時 對上述情形,吾人計算cD_f根射㈣胸⑷的U否大於/ 。亦即: 為〜1,0或1 i=sign{0(n+d)-6>(n)} 否則i= 0In the redundancy decision, the Φ value is further subtracted or added by 2; r Φ4(ηΜ)-0(n)−2; ri,i when 丨Θ(n+d)− 0 (η) I is greater than π , we calculate cD_f root shot (four) chest (4) U is greater than /. That is: ~1,0 or 1 i=sign{0(n+d)-6>(n)} otherwise i= 0

綜合以上實施例中對第-階段、第二階端運算的描述,整理得到完整 的殘餘鮮麵綠之錄。賴由本發明財法,吾人可迅速、精確的 估測補償_以降低Ια效應,增加通訊祕接_的效能。 以上所述僅為本㈣之較佳實施例*已,並非肋限定本發明之申請專 利範圍;凡其它未脫離本發騎揭示之精神下所完成之等效改變或修娜, 均應包含在下述之申請專利範圍中。 【圖式簡單說明】 弟一圖為OFDM通訊系統接收端架構; 第二圖為OFDM資料符號與相位變化關係的示意圖; 第三圖為本發明第-階段運算,其㈣M資料符號與相位變化關係的示 意圖;以及 第四圖為本發明第二階段運算,其〇FM資料符號與相纟之間關係的示 意圖。 14 ⑧ 1291817 【主要元件符號說明】 100本地振盪器 102離散富利葉轉換單元 104相位變化偵測器 106殘餘頻率估測器Combining the descriptions of the first-stage and second-order operations in the above embodiments, the complete residual fresh-faced green records are compiled. Thanks to the invention of the financial method, we can quickly and accurately estimate the compensation _ to reduce the Ια effect and increase the communication _ _ performance. The above is only the preferred embodiment of the present invention (4), and the scope of the patent application is not limited to the scope of the invention; any other equivalent changes or repairs performed without departing from the spirit of the present disclosure shall be included. In the scope of the patent application. [Simple diagram of the diagram] The first picture is the receiving end architecture of the OFDM communication system; the second picture is the schematic diagram of the relationship between the OFDM data symbol and the phase change; the third picture is the first stage operation of the invention, and the (4) M data symbol and phase change relationship The schematic diagram of the second stage operation of the present invention is a schematic diagram of the relationship between the FM data symbols and the phase. 14 8 1291817 [Key component symbol description] 100 local oscillator 102 discrete Fourier transform unit 104 phase change detector 106 residual frequency estimator

Claims (1)

1291817 十、申請專利範圍: 1· 一種多載波通訊系統的殘餘頻率位移補償方法,包括· 進行-第-階财算,每錢理—目前轉符叙她變錄,取得— 第-殘餘頻率估難独該第—殘餘鱗相值回授麵—觸變載波, 其中該第—殘細雜聰起始錢鮮解讎目前資料符號 時之值;以及 進仃-第二階段運算,每次運算複數個資料符號以得到一第二殘餘頻率 估測值並以該第二殘餘頻率估測值補償該解調變載波。 2.如申請專利範圍第丨項所述之多載波通訊系統的殘餘鮮鮮位移補償 方去’其中該目前資料符號與該複數個資料符號包含正交分頻多工資料符 3.如申請專利細第丨項所述之多載波通訊魏的殘餘頻率頻率位移補償 方法,其中該第一階段運算進行3次運算,包括:1291817 X. Patent application scope: 1. A residual frequency displacement compensation method for a multi-carrier communication system, including · performing-first-order financial calculations, each money--currently re-characterizing her change, obtaining-first-residual frequency estimation It is difficult to uniquely the first-residual scale-value-receiving surface-thixo-carrier, where the first------------------------------------------------------------------------------------------- The plurality of data symbols are used to obtain a second residual frequency estimate and the demodulation variable carrier is compensated by the second residual frequency estimate. 2. The residual fresh displacement compensation method of the multi-carrier communication system as described in the scope of the patent application is as follows: wherein the current data symbol and the plurality of data symbols comprise orthogonal frequency division multiplex data symbols. The residual frequency frequency displacement compensation method of the multi-carrier communication Wei described in the third item, wherein the first stage operation performs three operations, including: 取得該第 價/貝】弟i個資料符號的相位變化0 (i),根據算式=色立 、 ixT 殘餘頻率估測值F2補償該解調變載波; (j),根據算式 補償該解調變載波; 價測一第j個資料符號的相位變化0 F2」(/K 學 X(W)r ... -取得該第一殘餘頻率估測值F2 以及 偵測一第k個資料符號的相位變化θ (k),根據瞀式 ⑧ 16 1291817 η(ΐΓλ,θ(ϊ) eU) + —x(j-i)T F2. 取侍該第_殘餘頻率估 ---~_ j 乂T kxT 測值F2補償該解調變•載波; 其中,T為兩資料符號的時間間隔。 、申β專彳〗關第1項所述之多載波通訊线的殘餘頻率位移補償方 去,其中係藉由資料符號中安插之導音訊號以取得該相位變化值。 5. 如申請專利範圍第丨項所述之多載波通訊系統的殘餘頻率位移補償方 法’其中係經一封包之前置序列取得之一頻率位移補償本地振盈器產生之 载波,以得到該起始載波頻率。 6. 如申請糊制第i顧述之錢波通_統_餘頻雜移補償方 2其中該弟二階段運算之該運算係將該複數個資料符號中第一個與最後 個貝料符號之她變化錄_概個雜符叙時間間隔。 7. 如申請專利第6項所述之多載波通鱗統的殘餘解頻率位移補償 方法,其令該第二階段運算根據算式F2= 取得該第二 殘餘_估測值F2,其中Θ⑻為-^师料符號之相位變化值,咖) 為―第_個資料符號之相位變化值,τ為兩資料符號的時間間隔且d為 正整數。 … ^如申請專利範圍第7項所述之錢分頻多工通訊系統的殘餘頻率位移補 償方法,射該第二階段運算中於該第n個資料符號與該第n+d個資料符 ⑧ 17 1291817 就之相位變化差大於π時,將該相位變化差減去小於-π時,將該相 位變化差加上2 7Γ。 •一種多載波通訊系統的殘餘頻率位移補償系統,分為兩階段運算處理一 訊號封包,包括: 一離散富利葉轉換單元,接收該訊號封包作離散富利葉轉換運算; 一相位變化彳貞測器’連接該離散富利葉轉換單元以偵測該訊號封包複數 個資料符號之一相位變化值;以及 一殘餘頻率估測器,連接該相位變化偵測器根據該複數個資料符號之相 位變化值取得一殘餘頻率估測值。 10·如申請專利範圍第9項所述之多載波通訊系統的殘餘頻率位移補償系 統,其中該多載波系統包含正交分頻多工通訊系統。 11·如申請補細第9項所述之多紐通訊祕的殘餘鮮位移補償系 統,其中該殘餘頻率估測器於一第一階段運算時每次對一目前資料符號之 相位變化值作-處理,而取得相同於以—起始載波頻率解調該目前資料符 7 虎時所得到一殘餘頻率估測值以回授補償解調變載波。 移補償系 取得該殘 12·如申請專職圍第n項所述之多載波通鱗統的絲頻率位 統’其中該第一階段運算進行3次運算,包括: 偵測一第i個資料符號的相位變化0(i),根據算式= ^ i 乂T 餘頻率估測值F2補償該解調變載波; 偵測…個資料符號的相位變化θ0),根據算式 ⑧ 18 1291817 ^0 ) + x(j - {\f F2 = —— 卜T 取得該殘餘頻率估測值F2補償該解調變栽波; 偵測第k個育料符號的相位變化Θ (k), 以及 根據算式 F2 j 乂T >(k 一 j)T kxT 取得該殘餘頻率估測值 F2補償該解調變載波; 其中,T為兩資料符號的時間間隔。 =.如=專·圍第9項所述之多載波通減麵殘餘頻率位移補償夺 ,、·充、中該殘餘頻率估測器於一第二階段運算 … 作一運曾以俨$丨μ * 對设數個資料符號 運开以传到一對應之殘餘頻率估測值,以補償該解調變載波。 14. 如申請專利範圍第13項所述之多載波通訊系統的殘餘頻率位移補償李 統:其中該第二階段料之該運算係賴d㈣料符財第-個與最後二 個貝料符號之相位變化差除以該複數個資料符號之時間間隔。 15. 如申請專利範圍第13項所述之多載波通訊系統的殘餘頻率位移補償系 統’其中该第二階段運算之該運算根據算式— '一Τ^τ一''取得 該對應之殘餘鮮侧值F2,第晴讀符叙相位變化 值’⑽刪-第_個資料符號之相位變化值,τ為兩麵號的時間 間隔且d為正整數。 16. 如申請專利範圍第15項所述之多載波通訊系統的殘餘頻率位移補償系 統’其中該第二階段運算中於該第n個資冊號與鄕_师料符號之Obtaining the phase change 0 (i) of the i-data symbol of the first price/Bei, and compensating the demodulation variable carrier according to the formula = color standing, ixT residual frequency estimated value F2; (j), compensating the demodulation according to the formula Variable carrier; price measurement - phase change of the jth data symbol 0 F2" (/K learning X(W)r ... - obtaining the first residual frequency estimate F2 and detecting a kth data symbol The phase change θ (k) is determined according to the formula 8 16 1291817 η(ΐΓλ, θ(ϊ) eU) + —x(ji)T F2. The _ residual frequency estimate is taken ————~_ j 乂T kxT The value F2 compensates the demodulation variable carrier; where T is the time interval of two data symbols. The residual frequency displacement compensation of the multi-carrier communication line described in Item 1 is performed by The sound signal is inserted in the data symbol to obtain the phase change value. 5. The residual frequency shift compensation method of the multi-carrier communication system as described in the scope of the patent application is one of the packets before the sequence is obtained. The frequency shift compensates the carrier generated by the local oscillator to obtain the starting carrier frequency. The system of the i-th description of the money wave _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 7. A multi-carrier pass-sequence residual frequency shift compensation method as described in claim 6, which causes the second stage operation to obtain the second residual estimate according to the formula F2= Measured value F2, where Θ(8) is the phase change value of the -^ teacher symbol, coffee) is the phase change value of the _th data symbol, τ is the time interval of the two data symbols, and d is a positive integer. ... ^Residual frequency displacement compensation method of the money division multiplexing communication system described in claim 7 of the patent application, in the second stage operation, the nth data symbol and the n+d data symbol 8 17 1291817 If the phase change difference is greater than π, the phase change difference is subtracted by less than -π, and the phase change difference is added by 2 7 Γ. • A residual frequency displacement compensation system for a multi-carrier communication system, which is divided into two-stage arithmetic processing, a signal packet, comprising: a discrete Fourier transform unit that receives the signal packet for discrete Fourier transform operation; The detector 'connects the discrete Fourier transform unit to detect a phase change value of one of the plurality of data symbols of the signal packet; and a residual frequency estimator connected to the phase change detector according to the phase of the plurality of data symbols The change value takes a residual frequency estimate. 10. The residual frequency displacement compensation system of the multi-carrier communication system of claim 9, wherein the multi-carrier system comprises an orthogonal frequency division multiplexing communication system. 11. The residual fresh displacement compensation system of the multi-link communication secret described in Item 9 of the application, wherein the residual frequency estimator performs a phase change value of a current data symbol each time in a first stage operation- Processing, and obtaining a residual frequency estimation value obtained by demodulating the current data symbol 7 with the initial carrier frequency to feedback the demodulation variable carrier. The shift compensation system obtains the residual 12·If the multi-carrier pass scale system of the silk frequency system described in the nth item of the full-time application is applied, the first stage operation performs 3 operations, including: detecting an i-th data symbol The phase change 0(i) compensates the demodulated variable carrier according to the formula = ^ i 乂T residual frequency estimated value F2; detects the phase change of the data symbols θ0) according to the formula 8 18 1291817 ^0 ) + x (j - {\f F2 = —— 卜 T obtains the residual frequency estimate F2 to compensate the demodulated morphing wave; detects the phase change Θ (k) of the kth cultivating symbol, and according to the formula F2 j 乂T > (k - j) T kxT obtains the residual frequency estimate F2 to compensate the demodulation variable carrier; wherein T is the time interval of two data symbols. The carrier frequency reduction surface residual frequency displacement compensation, , · charging, the residual frequency estimator in a second stage of operation... After a shipment has been 俨 $ 丨 μ * set a number of data symbols to open to a Corresponding residual frequency estimate to compensate for the demodulation variable carrier. 14. As stated in claim 13 The residual frequency displacement of the wave communication system compensates Li Tong: the operation of the second stage is based on the difference between the phase difference of the first and last two beet symbols divided by the time interval of the plurality of data symbols. 15. The residual frequency displacement compensation system of the multi-carrier communication system as described in claim 13 wherein the operation of the second-stage operation is based on the formula - '一Τ^τ一'' The side value F2, the seventh reading phase change value '(10) delete - the phase change value of the _th data symbol, τ is the time interval of the two-face number and d is a positive integer. 16. As claimed in item 15 The residual frequency displacement compensation system of the multi-carrier communication system is described in the second stage of the operation of the nth number and the symbol of the teacher 19 J291817 目位、遂化差大於;raf ’將該她變化差減去“,小於_π時,將該相位變 化差加上2ττ。 17 •如申請專利範圍第9項所述之多載波通訊系統的殘餘頻率位移補償系 、、先’其中係藉由資料符號中安插之導音訊號以取得該相位變化值。 18·如申請專利範圍第9項所述之多載波通訊系統的殘餘頻率位移補償系 . 統,其中經該訊號封包之前置序列取得之一頻率位移補償本地振盪器產生 • 之載波’以得到該起始載波頻率。 ⑧19 J291817 The position and the difference are larger than that; raf 'subtracts her variation from the difference. When it is less than _π, the phase difference is added by 2ττ. 17 • Multi-carrier communication as described in claim 9 The residual frequency offset compensation system of the system first obtains the phase change value by the sound signal inserted in the data symbol. 18. The residual frequency shift of the multi-carrier communication system as described in claim 9 a compensation system, wherein a frequency shift compensated local oscillator generates a carrier's frequency by the pre-sequence of the signal packet to obtain the starting carrier frequency. 2020
TW94124124A 2005-07-15 2005-07-15 Method and system for residual frequency offset compensation of multi-carrier communication TWI291817B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW94124124A TWI291817B (en) 2005-07-15 2005-07-15 Method and system for residual frequency offset compensation of multi-carrier communication

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW94124124A TWI291817B (en) 2005-07-15 2005-07-15 Method and system for residual frequency offset compensation of multi-carrier communication

Publications (2)

Publication Number Publication Date
TW200703989A TW200703989A (en) 2007-01-16
TWI291817B true TWI291817B (en) 2007-12-21

Family

ID=39461288

Family Applications (1)

Application Number Title Priority Date Filing Date
TW94124124A TWI291817B (en) 2005-07-15 2005-07-15 Method and system for residual frequency offset compensation of multi-carrier communication

Country Status (1)

Country Link
TW (1) TWI291817B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI575900B (en) * 2014-06-05 2017-03-21 英特爾智財公司 Methods and devices for channel estimation and ofdm receiver
US9948415B2 (en) 2015-03-27 2018-04-17 Intel IP Corporation Method of processing a plurality of signals and signal processing device

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI575900B (en) * 2014-06-05 2017-03-21 英特爾智財公司 Methods and devices for channel estimation and ofdm receiver
US9806912B2 (en) 2014-06-05 2017-10-31 Intel IP Corporation Methods and devices for channel estimation and OFDM receiver
US9948415B2 (en) 2015-03-27 2018-04-17 Intel IP Corporation Method of processing a plurality of signals and signal processing device

Also Published As

Publication number Publication date
TW200703989A (en) 2007-01-16

Similar Documents

Publication Publication Date Title
CN104579476B (en) Method and device for estimating dispersion in optical coherent communication
JP6866725B2 (en) Antialiasing channel estimator, method and receiver
WO2018054236A1 (en) Method and device for ofdm system synchronous tracking
JP4153907B2 (en) OFDM receiving apparatus and OFDM receiving method
US20040202234A1 (en) Low-complexity and fast frequency offset estimation for OFDM signals
CN112887238A (en) IQ imbalance correction method and device, and receiver
CN104601509B (en) The timing synchronization device and method of multicarrier modulated signal
US7639750B2 (en) Phase tracking method and device thereof
JP3905541B2 (en) Delay profile estimation apparatus and correlator
TWI291817B (en) Method and system for residual frequency offset compensation of multi-carrier communication
CN106101039B (en) A Data-Aided Frequency Offset Estimation Method with Adjustable Precision
JP4115466B2 (en) OFDM receiver
CN118646621A (en) OFDM waveform frequency deviation measurement method suitable for cross-subcarrier frequency deviation scenarios
CN108989261B (en) A timing synchronization method, device and related equipment for a communication system
CN116016049B (en) A frequency offset estimation method suitable for high dynamic and low signal-to-noise ratio
JP3945623B2 (en) Frequency synchronization method and OFDM receiver using the same
TWI717864B (en) Baseband system for wireless receiver and baseband signal processing method thereof
US10924320B2 (en) IQ mismatch correction module
JP4335125B2 (en) Timing synchronization circuit
CN100355255C (en) Synchronous method of large search range OFDM system carrier based on statistical uniform
CN106453193A (en) Frequency offset estimation method and device
CN115412417B (en) Carrier initial phase determination method, device, terminal and storage medium
JP2015073223A (en) Receiving apparatus and method
CN107105222B (en) Symbol detection circuit and method
JP4827449B2 (en) Amplitude phase control device and receiving system

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees