TW384582B

TW384582B - Method for correcting the carrier frequency errors and spread spectrum coding sequence clock in direct sequence spread spectrum communication system and the receiver using the method

Info

Publication number: TW384582B
Application number: TW87108069A
Authority: TW
Inventors: Jia-Chi Huang; Shin-Yuan Wang; Yu-Miau Jung
Original assignee: Chungshan Inst Of Science & Te
Priority date: 1998-05-25
Filing date: 1998-05-25
Publication date: 2000-03-11

Abstract

The present invention provides a kind of correcting method for the carrier frequency errors and spread spectrum coding sequence clock in direct sequence spread spectrum communication system and the receiver using the method wherein the carrier frequency errors are divided into the integer and non-integer portions. The integer carrier frequency error is defined as integer multiple to the inverse of spread spectrum code sequence period. By matching the received signal with the spread spectrum code sequence with center frequency as the integer multiple of the inverse of spread spectrum code sequence period and searching the maximum amplitude in the matching result, the integer carrier frequency errors can be obtained; by comparing the said amplitude and the amplitude before and after the integer matching to obtain a directional flag; after correcting the integer carrier frequency errors, calculating the difference of rotation angles between two timely adjacent matching results based on the directional flag and calculating the non-integer carrier frequency errors; after correcting the carrier frequency errors, the correction of spread frequency code sequence clock can search the initial point of spread frequency code according to the minimum phase displacement system with minimum delayed spread properties.

Description

A7 五、發明説明（1 ) 發明領域本發明是關於直接序列展頻通信的領填、，特別是關於直接序列展頻通信系統中載波頻率誤差及展頻碼序列時脈之絛正。發明背景在^接^序列展頻通信中，資訊解調前需要修正載波頻率誤差及展頻碼序列時脈以遮+載波與展頻碼同步。習知的作法中有用滑動關連器來 - ‘ 、 >. 作時脈修正者，然而這<種作法不僅費時，且由於時脈課差係與載波頻率誤差同時存在且耦舍在一，故上述習知作法實質上^無洼：有效修i時脈誤差。經濟部中央標準局負工消費合作社印聚；請先聞讀背面之注.意事項再填寫本頁) 另一種習用作法是以匹配遽波H參達成載波與展頻今同I，然後再以鎖相迴路進行進一‘步的修正。但是這種作法限於載波頻率誤差小於展頻碼序列週期之倒數的情況下.，才能正確運作；然而事實上，直接序列展頻通信中之載波頻率誤差往往大於展頻碼序列週期之倒數，而使致此種作法不適用。美氩i鈾篕號及美國專鰣箨號中提出以展頻碼的快速傅利葉轉換來匹配接收信號的快揀構利葉轉換之作法，這兩個專利均是針對衛星定位系統（GPS)信號的特性來設計。上述兩項專利均是藉增加匹配濾波器的快速傅利 L:\EXT\S2\52160.DOC\4 — 4 — 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐）經濟部中央標準局員工消费合作社印製 A7 B7 五、發明説明（2 ) 葉轉換運算之點數冬增加1波_頻瘗誤盖佐钍之精確度+此種方式對於資料傳送速率只有50 bits/secJ1_Q_P_S_^言尚稱可jf，但對於一般的直接序列展頻通信系統如分碼多重進接（QPMA)系統而言 ”資訊傳送速率係遠大於50 bits/sec，故^上述作J法考於一般的直接序列展頻通信系羞ϋ可行。此外，上述兩項專利均是將匹配濾波器輸出的最高尖峰當作展頻碼序列時脈，這是因爲GPS的 ^_道只有一個主要路餘，故只會產生一個最声尖峰；然而一般直接序列展趨通信系統的傳輸頻道常是具有多重路徑，例如CDMA系統的通信頻道即是多重路徑衰減頻道，因此，並不能以匹配濾波器輸出之最高點作爲展頻碼序列時脈〇基於上述習知技藝所具有之缺憾及所面臨之難題，本發明之一目的即在於提出一種適用於修正一般直接序列展頻通信系統具有任意大小之載波頻率誤差的方法及裝置；本發明之另一目的在於提出一種可一併修正載波頻率誤差及展頻碼序列時脈的方法及裝置 ♦ 本發明之又一目的在於提出一種適用於所有直接序列展頻通信系統“包括具有多重路徑衰減頻道及具有任意資訊傳送速率之系統)之載波頻率 * L:\rmS2\52160.DOC\4 — 5 — 本紙張尺戽適用中國國家標準（CNS ) A4規格（210X297公釐） (請先Μ讀背面之汰意事項•再填寫本頁) -裝- 訂 A7 B7 五、發明説明（3 ) 誤差修正及展頻碼序列時脈修正的方法及裝i » 本發明之再一目的在於提出一種全數位化之修正載波頻率誤差及展頻碼序列時脈的方法$ 裝置，使其適用以積體電路鉑ϋ參度施，而' 符合”輕薄短小"的設計要求。發明综述本發明首揭將載波頻率誤差之修正區分爲整數載波頻率誤差修正及非整數載波頻率誤差修正；其中整數載波頻率誤差係定義爲展頻碼序列週期倒數之某一整數倍，而非整數載波頻率誤差即爲小於展頻碼序列週期之倒數者。利用整數載波頻率誤差被賦予之規則性，可藉由將接收信號匹配中心頻率爲展頻碼序列週期倒數之整數倍的該展頻碼序列，，以及搜尋匹配結果中振輻最太者.，即可迅速得出整數載波頻率誤差。經濟部中央標窣局負工消費合作社印製 t請先閱讀背面之注t事項界填寫本I ) 利用非鸯數t載波竭％誤差小.於展頻碼序列週期倒數之事實"可藉前述匹配結卷中_旋轉角度差啲tt算而得出’非整數載波頻率誤差。本發明之上述方法可簡便有效地修正具有任意大小之載波頻率誤差，且可以全數位化的方式來實施；更可與任何習用之展頻碼序列時脈的修正技術相配合=» L:\EXT\52\S2160.DOC\4 — 6 — 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐）經濟部中央標準局員工消費合作社印繁 A7 B7五、發明説明（4 ) 此外，對於已修正載波頻率誤差之直接序列展頻通信系統而言，本發明首揭利用頻道爲一最小相位移_ m應具有最小延遲分散（delay spread) 之性質〜，而搜索頻道脈衝f摩冬主要路徑琴遲分散最小I，以決定鏖蠤廳扁齟蜍磁·；本發明之修正展頻碼序列時脈之方法可以數位化的方式來實施。本發明上述之修正載波頻率誤差及修正展頻碼序列時脈之方法可適用於具有多重路徑及具有任意大小之資訊傳送速率的直接序列展頻通信系統。直接序列展頻通信系統中*有些系統具有領航信號*有些系統則否；對於存在領航信號或不存在領航信號之直接序列展頻通信系統，本發明提出之方法多裝置均可藉同一架構來進行載波頻率誤差修正及屏，哼暾農涤正。本發明之其它技術特徵和詳細技術内容可藉下列之圖式及詳細説明而瞭解。圖式之簡單説明：圖1爲本發明第一實施例之系統架構的方塊圖圖2爲本發明第一實施例中匹配濾波器之一匹配結果圖3爲圖1中整數載波頻率誤差修正之方塊圖圖4爲圖1中非整數載波頻率誤差修正之方塊圖 L:\EXT\S2\S2160.DOC\4 — 1 ^ *(請先閲婧背面之注*意事項—填寫本莧) .-ο-裝.A7 V. Description of the invention (1) Field of the invention The present invention relates to the acquisition of direct-sequence spread-spectrum communication, and in particular to the carrier frequency error and the spread of the spread-spectrum code sequence in a direct-sequence spread-spectrum communication system. Background of the Invention In sequence spread spectrum communication, it is necessary to correct the carrier frequency error and the frequency of the spread code sequence before demodulating the information so that the + carrier is synchronized with the spread spectrum code. In the conventional practice, there is a sliding connector to be used as a clock corrector. However, this method is not only time consuming, but because the clock difference and the carrier frequency error coexist and are rounded together, Therefore, the above-mentioned conventional practice is essentially non-defective: effectively repairing the clock error. The Central Standards Bureau of the Ministry of Economic Affairs ’s Consumer Cooperatives printed and printed together; please read the notes on the back. Please note this page before filling out this page.) Another practice is to match the carrier wave and the spread frequency by matching the wave H parameters, and then use the The phase-locked loop is further modified. However, this method can only work correctly if the carrier frequency error is less than the reciprocal of the spreading code sequence period. However, in fact, the carrier frequency error in direct sequence spread spectrum communication is often greater than the reciprocal of the spreading code sequence period. Renders this practice inapplicable. The U.S.U.U. and U.S.S.A. put forward the method of using fast spreading code of spread spectrum codes to match the fast picking structure of receiving signals. Both patents are for GPS signals. To design. The above two patents are based on the fast Fourier L of the matched filter: L: \ EXT \ S2 \ 52160.DOC \ 4 — 4 — This paper size applies the Chinese National Standard (CNS) A4 specification (210X297 mm) Central Ministry of Economic Affairs Printed by the Consumer Bureau of Standards Bureau A7 B7 V. Description of the invention (2) The number of points in the leaf conversion operation is increased by 1 wave in winter _ frequency error is incorrect and the accuracy is + this method has only 50 bits / sec for data transfer rate The word is still called jf, but for a general direct sequence spread spectrum communication system, such as a code division multiple access (QPMA) system, the "information transmission rate is much greater than 50 bits / sec, so the above J method is considered for general Direct-sequence spread-spectrum communication is ashamed to be feasible. In addition, the above two patents both use the highest peak output of the matched filter as the spread-spectrum code sequence clock. This is because the GPS track has only one main path, so Only one of the loudest peaks will be generated; however, the transmission channels of direct-sequence communication systems often have multiple paths. For example, the communication channel of a CDMA system is a multi-path attenuation channel. The highest point is used as the clock of the spread spectrum code sequence. Based on the shortcomings and difficulties faced by the above-mentioned conventional techniques, one object of the present invention is to propose a method suitable for modifying the general direct sequence spread spectrum communication system with any size. Method and device for carrier frequency error; Another object of the present invention is to provide a method and device that can simultaneously modify the carrier frequency error and the clock of the spreading code sequence ♦ Another object of the present invention is to provide a method suitable for all direct sequences Spread spectrum communication system "including carrier frequency with multi-path attenuation channels and systems with arbitrary information transmission rate) * L: \ rmS2 \ 52160.DOC \ 4 — 5 — This paper size applies to China National Standard (CNS) A4 specifications (210X297mm) (Please read the notice on the back of the page and then fill out this page)-Installation-Order A7 B7 V. Description of the invention (3) Method and installation of error correction and spreading code sequence clock correction » Another object of the present invention is to propose a fully digital method for correcting carrier frequency error and clock of spreading code sequence. The device makes it applicable to integrated circuit platinum I ’m enthusiastic about the design requirements of “light, thin, short”. Summary of the Invention The first disclosure of the present invention distinguishes the correction of carrier frequency error into integer carrier frequency error correction and non-integer carrier frequency error correction; where the integer carrier frequency error is It is defined as an integer multiple of the reciprocal of the period of the spreading code sequence, and the non-integer carrier frequency error is less than the reciprocal of the period of the spreading code sequence. By using the regularity given to the integer carrier frequency error, the received signals can be matched by The spreading code sequence whose center frequency is an integer multiple of the reciprocal of the period of the spreading code sequence, and the one with the largest amplitude in the search result, can quickly obtain the integer carrier frequency error. Printed by the Central Bureau of Standards, Ministry of Economic Affairs and Consumer Cooperatives, please read the note on the back and fill in this I) Use of non-numbered carriers to reduce the% error is small. The fact that the cycle of the spreading code sequence counts down " can be borrowed The non-integer carrier frequency error is calculated by calculating the _rotation angle difference 啲 tt in the aforementioned matching volume. The above method of the present invention can simply and effectively correct a carrier frequency error having any size, and can be implemented in a fully digitalized manner; it can also cooperate with any conventional technique for correcting the frequency of a spreading code sequence = »L: \ EXT \ 52 \ S2160.DOC \ 4 — 6 — This paper size is applicable to Chinese National Standard (CNS) A4 (210X297 mm) Employees' Cooperatives of the Central Standards Bureau of the Ministry of Economic Affairs, Printing Cooperative A7 B7 V. Description of Invention (4) In addition, For a direct-sequence spread-spectrum communication system with modified carrier frequency error, the present invention firstly utilizes the property that the channel is a minimum phase shift_m should have a minimum delay spread ~, and the search channel pulse f The path delay is minimized by I to determine the magnetic properties of the flat toad. The method of correcting the clock of the spreading code sequence of the present invention can be implemented in a digital manner. The method for correcting the carrier frequency error and correcting the clock of the spreading code sequence according to the present invention can be applied to a direct sequence spread spectrum communication system with multiple paths and an information transmission rate of any size. In direct sequence spread spectrum communication systems * some systems have pilot signals * some systems do not; for direct sequence spread spectrum communication systems with or without pilot signals, the method proposed by the present invention can be performed by multiple devices with the same architecture Carrier frequency error correction and display. Other technical features and detailed technical contents of the present invention can be understood through the following drawings and detailed description. Brief description of the drawings: FIG. 1 is a block diagram of the system architecture of the first embodiment of the present invention. FIG. 2 is a matching result of one of the matched filters in the first embodiment of the present invention. Block diagram Figure 4 is a block diagram of the non-integer carrier frequency error correction in Figure 1 L: \ EXT \ S2 \ S2160.DOC \ 4 — 1 ^ * (Please read the note on the back of Jing first * Notes-fill out this note). -ο- 装.

.1T om. 本紙張尺度適用中國國家標準（CNS > Α4規格（2Ι0Χ297公釐）經濟部中央標準局貝工消费合作社印褽 A7 B7 五、發明説明（5 ) 圖5爲圖1中展頻碼序列時脈修正之方塊圖圖6爲本發明第二實施例之流程圖囷7爲本發明第二實施例中匹配濾波器之一匹配結果，圖8爲本發明第二實施例中非整數載波頻率镁差修正之方塊圖發明之詳細説明: 本發明所提出之方法及裝置可適用於所有直接序列展頻通信系統，尤其適用於具有高侈送速率及多重路徑4 m釔1 m i熏.— 系統。依據本發明之第一實施例的系統架挺如圖1 所示。高頻信號由天線101接收後經高頻接收模組102(RF front-end)降爲複數等效基頻信號（complex equivalent baseband signal)，，此基頻信號有實數部份 (I(t))及虛數部份（Q⑼。基頻信號經類比數位轉換器103以一個展頻碼時間（Tc)爲週期抽樣後 < 成爲一複數數位序列r[n](r[n]=I[n]+jQ[n]) 〇此複數數位序列儲存在記憶體104中。將儲存之複數數位序列分割爲長度是N的區段可以是一個領航展頻碼序列的長度（Νρ)或一個資訊展頻碼序列的長度（Nc) 分割後的數位序列（r[n])經快速傅利葉轉換（FFT) 運算後¥，產生在頻域的接收信號（R[k])。 L:\EXT\52\52160.DOC\4 — 8 — 本紙張尺度適用中國國家標準（CNS ) Α4規格（210Χ297公釐） -(請先Μ讀背面之注-意事項界填寫本頁) 訂 A7 B7 五、發明説明（6 ) 在尚未討論載波頻率誤差修正及展頻碼時脈修正之前，先來描述快速傅利葉轉換匹配濾波器的工作原理及快速傅利葉轉換的調變特性如下： I.快遣i逛重替遙氐I遙H姐iS < xl[n]及χ2[η]是二個長度爲N的數位信號（η是介於0和Ν - 1之間），其FFT分別爲Xl[k]及X2[ k ]。若 X3[k]等於 Xl[k]和 X2*[k]的乘積，則 x3[n]是 xl[n]和 x2*[((-n))N](註：（00)>1球m〇d N，（(x))n介於0和N-1之間）循環迴旋（circular convolution)的結果。 X3[k] = Xl[k]X2*[^] N-1 * (1).1T om. This paper size applies to Chinese national standard (CNS > Α4 size (2IO × 297 mm)) Central Printing Bureau of the Ministry of Economic Affairs, Shellfish Consumer Cooperative, India A7 B7 5. Description of the invention (5) Figure 5 is the spread spectrum in Figure 1. Block diagram of code sequence clock correction. Figure 6 is a flowchart of the second embodiment of the present invention. Figure 7 is a matching result of one of the matched filters in the second embodiment of the present invention. Figure 8 is a non-integer in the second embodiment of the present invention. Block diagram of carrier frequency magnesium difference correction detailed description of the invention: The method and device proposed by the present invention can be applied to all direct-sequence spread spectrum communication systems, especially suitable for high-speed and multi-path 4 m yttrium 1 mi fumigation. — System. The structure of the system according to the first embodiment of the present invention is shown in Figure 1. The high-frequency signal is received by the antenna 101 and reduced to a complex equivalent fundamental frequency signal by the high-frequency receiving module 102 (RF front-end). (Complex equivalent baseband signal), this baseband signal has real part (I (t)) and imaginary part (Q⑼). The baseband signal is sampled by the analog-to-digital converter 103 with a spreading code time (Tc) as a period. After < becomes plural Bit sequence r [n] (r [n] = I [n] + jQ [n]) 〇 This complex digital sequence is stored in the memory 104. The stored complex digital sequence is divided into segments of length N, which can be The length of a pilot spreading code sequence (Nρ) or the length of an information spreading code sequence (Nc) is divided into a digital sequence (r [n]) after a fast Fourier transform (FFT) operation. Receive signal (R [k]). L: \ EXT \ 52 \ 52160.DOC \ 4 — 8 — This paper size applies to China National Standard (CNS) A4 specification (210 × 297 mm)-(Please read the note on the back first -Please fill in this page on the subject matter page) Order A7 B7 V. Description of the invention (6) Before discussing carrier frequency error correction and spreading code clock correction, first describe the working principle of the fast Fourier transform matched filter and fast Fourier transform The modulation characteristics of I are as follows: I. Quickly send i to replace Yao Yi i Yao H iS < xl [n] and χ2 [η] are two digital signals of length N (η is between 0 and N- 1), the FFTs are Xl [k] and X2 [k]. If X3 [k] is equal to the product of Xl [k] and X2 * [k], then x3 [n] is xl [n] and x2 * [((-n)) N] (Note: (00) & gt 1 ball m〇d N, ((x)) n is between 0 and N-1) the result of circular convolution. X3 [k] = Xl [k] X2 * [^] N-1 * (1)

x3[n] = ^Z〇xl[m]x2 [((n + m))NJ 其中（(n+m))N是11 + 111除以N的模數且（(n+m))N是介於0 和N - 1之間。 Π .快速傅利葉轉換.的調變特性，經濟部中央標準局員工消費合作社印製 (請先閲讀背面之注-意事項#'填寫本I) 一個長度爲N的數位信號x[n]，其FFT爲X[k]。將x[n]調變到載波，即是x[n]i^/n，〜是 exp(-y勢）。調變後的信號之FFT爲辱-/·))#]。載波頻率誤差修正及展頻碼序列時脈修正的方法，可依據領航信號是否存在有不同的流程及做法，但均可利用同一之架構來實施。本發 L:\EXT\52\52160.DOC\4 — 9 — 本紙張尺度適用中國國家標準（CNS ) Α4規格（210Χ297公釐） A 7 B7 五、發明説明（7 ) 明之第一實施例是考慮領航信號存在時的情況〇 I.當領航信號存在時當領航信號存在時，先搜尋整數載波頻率誤差，然後估計非整數載波頻率誤差，最後棱尋展頻碼時脈。φ a.整數載波頻率誤差修正整數載波頻率誤差是一個領航展頻碼序列週期倒數（^)的整數.倍。我們以快速傅利葉轉換 y ΡΙ(：匹配濾波器（FFT Matched Filter)來搜尋整數載波頻率x3 [n] = ^ Z〇xl [m] x2 [((n + m)) NJ where ((n + m)) N is the modulus of 11 + 111 divided by N and ((n + m)) N Is between 0 and N-1. Π. Modulation characteristics of the fast Fourier transform. Printed by the Consumer Cooperatives of the Central Standards Bureau of the Ministry of Economic Affairs (please read the note on the back-Issue # 'Fill this I). A digital signal of length N [x], which The FFT is X [k]. Adjust x [n] to the carrier, which is x [n] i ^ / n, and ~ is exp (-y potential). The FFT of the modulated signal is shame-/ ·)) #]. The methods of carrier frequency error correction and spread-spectrum code sequence clock correction can be implemented according to different procedures and methods of pilot signals, but they can be implemented using the same architecture. This issue L: \ EXT \ 52 \ 52160.DOC \ 4 — 9 — This paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 × 297 mm) A 7 B7 5. Description of the invention (7) The first embodiment of the invention is Consider the situation when the pilot signal exists. I. When the pilot signal exists When the pilot signal exists, first search for the integer carrier frequency error, then estimate the non-integer carrier frequency error, and finally search for the spreading code clock. φ a. Integer carrier frequency error correction Integer carrier frequency error is an integer. multiple of the period reciprocal (^) of the pilot sequence code sequence. We use fast Fourier transform y Π (: FFT Matched Filter) to search for integer carrier frequency

II

誤差。這一個方法是將接收信號匹配ί公邋枣義整數載波頻率誤差的領航展頻碼序列，然後依I 匹配濾波器輸出的振幅決定整數載波頻率誤差〇假設領航展頻碼是cp[n]，（^閃是Cp[n]的FFT。接收信號r[n]是 r[n] = cp[((« + a)) N ]exp〇'«offn + χ} + φ] (2) iy Ρ 經濟部中央標準局貝工消費合作社印掣 *(請先閱讀背面之注-意事項#'填寫本頁) 其中α是分割區段起點和領航展頻碼序列起點的差値，iiy是載波頻率誤差V X是隨機相位，η[η] 是雜訊ί這裏考慮頻道只有一個主要路徑的情況，當頻道有多重路徑時，則每一個路徑都和單一路徑相同）。定義R[k]C/[((k-i))Np]的IFFT爲 CRi[n] ( CRi[n]是匹配濾波器的輸出）。CRi[n]可寫爲 L:\EXT\S2\52160.DOC\4 — 1 〇 — 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐） A7 B7 五、發明説明（8 ) Ν~ι CR；[n]= X c [((m + a))NJc *[((m + n))N Jexpiy'iy^w-y^iCim + n))^ m = 〇 P P p p (3) 其中η·[η]是輸入雜訊所產生V當《例爲0時，CRi[a] 是 N-1 = ^ Σ exp〇'(iy^-~ΐ)πι}6χρ{-7'-^ΐα + Ζ} m = 0 ^ p 描繪在圖2。由圖 (4) 依據方程式4並令i==〇,將 N. 2可得知當 < off 時，!2^1小於0 22 无時，1^1大於0.636。當丨〜μ驚 (請先閲讀背面之汰意Ϋ項.再填寫本頁) 依據方程式4改變不同的i値，即是將匹配濾波器的脈衝響應（領航展頻碼序列）調變到頻率error. This method is to match the received signal with the pilot frequency code sequence of the integer carrier frequency error, and then determine the integer carrier frequency error according to the amplitude of the I matched filter output. Assuming the pilot frequency code is cp [n], (^ Flash is the FFT of Cp [n]. The received signal r [n] is r [n] = cp [((«+ a)) N] exp〇 '« offn + χ} + φ] (2) iy Ρ Printed by the Central Standards Bureau of the Ministry of Economic Affairs of the Bayer Consumer Cooperatives * (please read the Note-Issue # 'on the back page first to fill out this page) where α is the difference between the start of the segment and the start of the pilot code sequence, and iiy is the carrier frequency The error VX is a random phase and η [η] is the noise. Here we consider the case where the channel has only one main path. When the channel has multiple paths, each path is the same as a single path). Define the IFFT of R [k] C / [((k-i)) Np] as CRi [n] (CRi [n] is the output of the matched filter). CRi [n] can be written as L: \ EXT \ S2 \ 52160.DOC \ 4 — 1 〇 — This paper size applies to China National Standard (CNS) A4 specification (210X297 mm) A7 B7 V. Description of the invention (8) Ν ~ ι CR; [n] = X c [((m + a)) NJc * [((m + n)) N Jexpiy'iy ^ wy ^ iCim + n)) ^ m = 〇PP pp (3) where η · [η] is the V generated by the input noise. When the example is 0, CRi [a] is N-1 = ^ Σ exp0 '(iy ^-~ ΐ) πι} 6χρ {-7'-^ ΐα + Z} m = 0 ^ p is depicted in Figure 2. From Figure (4), according to Equation 4 and let i == 〇, N. 2 can be known that when < off,! 2 ^ 1 is less than 0 22 without, 1 ^ 1 is greater than 0.636. When 丨 ~ μ is surprised (please read the 汰 item on the back. Then fill in this page) To change different i 値 according to Equation 4, it is to adjust the impulse response of the matched filter (pilot show code sequence) to the frequency

爲的載波上v將接收信號和中心頻率爲这;·之 iV N 匹配濾波器脈衝響應匹配後-，其輸出振幅是 CR^ai N. 。祭載波頻率誤差和匹配滤波器脈衝響應的中心頻率的差値小於丄時（ Np foff 2π . --i Np <—-) Np；則匹經濟部中央標隼局負工消费合作社印製 &濾施盎鐘达钧簏馇倉土族以站，其他情況振幅 ·- —* — 則會小於0.22。我們可以依據方程式4，迴旋移動領航展頻碼的FFT (改變不同的i値），得到匹配濾波器輸出振幅（。若整數載波頻率誤差是，\£Rj_[ai N. N.For the carrier v, the received signal and the center frequency will be this; · After the iV N matched filter impulse response is matched, its output amplitude is CR ^ ai N. The difference between the carrier frequency error and the center frequency of the impulse response of the matched filter is less than 丄 (Np foff 2π. --I Np < ---) Np; ; Filter Shi Ang Zhong Da Jun Ao Cang Tu ethnic group to stand, in other cases amplitude--* *-will be less than 0.22. According to Equation 4, we can circulate the FFT of the pilot frequency display code (change different i 依据) to obtain the matched filter output amplitude. (If the integer carrier frequency error is \ £ Rj_ [ai N. N.

L:\EXT\S2\52160.D 本紙張尺度適用中國國家標準（CNS ) A4規格（2丨0X297公釐） A7 B7 五、發明説明（9 ) 會有最大値*經整數載波頻率誤差修正後，非整數載波頻率誤差可寫爲請先閲讀背 ιέ 之注意· ί裝頁 ί^ωoff - ω〇ίΓ (5) 是估計之整數載波頻率誤差，且Ιδο^Ι小於 π 或等於由圖2可得知對稱於頻率爲0 ΝΡ Νρ 的値會近似 < 且頻道脈衝響應的旋轉方向是反時鐘9同理當且接近時，，丨C巧;1-[-θ和丨巧》㈣的値會近似，且頻道脈衝響應的旋 ΝΡ Νρ 訂轉方向是順時鐘。將R[k]和Cp#[((k-i))Np]相乘後儲存在記憶體110中 #經反快速傅利葉轉換（IFFT)運算111後計算其振幅112。之後即可進行整數載波頻率誤差修正114 ，其系統方塊圖如圖3所示，即搜尋IFFT運算後的最大振幅並記錄爲Αρμ]301。在可能的載波頻率誤差範圍内，改變i値並重複上述運算。以經濟部中央標準局貝工消費合作社印製：lAGhz Ϊ :IOMjz，\ = 1024爲例、若載波頻率誤差在10 ppm以下，則i的搜尋範圍是{-3, -2, -1，0, 1, 2, 3} P 比較Ap[i]並選擇最大者是爲ApK]3〇2 ·所以整數載波頻率誤差決定爲^^w。比較七丨€]和Ή + 1]和 νίρμ-η的値303，若\⑷和ylpy + l]相近則決定方向旗. L:\EXT\52\S2160.DOC\4 -12 - 本紙張尺度適用中國國家標準（CNS ) Α4規格（210Χ297公釐） A7 B7 經濟部中央標準局貝工消費合作社印製五、發明説明（1(3 ) 標爲 1 3若七[£]和、ρ-ι]的値相近_ 決定方向旗標爲 -1 。若非上述兩種狀況則決定：方向旗標爲0 最後輸出整數載波頻率誤差控制信號給展頻碼的 FFT之共軛複數109或自動頻率控制 10f ! 用以修正整數載波頻率誤差。 (修正之後 > 將 i値設爲0 9 若由展頻碼的FFT之共輊複教109修正則i 値設爲 ί >若由自動頻率控制 1 0 5修正則 i値設 «λ 爲 0 。命 0 重新計算匹配濾波器輸出並搜尋最大振幅 (APmax)。'輸出啓動信號、方向旗標及 APraax 0 鳞非整數載波頻率誤差修正非整數載波頻率誤差是估計相鄰兩個快速傅利葉轉換匹_配滤波器輸出的旋轉角度差而得依據方程式4，假設相鄰兩個匹配濾波器輸出之第一個輸出的第α點如下（已知整數載波頻率誤差爲 2jt ) NP N-l CR〇[a] = iVp Σ exp〇'A6；o(rm}exp{-7—^α + ^} (6) m = 0 Ρ ，第二個輸出的第ot點則是 CR0[or] = iVp Σ exp〇'Aiy〇irm}exp{-7(-^L^ar-Aiy〇irNp) + z} m= 0 p ⑺ 由方程式6和7可得知相鄰兩個匹配滤波器輸出的旋轉角度差（ΔΘ)即是Δα 〇ffNp，1 經上述整數載波頻率誤差修正後， |小於或等於- π 9 所以 |Δλ 小於或等於π。計算相鄰兩個匹配濾波器輸出的旋轉角度差（Δθ)，再將此旋轉角度差除以 L:\EXT\52\52160.DOC\4 · 13 - 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐）請先閲讀背面之注. 意事項再- 填寫本頁 A7 B7 五、發明説明（11 ) 一個領航展頻碼序列週期·，即爲非整數載波頻率誤差（#，若以數化方式表示Tc = 1 )，在非整丄、p上c 數載波頻率誤差修正時，方向旗標可以決定旋轉角度差的計算方法。若方向旗標是· 1時^，以蝻時鐘方向計算旋轉角度差，ΔΘ介於0度及-360度 *若方向旗標是1時 < 以反時鐘方向計算旋轉角度差，ΔΘ介於0度及360度。若方向旗標是0時，旋轉方向未決定，以反時鐘方向或順時鐘方向計算旋轉角度差但選擇ΔΘ介於〇度至180度或介於 0度於至-180度》假！此差統是t邋盆巍羞蠡施揚1鍈、，所以頻道有多：1：盛ϋ衰.減iadingj。爲_了得到較佳的估計，我們可以選擇幾個主要路徑來估計非整數載波頻率！差。非整數載波頻率誤差修正115之系統方塊圖如圖4所示。 •輸入^_„並決定一個門拳，選擇J個匹配濾波器輸出振幅大於門檻句i |螯徑4以，計算夸經濟部中央標準局—工消費合作社印策 (請先閲讀背面之注·意事項耳填寫本頁) 一主要路徑的角度（死/·]) 4〇2 再依撼方逸遽撮計 - 算每一個主要路徑的旋轉角度差(△呵η) 403 ;計算每一路徑之旋^轉J .度差的M m棄拖，(爾，帅：，·以及平均每一路徑的哪,{Δ^：/Ί}爲Α否：，L: \ EXT \ S2 \ 52160.D This paper size applies to Chinese National Standards (CNS) A4 specifications (2 丨 0X297 mm) A7 B7 V. Description of the invention (9) There will be a maximum value * After correction by integer carrier frequency error The non-integer carrier frequency error can be written as Please read the note first. The installation page ^^ off-ω〇ίΓ (5) is the estimated integer carrier frequency error, and Ιδο ^ Ι is less than π or equal to It is known that 对称, which is symmetrical to the frequency 0 NP ρ, is approximately < and the direction of rotation of the channel impulse response is anticlockwise. Similarly and close, 丨 Cqiao; 1-[-θ 和丨巧》㈣'s 値会Approximately, and the rotation NP NRp subscription direction of the channel impulse response is clockwise. Multiply R [k] and Cp # [((k-i)) Np] and store in memory 110. # Inverse fast Fourier transform (IFFT) operation 111 is used to calculate its amplitude 112. After that, the integer carrier frequency error correction 114 can be performed. The system block diagram is shown in Figure 3, that is, the maximum amplitude after the IFFT operation is searched and recorded as Αρμ] 301. Within the range of possible carrier frequency error, change i 値 and repeat the above operation. Printed by the Shellfish Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs: lAGhz Ϊ: IOMjz, \ = 1024 as an example. If the carrier frequency error is less than 10 ppm, the search range of i is {-3, -2, -1, 0 , 1, 2, 3} P compares Ap [i] and selects the largest one as ApK] 3〇2. So the integer carrier frequency error is determined as ^^ w. Compare 丨 €] and Ή + 1] and ν303 of νίρμ-η. If \ ⑷ and ylpy + l] are close, the direction flag is determined. L: \ EXT \ 52 \ S2160.DOC \ 4 -12-paper size Applicable to Chinese National Standards (CNS) A4 specifications (210 × 297 mm) A7 B7 Printed by the Shellfish Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs 5. Description of the invention (1 (3) marked as 1 3 Ruoqi [£] and, ρ-ι The 値 similarity of _ determines the direction flag to be -1. If it is not the above two conditions, it is determined: The direction flag is 0. Finally, the integer carrier frequency error control signal is output to the conjugate complex number 109 of the FFT or the automatic frequency control 10f ! Used to correct the integer carrier frequency error. (After correction > Set i 値 to 0 9 If it is corrected by the FFT of the spreading code FFT, then i 値 is set to ί > If it is controlled by automatic frequency 1 0 5 To correct, set «λ to 0. Recalculate the output of the matched filter and search for the maximum amplitude (APmax). 'Output start signal, direction flag, and APraax 0 scale non-integer carrier frequency error Correct non-integer carrier frequency error Is estimated The difference between the rotation angles of the two adjacent fast Fourier transform matching filter outputs is based on Equation 4. Assume that the α point of the first output of the two adjacent matched filter outputs is as follows (known as the integer carrier frequency error is 2jt) NP Nl CR〇 [a] = iVp Σ exp〇'A6; o (rm) exp {-7— ^ α + ^} (6) m = 0 ρ, the ot point of the second output is CR0 [or] = iVp Σ exp〇'Aiy〇irm} exp {-7 (-^ L ^ ar-Aiy〇irNp) + z} m = 0 p ⑺ Equations 6 and 7 show that two adjacent matched filters The rotation angle difference (ΔΘ) of the output of the filter is Δα 〇ffNp. 1 After the above integer carrier frequency error correction, | is less than or equal to-π 9 so | Δλ is less than or equal to π. Calculate the output of two adjacent matched filters. Rotation angle difference (Δθ), then divide this rotation angle difference by L: \ EXT \ 52 \ 52160.DOC \ 4 · 13-This paper size applies to China National Standard (CNS) A4 (210X297 mm) Please read first Note on the back. Matters again-fill in this page A7 B7 V. Description of the invention (11) A pilot show frequency code sequence period, that is, a non-integer carrier frequency error (#, if Tc = 1 is expressed numerically), When the carrier frequency error correction of the number c on p, the direction flag can determine the calculation method of the rotation angle difference. If the direction flag is · 1 ^, the rotation angle difference is calculated in the 蝻 clock direction, ΔΘ is between 0 degrees and -360 degrees * If the direction flag is 1, < the rotation angle difference is calculated in the anticlockwise direction, ΔΘ is between 0 degrees and 360 degrees. If the direction flag is 0, the rotation direction is not determined. Calculate the rotation angle difference in the counterclockwise or clockwise direction but select ΔΘ between 0 degrees and 180 degrees or between 0 degrees and -180 degrees. False! This difference is t 邋 penweiwei 蠡 shiyang1 鍈, so there are many channels: 1: Sheng ϋfading. Less iadingj. In order to get a better estimate, we can choose several main paths to estimate the non-integer carrier frequency! difference. The system block diagram of non-integer carrier frequency error correction 115 is shown in Figure 4. • Enter ^ _ „and decide a gate punch, select J matched filters whose output amplitude is greater than the threshold sentence i | cheek diameter 4 to calculate the printed policy of the Central Standards Bureau of the Ministry of Economic Affairs-Industrial and Consumer Cooperatives (please read the note on the back first. Note the ears to fill in this page) The angle of a main path (dead / ·]) 4 〇 2 According to Fang Yiyi calculation-calculate the rotation angle difference of each main path (△ 呵 η) 403; calculate each path The rotation ^ turns J. The degree difference of M m is abandoned, (Here, handsome:, ·, and which average of each path, {Δ ^: / Ί} is Α No :,

—— ； I Δ^ = 1 iavg.mm (8) pj=l L:\EXT\52\52160.DOC\4 一 14 一本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐） B7 五、發明説明（12 ) 即是非整數載波頻率誤差估計値。最後輸出非整數載波頻率誤差控制信號給自動頻率控镛制105，激正非整數載波頻率誤差。輸出啓動信號用以啓動時脈修正。，c .展頻碼序列時脈修正經載波頻率誤差修正後，快速傅利葉轉換匹配滤波器的輸出振幅可寫|Μ((η-α))〜]|。|Μ((«-α))~]|是迴旋移動的頻道脈衝響應振幅。因爲迴旋移動的性質，展頻碼時脈不一定是匹配濾波器輸出的第一個主要路徑，所以需要搜尋頻道脈衝響應的起始點β我們假設頻道是一個最小相位移系統》最小相位移系統具有最小能量延遲的性質，所以頻道脈衝響應的主要路徑較集1k，即是其主要路徑分散範圍最小或延遲分散（delay spread)最小？展頻碼序列時脈修正116之系統方塊如圖5所示。經濟部中央標準局負工消費合作社印袋 (請先閲讀背面之注意事項再填寫本頁) 輸入並決定一門檻*若匹配濾波器輸出之振幅大於此門檻定義爲主要路徑501。然後假設每個主要路徑爲時脈起始點計算主要路徑分散範圍（DL/]) 502 *£>[/]的計算方法如T :定義以第 j個主要路徑爲起始點的一個週期，，此週期是由第j個主要路徑到匹配濾波器輸出的最後一黟，接著是匹配滤波器輸出的第一點到第j個主要路徑的前一點爲止。是以第j個主要路徑爲起始 L:\EXT\52\52160.DOC\4 — 15 — 本紙張尺度適用中國國家標準（CNS ) A4規格（2丨0X29*7公釐） A7 B7 五、發明説明（13 ) 點的一個週期中第一條主要路徑的位置和最後一個主要路徑的間隔。選擇最小的£)[/!爲-則第α個主要路徑爲頻道脈衝響應時脈起始點即是展頻碼序列時脈。 II .當領航信號不存在時 / 經濟部中央標準局員工消費合作社印聚 (請先閲讀背面之注·意事項再填寫本頁) 本發明之第二實施例是考慮領航信號不存在時的情況，當領航信號不存在時，我們則以資訊信號來估計載波頻率誤差及展頻碼序列時脈這裡的資訊調變假設爲BPSK a因爲資訊調變使得我們需要先得到展頻碼序列時脈〜才能修正載波頻率誤差《.圖6是第二實施例之整個載波頻率及時脈修正的流程圖"我們先以快速傅利葉轉換匹配濾波器來匹配接收信號並產生"有效輸出 "601。此"有效輸出"是指匹配濾波器的輸出振幅會產生尖峰（peak)，這個尖峰是因爲展頻碼匹配和載波頻率誤差調變而產生1^利用此"有效輸出" 搜尋展頻碼序列時脈602。展頻碼序列時脈修正後v搜尋整數載波頻率誤差603 _再以匹配濾波器輸出的相位旋轉來估計非整數載波頻率誤差 604，‘後又再一次搜尋展頻碼時脈605。 a J生"有效輸ά / 以快速傅利葉轉換匹配濾波器匹配接收信號來搜尋"有效輸出"。因爲分割區段的時脈未能 L:\EXT\S2\52160.DOC\4 ~~ 1 6 "" 本紙張尺度適用中國國家標準（CNS ) Α4規格（2丨0>< 297公釐）經濟部中央橾準局貝工消費合作社印製 A7 B7 五、發明説明（14 ) 和展頻碼序列時脈同步f所以資訊調變及載波頻率誤差皆會影響匹配濾波器輸出的振幅，假設資訊展頻碼是cd[n]，（：#]是cd[n]的FFT -接收信號r[n]是 r[ri] = dniri\cd[({n + α))Ν^χ^ω offn + χ} + ri[n] (々 drr^n] = d\ {t^ri\ - t^n - a]) + d2- (w[n -a]- i^n - Nd ]) 其中a是分割區段的起使點和資訊展頻碼序列的起使點的差値，是載波頻率誤差'，X是隨機相位，n[n]是雜訊· dm[n]是分割區段中前後兩個不同的資訊調變位元组成的序列，其中u [ η ]是步階函數，dl是區段中第一個位元> d2是第二個位元。定義 R[k]Cd[((k-i))Nd]的 IFFT 爲 CRi[n]。CRi[n]可寫爲 $ Ν-1 * CR,[n]= Σ dm[m]cd[((m + a))NJcd [((m + n))NJexp{yiy〇#/w-y^i((m + n))Nd + m = 0 d (10) 其中n|[n]是輸入雜訊所產生—當n[n]爲0時f CRi[n] 是 N — 1 CR,[a] = Nd Σ dm[m]exp{;(iy0# + χ) (11) m = 0 % d 依據方程式11並令i = 0及dl=-d2,將|C》[a]丨在a = 0， a = Νά卜反α = Ν」2採繪在截Ί。由圖7可得知，當 a = Ndl2i 〇)off = 0@ ， i》[-]丨爲〇 0 當 a =//d/2且 = ±祭時，^^爲0.636。由上述第一個例子可知，雖然——; I Δ ^ = 1 iavg.mm (8) pj = l L: \ EXT \ 52 \ 52160.DOC \ 4 -14 A paper size is applicable to China National Standard (CNS) A4 specification (210X297 mm) B7 Fifth, the invention description (12) is a non-integer carrier frequency error estimation 値. Finally, a non-integer carrier frequency error control signal is output to the automatic frequency control system 105, and the non-integer carrier frequency error is excited. The output start signal is used to start clock correction. C. Spreading code sequence clock correction After carrier frequency error correction, the output amplitude of the fast Fourier transform matching filter can be written as | M ((η-α)) ~] |. | M ((«-α)) ~] | is the amplitude of the channel impulse response of the convolutional movement. Because of the nature of the convolutional movement, the spreading code clock is not necessarily the first main path of the matched filter output. Therefore, we need to search for the starting point of the channel impulse response. We assume that the channel is a minimum phase shift system. It has the property of minimum energy delay, so the main path of the channel impulse response is more concentrated than 1k, that is, its main path has the smallest dispersion range or the smallest delay spread? The system block of the spread spectrum code sequence clock correction 116 is shown in FIG. 5. Central Government Bureau of Standards, Ministry of Economic Affairs, Printed Bags for Consumer Cooperatives (Please read the notes on the back before filling this page) Enter and determine a threshold * If the amplitude of the matched filter output is greater than this threshold, it is defined as the main path 501. Then assume that each main path is the starting point of the clock to calculate the main path dispersion range (DL /)) 502 * £> [/] The calculation method is T: Define a cycle with the j-th main path as the starting point , This period is from the j-th main path to the last step of the matched filter output, and then from the first point of the matched filter output to the previous point of the j-th main path. Starting with the j-th main path L: \ EXT \ 52 \ 52160.DOC \ 4 — 15 — This paper size applies the Chinese National Standard (CNS) A4 specification (2 丨 0X29 * 7 mm) A7 B7 V. Description of the invention The position of the first main path and the interval of the last main path in a period of (13) points. Choose the smallest £) [/! Is-then the first major path is the channel pulse response clock starting point which is the spreading code sequence clock. II. When the pilot signal does not exist / Printed by the Consumer Cooperatives of the Central Standards Bureau of the Ministry of Economic Affairs (please read the notes and notices on the back before filling out this page) The second embodiment of the present invention considers the situation when the pilot signal does not exist When the pilot signal does not exist, we use the information signal to estimate the carrier frequency error and the spreading code sequence clock. The information modulation here is assumed to be BPSK a. Because the information modulation makes us need to obtain the spreading code sequence clock first ~ Only the carrier frequency error can be corrected. "Figure 6 is a flowchart of the entire carrier frequency and clock correction in the second embodiment." We first use a fast Fourier transform matched filter to match the received signal and generate "effective output" 601. This "effective output" means that the output amplitude of the matched filter will produce a peak. This peak is caused by the spreading code matching and the modulation of the carrier frequency error. 1 ^ Use this " effective output " Frequency code sequence clock 602. After the spreading code sequence clock is corrected, v searches the integer carrier frequency error 603 _ and then uses the phase rotation output of the matched filter to estimate the non-integer carrier frequency error 604, ‘and then searches for the spreading code clock 605 again. a J Student " Effective Input / Match the received signal with a fast Fourier transform matched filter to search for " Effective Output ". Because the timing of the segmentation segment cannot be L: \ EXT \ S2 \ 52160.DOC \ 4 ~~ 1 6 " " This paper size applies the Chinese National Standard (CNS) Α4 specification (2 丨 0 > < 297 public (%) Printed by A7 B7, Shellfish Consumer Cooperative, Central Bureau of Standards, Ministry of Economic Affairs. 5. Description of the invention (14) and clock synchronization of the spread-spectrum code sequence. So information modulation and carrier frequency errors will affect the amplitude of the matched filter output. Suppose the information spreading code is cd [n], (: #] is the FFT of cd [n]-the received signal r [n] is r [ri] = dniri \ cd [({n + α)) N ^ χ ^ ω offn + χ} + ri [n] (々drr ^ n] = d \ {t ^ ri \-t ^ n-a]) + d2- (w [n -a]-i ^ n-Nd]) Where a is the difference between the starting point of the segment and the starting point of the information spreading code sequence. It is the carrier frequency error. X is a random phase. N [n] is the noise. Dm [n] is the partition. A sequence of two different information modulation bits in the segment, where u [η] is a step function, dl is the first bit in the segment, and d2 is the second bit. Define the IFFT of R [k] Cd [((k-i)) Nd] as CRi [n]. CRi [n] can be written as $ Ν-1 * CR, [n] = Σ dm [m] cd [((m + a)) NJcd [((m + n)) NJexp {yiy〇 # / wy ^ i ((m + n)) Nd + m = 0 d (10) where n | [n] is generated by input noise—when n [n] is 0, f CRi [n] is N — 1 CR, [a ] = Nd Σ dm [m] exp {; (iy0 # + χ) (11) m = 0% d According to equation 11 and let i = 0 and dl = -d2, set | C》 [a] 丨 in a = 0, a = Νά 卜反 α = Ν ″ 2 The plot is being intercepted. It can be known from FIG. 7 that when a = Ndl2i 〇) off = 0 @, i》 [-] 丨 is 〇 0 When a = // d / 2 and = ± offering, ^^ is 0.636. From the first example above, although

Nd - 沒有載波頻率誤差但是資訊調變會造成匹配濾波器輸出爲0。在第二個例子可得知若將輸入信 L:\EXT\52\52160.DOC\4 一 17 — 本紙張尺度適用中國固家標準（CNS ) A4規格（210X297公釐）； ^------οτ^-- (請先閲讀背面之注-意事項#填寫本瓦) 訂 A7 五、發明説明（15 ) 號和中心頻率爲±^L之資訊展頻碼匹配良，其輸出振幅會有0.636。所以我們可以得私，輸入信號和不同中心頻率資訊展頻碼匹配後，在某一次的匹S&輸出會產生尖峰。當dl=d2時，其狀況和實施例I (領航信號存在時）相同，所以必能產生尖峰〇將R[k]和（：/[(&-〇)&]相乘後儲存在記憶體經IFFT運算後計算其振幅。搜尋IFFT運算後的最大振幅並記錄爲Ap【i]。在可能的載波頻率誤差範圍内，改變不同的i並重複上述運枭。以/c=2.4Gfe ，+ = ，' =1024爲例，若載波頻率誤差在10Nd-There is no carrier frequency error but information modulation will cause the matched filter output to be zero. In the second example, we can know that if the input letter is L: \ EXT \ 52 \ 52160.DOC \ 4 -17 — This paper size is applicable to the Chinese solid standard (CNS) A4 specification (210X297 mm); ^ --- --- οτ ^-(Please read the note on the back-Issue #Fill this tile first) Order A7 V. Invention Description No. (15) and the information spread code whose center frequency is ± ^ L is good, and its output amplitude There will be 0.636. So we can get private. After the input signal matches with the spreading code of different center frequency information, the S & output of a certain time will produce a spike. When dl = d2, the condition is the same as that in Example I (when the pilot signal is present), so a peak must be generated. R [k] and (: / [(& -〇) &] are multiplied and stored in The memory calculates its amplitude after IFFT operation. Search the maximum amplitude after IFFT operation and record it as Ap [i]. Within the range of possible carrier frequency error, change different i and repeat the above operation. Take /c=2.4Gfe , + =, '= 1024 as an example, if the carrier frequency error is 10

1C ppm以下，則需要搜尋i的範圍是{-3, -2, -1，0, 1，2, 3}。比較Ap[i]並選擇最大者是爲ΑΡ„^，則R[k]和 Cd*[((k-〇)NJ匹配後的輸出是”有效輸出"，在a步驟中得到的••有效輸出"，提供給b步驟用以第一次時脈修1。其系統方塊圖與圖3所示者相。經濟部中央標準局—工消费合作社印製 (請先閲讀背面之注•意事項'再填寫本頁) b . 第一次時脈修正經a步驟後，匹配濾波器會產生"有效輸出" 。"有效輸出"的尖峰仍是迴旋移動後的主要路徑。所以時脈修正的原理及方法都和上述領航信號存在時相同•·只是改爲以在a步驟得到的" 有效輸出"來修正時脈。其系統方塊圖如圖5所示。 L:\EXT\52\52160.DOC\4 — 18 — 本紙張尺度適用中國國家標準（CNS ) Α4規格（210Χ297公釐） A7 B7 五、發明説明（ 16 經濟部中央標準局員工消費合作社印製 C .整數載波頻率誤差修正經b步驟後，資訊展頻碼時脈和分割區段時脈已經同步，所以資訊調變不會影響匹配濾波器輸出振幅，《0此整數載波頻率誤差修正的方法和原理和上述領航信號存在時相同·" /‘ 在c步驟中修正整數載波頻率誤差，並且產生方向旗標提供給d步驟使用。因爲整數載波頻率誤差修正和a步驟產生"有效輸出"的方法相似，都·是搜尋及比較匹配遽波器輸出之最大振幅，所以整數載波頻率誤差修正和產生"有效輸出 "可以共用同一子系統，其系統方塊圖與圖3所示者相同。 d.銮整數載波頻率誤差修正非整數載波頻率誤差是估計非整數載波頻率誤差造成相鄰之匹配濾波器輸出的旋轉角度差（ΔΘ)而得#但是我們只能量測到相鄰之匹配濾波器輸出的旋轉角度差（△(}>)，又資訊調變會造成 △φ和ΔΘ有180度模糊量，因此我們需要修正Δφ來估計雄。依據方程式11，假設相鄰兩個匹配滤波器輸出之第一個輸出的第α點如下_(已知整數載波頻率誤差爲 Nd N-1 CR0[a] = i/l-^d Σ exp{jAmoffm}&^{-j^la +χ] ，第二個輸出的第α點則是 (12) ---^------cr^~------ΐτ----- (請先閲讀背面之注意事項'再填寫本頁) L:\EXT\52\52160.DOC\4 19 - 本紙張尺度適用中國國家標準（CNS ) Α4規格（210Χ297公釐）經濟部中央標隼局貝工消費合作社印裝 A7 B7 五、發明説明（n ) N-1 CRQ[a]^d2 Nd Σ exp{j^〇Fm}exp{-j(j^ia-Αω0^Νά) +χ} (13) m = Ο d 其中dl是第一個輸出的資訊位元，d2是第二個輸出的資訊位元。*雖然經上述整數載波頻率誤差修正後，小於或等於π，但是當 dl = -d2時，_ Δφ等於Δθ + π或Δθ - π «因爲以反正切函數計算角度可解析的角度只有360度ν所以Δθ + π或 △θ-π認爲是相同的角度。我們仍是依據方向旗標計算匹配濾波器輸出的旋轉角度差#然後修正資訊調變造成的180 度模糊量得到非整數載波頻率誤差造成的旋轉角度差。當方向旗標是1時，#整數載波頻率誤差造成匹配濾波器輸出反時鐘方向旋轉。因爲Δφ及ΔΘ皆是正値，將符號Δφ及ΔΘ分別改爲 Δφ+及ΔΘ+。以反時鐘方向計算Δφ*，且Δφ+是介於0 度到360度之間。Δφ+可能是ΔΘ+或Δθ+ + π &爲了解決資訊調變造成的180度模糊量我們可以利用當方向旗標爲1時ΔΘ+是大於90度且小於270度的性質，依據表一修正Δφ+爲ΔΘ+。表一 △Φ+ Δθ+ 0 〜0.5π Δφ+ + π 0.5π 〜1.5π Δφ+ 1.5π 〜2π △φ+ - π L:\EXT\52\52160.DOC\4 _ 2 0 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐） (請先閲讀背面之注意事項再填寫本頁) .-ο-裝· 、1Τ A 7 __^_B7_ 五、發明説明（18 ) 當方向旗標是-ι時，非整數載波頻率誤差造成匹配渡波器輸出順時鐘方向旋轉，.。因爲△<(>及 △Θ皆是負値_將符號△({>及ΔΘ分別改爲Δ(|Γ及ΔΘ·。以順時鐘方向計算Δφ_ π且△(()-是介於〇度到-360度之間。<Λφ-可能是ΔΘ-或Δθ· - π。·爲了解決資訊調變ΐέ 成的180度模糊量，我們可以利用當方向旗標爲-1時ΔΘ·是小於-90度且大於-270度的性質，依據表二修正Δφ·爲ΔΘ·。 (請先閱讀背面之汰意事項再填寫本頁) 經濟部中央標準局員工消費合作社印聚 L:\EXT\S2\S2160.DOC\4 -21 本紙張尺度適用中國國家標準（CNS ) Α4规格（210Χ297公釐）經濟部中央標準局貝工消費合作社印製 A7 B7 五、發明説明（19 ) 表二 Δφ' ΔΘ' 0 ~ 0.5π Δφ' - π -0.5π ~ - 1·5π Δφ' -1.5π 〜-2π Δφ" + π 當方向旗標及爲〇時 <旋轉方向未定，但其非整數載波頻率誤差造成的旋轉角度差的絕對値小於180度（|ΔΘ|<180)。因此需要以順時鐘方向估計旋轉角度差（△<)〇及反時鐘方向估計旋轉角度差 (△φ+)«^Δ(|Γ是介於0度到-360度，Δφ+是介於0度到360 度》:我們假設ΔΘ+是介於0度到180度，ΔΘ·是介於0 度到-180度。若Δφ+大於180度或Δφ_小於-180度時，則需要修正資訊調變造成180度模糊量《當Δφ+大於180度時則ΔΘ+等於Δφ+ - 7Γ 。當Δφ_小於-180度時則 △Θ—等於△<})·+ «« 表三描述 Af，△<()·，ΔΘ+ 及 ΔΘ·間的關係。由表三我們可得知Δφ-等於Δφ+ - 2 7Γ且ΔΘ·等於Δθ+ - π 。所以當方向旗標爲0時*我們只要先估計ΔΘ+，然後再將ΔΘ+減去180就可得到Δθ_ ” 表三 Δφ+ Δφ=Δφ+-2π Δθ+ ΔΘ =Δ θ +-π 0〜π -2π —π △φ+ Δφ+-π π〜2π -π〜0 Δφ+-π Δφ+-2π 假設此系線是f邋分磉進接4镕，所以頻道有多重路徑衰減現象（multipath fading)。爲了得到 L:\EXT\S2\52160.DOC\4 22 ~ 本紙張尺度適用中國國家梯準（CNS ) A4規格（210X297公釐）； Q-私衣------ΐτ-----Αν t請先閱讀背面之注·意事項再填寫本頁) A7 B7 五、發明説明（2〇 ) 較佳的估計，我們可以選擇幾個主要路徑來估計非整數載波頻率誤差，其系砗方塊圖...如圖8所示 0 輸入並選擇P個匹配濾波器輸出振幅大於門檻的主要路徑801，計算每一主要路徑的角度（φ[ΐ])802，然後依據方向旗標計算每一個主要路徑的旋轉角度差803並修正之8似。隨後計算每一主要路徑之旋轉角度差的時間平均805 : 當方向旗標爲1時+計算每一主要路徑之旋轉角度差的時間平均（αν&{ΔΘ+[7·]})。平均每一路徑的 ovg,{M+L/]}爲 Δ歹+ : Α~θ+ =^avgt{A0+[j]} (14) >=1 」心/#,即是非整數載波頻率誤差估計値。當方向旗標爲-1時，計算每一主要路徑之旋轉角度差的時間平均（αν&{Δθυ}) *平均每一路徑的爲 Δ》-： Δθ' =Υ^^{ΔΘ~Ιί]} (15) 經濟部中央標準局貝工消费合作社印製 (請先閲讀背面之；>χ·意事項4·填寫本瓦) 7=1 df/Λ^即是非整數載波頻率誤差估計値。當方向旗標不爲0時，輸出非整數載波頻率誤差控制信號給自動頻率控制105，修正非整數載波頻率誤含。輸出啓動信號用以啓動時脈修正〇 L:\EXT\52\S2160.DOC\4 一 ·23 — 本紙張尺度適用中國國家標準（CNS ) Α4規格（210Χ297公釐） A7 B7 五、發明説明（21 ) 當方向旗標爲〇時，玖反時鐘方向計算每一個主要路徑的旋轉角度差並修正之 <計算每一路徑之旋轉角度差的時間平均（αν&{ΔΘ+(7·)})。平均每一路徑的αν&{ΔΘ+(/)丨爲△歹+。順時鐘方向的旋轉角度差估計 (2lf)爲-π。順時鐘方向及反時鐘方向的非整▲ 載波頻率誤差估計分別爲Z^/乂和。這兩個估計値只有一個是正確，我們以試誤法決定正確的估計値。比較兩個非整數載波 ' ·* -»r 一·- .λ. 1 一一' 頻率誤差修正後的匹配慮波器輸出的最大振幅振，具有較大匹配濾波器輸出的最大振幅振者，其對應之非整數載波頻率誤差爲正確估計。先將輸入自動頻率修正器修正載波頻經濟部中央標準局負工消費合作社印裝 (請先閲讀背面之·ίϊ意事項#填寫本頁) 率誤差，記錄修正後匹配；慮波器輸出的最大振幅（^二）。再將輸入自動頻率修正器修正頻率，記錄修正後匹配濾波器輸出的最大振幅 (必_)。比較^«及^，選擇較大者並以其對應之非整數載波頻率誤差爲正確估計。輸出非整數載波頻率誤差控制信號給自動頻率修正器、，修正非整數載波頻率誤差。輸出啓動信號用以啓動時脈修正。 e .第二次時脈修正在第一次時脈修正時，可能因爲載波頻率誤差尚未修正造成匹配濾波器輸出信號的SNR (信號雜訊比）衰減，使得時脈修正不正確，所以再進行第二次時脈修正。在d步驟後，載波頻率誤差 L:\EXT\52\S2160.DOC\4 ~ 24 ~ 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐〉五、發明説明（ 22 A7 B7 好如敎圍較。i示及範到同揭示之得相份揭明以驟充之發可步已明本正h 容發脱修^内本不脈方術於均時的技基換次正及士替一修徵之及行脈特術改進時術技修再。技項種 e 計之本各正估明於之修脈發m.作經時本，所己的上示。；---一------------ir-----om. ·(請先閲讀背面之注·意事項#'填寫本頁) 經濟部中央標準局員工消費合作社印製本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐）Below 1C ppm, the search range of i is {-3, -2, -1, 0, 1, 2, 3}. Compare Ap [i] and select the largest one as AP. ^, Then the output after R [k] and Cd * [((k-〇) NJ match is "effective output", obtained in step a •• The effective output " is provided to step b for the first clock repair 1. Its system block diagram is similar to that shown in Figure 3. Printed by the Central Standards Bureau of the Ministry of Economic Affairs-Industrial and Consumer Cooperatives (please read the note on the back first) Note: Please fill in this page again.) B. After the first clock correction through step a, the matched filter will generate "effective output". The peak of "effective output" is still the main path after the convolution movement. Therefore, the principle and method of clock correction are the same as when the pilot signal mentioned above exists. It is only changed with the "effective output" obtained in step a to correct the clock. The system block diagram is shown in Figure 5. L: \ EXT \ 52 \ 52160.DOC \ 4 — 18 — This paper size applies to the Chinese National Standard (CNS) A4 specification (210 × 297 mm) A7 B7 V. Description of the invention (16 Printed by the Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs. Integer carrier frequency error correction After step b, the information spreading code clock and partition The clock is synchronized, so the information modulation will not affect the output amplitude of the matched filter. "0 The method and principle of correcting the integer carrier frequency error is the same as the above pilot signal exists. &Quot; / 'Correct the integer carrier frequency in step c Error, and the direction flag is provided for use in step d. Because the integer carrier frequency error correction is similar to the method of step a to generate "effective output", both are to search for and compare the maximum amplitude of the matched wave output, so the integer Carrier frequency error correction and generation " effective output " can share the same subsystem, and its system block diagram is the same as that shown in Figure 3. d. 銮 Integer carrier frequency error correction Non-integer carrier frequency error is an estimate of non-integer carrier frequency error Causes the rotation angle difference (ΔΘ) of the output of the adjacent matched filter to obtain # But we can only measure the rotation angle difference (△ () >) of the output of the adjacent matched filter, and the information modulation will cause Δφ and ΔΘ have 180 degrees of ambiguity, so we need to modify Δφ to estimate the male. According to Equation 11, assuming two adjacent matches The alpha point of the first output of the wave filter output is as follows: (The known integer carrier frequency error is Nd N-1 CR0 [a] = i / l- ^ d Σ exp {jAmoffm} & ^ {-j ^ la + χ], the α point of the second output is (12) --- ^ ------ cr ^ ~ ------ ΐτ ----- (Please read the notes on the back first 'Refill this page) L: \ EXT \ 52 \ 52160.DOC \ 4 19-This paper size applies to the Chinese National Standard (CNS) A4 specification (210 × 297 mm). Printed by A7, Shellfish Consumer Cooperative, Central Bureau of Standards, Ministry of Economic Affairs B7 V. Description of the invention (n) N-1 CRQ [a] ^ d2 Nd Σ exp {j ^ 〇Fm} exp {-j (j ^ ia-Αω0 ^ Νά) + χ} (13) m = Ο d where dl is the first output information bit, and d2 is the second output information bit. * Although it is less than or equal to π after the correction of the integer carrier frequency error, when dl = -d2, _ Δφ is equal to Δθ + π or Δθ-π «because the arc tangent function can only calculate the angle that can be resolved by 360 ° ν So Δθ + π or Δθ-π is considered the same angle. We still calculate the rotation angle difference # from the matched filter output based on the direction flag, and then correct the 180 degree blur caused by the information modulation to obtain the rotation angle difference caused by the non-integer carrier frequency error. When the direction flag is 1, the #integer carrier frequency error causes the matched filter output to rotate counterclockwise. Because Δφ and ΔΘ are positive 値, the symbols Δφ and ΔΘ are changed to Δφ + and ΔΘ +, respectively. Δφ * is calculated in the counterclockwise direction, and Δφ + is between 0 and 360 degrees. Δφ + may be ΔΘ + or Δθ + + π & In order to solve the 180-degree blur caused by information modulation, we can use the property that when the direction flag is 1, ΔΘ + is greater than 90 degrees and less than 270 degrees, according to Table 1. The correction Δφ + is ΔΘ +. Table 1 △ Φ + Δθ + 0 ～ 0.5π Δφ + + π 0.5π ～ 1.5π Δφ + 1.5π ～ 2π △ φ +-π L: \ EXT \ 52 \ 52160.DOC \ 4 _ 2 0 This paper size is applicable China National Standard (CNS) A4 specification (210X297 mm) (Please read the precautions on the back before filling out this page) .-ο-installation · 1T A 7 __ ^ _ B7_ V. Description of the invention (18) When the direction flag When it is -ι, the non-integer carrier frequency error causes the matching wavelet output to rotate clockwise. Because △ < (> and △ Θ are both negative 値 _ change the symbols △ ((> and ΔΘ to Δ (| Γ and ΔΘ ·, respectively. Calculate Δφ_π in a clockwise direction and Δ (()-is the median Between 0 ° and -360 °. &Lt; Λφ- may be ΔΘ- or Δθ ·-π. To solve the 180-degree blurring amount caused by information modulation, we can use ΔΘ when the direction flag is -1 · It is less than -90 degrees and greater than -270 degrees. Correct Δφ · as ΔΘ · according to Table 2. (Please read the notice on the back before filling out this page.) Staff Consumer Cooperatives, Central Standards Bureau, Ministry of Economic Affairs, Printed Collection L: \ EXT \ S2 \ S2160.DOC \ 4 -21 This paper size applies to the Chinese National Standard (CNS) Α4 size (210 × 297 mm) A7 B7 printed by the Bayer Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs 5. Description of the invention (19) Two Δφ 'ΔΘ' 0 ~ 0.5π Δφ '-π -0.5π ~-1 · 5π Δφ' -1.5π ～ -2π Δφ " + π When the direction flag is 0, the direction of rotation is not determined, but it is not The absolute value of the rotation angle difference caused by the integer carrier frequency error is less than 180 degrees (| ΔΘ | < 180). Therefore, the rotation angle difference (△ <) 〇 and counterclockwise estimated rotation angle difference (△ φ +) «^ Δ (| Γ is between 0 degrees and -360 degrees, Δφ + is between 0 degrees and 360 degrees": we assume ΔΘ + is Between 0 degrees and 180 degrees, ΔΘ · is between 0 degrees and -180 degrees. If Δφ + is greater than 180 degrees or Δφ_ is less than -180 degrees, it is necessary to modify the information modulation to cause a 180 degree blur when "Δφ + When it is greater than 180 degrees, ΔΘ + equals Δφ +-7Γ. When Δφ_ is less than -180 degrees, ΔΘ— equals Δ <}) · + «« Table 3 describes Af, △ < () ·, ΔΘ + and The relationship between ΔΘ ·. From Table 3 we can know that Δφ- is equal to Δφ +-2 7Γ and ΔΘ · is equal to Δθ +-π. So when the direction flag is 0 * we only need to estimate ΔΘ + first, and then ΔΘ + Subtract 180 to get Δθ_ ”Table 3 Δφ + Δφ = Δφ + -2π Δθ + ΔΘ = Δ θ + -π 0 ~ π -2π —π △ φ + Δφ + -π π ~ 2π -π ~ 0 Δφ + -π Δφ + -2π Suppose that this line is f 邋 points and 4 points, so the channel has multipath fading. In order to get L: \ EXT \ S2 \ 52160.DOC \ 4 22 ~ this Paper size applies to China National Standard (CNS) A4 (210X297 mm) ; Q-Private clothing ------ ΐτ ----- Αν t Please read the notes and notices on the back before filling out this page) A7 B7 V. Description of the invention (2) Better estimates, we can Select several main paths to estimate the non-integer carrier frequency error, which is a block diagram ... As shown in Figure 8 0 Enter and select P main paths 801 whose output amplitude is greater than the threshold. Calculate the Angle (φ [ΐ]) 802, and then calculate the rotation angle difference 803 of each main path according to the direction flag and correct it. Then calculate the time average 805 of the rotation angle difference of each main path: when the direction flag is 1 + calculate the time average of the rotation angle difference of each main path (αν & {ΔΘ + [7 ·]}). The average ovg of each path, (M + L /)} is Δ 歹 +: Α ~ θ + = ^ avgt (A0 + [j]} (14) > = 1 '' heart / #, which is a non-integer carrier frequency error Estimate 値. When the direction flag is -1, calculate the time average of the rotation angle difference of each main path (αν & {Δθυ}) * The average of each path is Δ》-: Δθ '= Υ ^^ {ΔΘ ~ Ιί] } (15) Printed by the Shellfish Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economic Affairs (please read the back; > χ · Matters 4 · Fill in the tile) 7 = 1 df / Λ ^ is the non-integer carrier frequency error estimate 値. When the direction flag is not 0, a non-integer carrier frequency error control signal is output to the automatic frequency control 105, and the non-integer carrier frequency error is corrected. The output start signal is used to start the clock correction 〇L: \ EXT \ 52 \ S2160.DOC \ 4 I · 23 — This paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 × 297 mm) A7 B7 V. Description of the invention ( 21) When the direction flag is 0, calculate the rotation angle difference of each main path in the counterclockwise direction and modify it < calculate the time average of the rotation angle difference of each path (αν & {ΔΘ + (7 ·)} ). The average αν & {ΔΘ + (/) 丨 of each path is Δ 歹 +. The clockwise rotation angle difference estimate (2lf) is -π. Non-integral in clockwise and anticlockwise directions ▲ Carrier frequency error estimates are Z ^ / 乂 and. Only one of these two estimates is correct. We use trial and error to determine the correct estimate. Compare two non-integer carriers '· *-»r a ·-.λ. 1 one one' with the maximum amplitude vibration of the matched filter output after frequency error correction, and the maximum amplitude oscillator with a large matched filter output, The corresponding non-integer carrier frequency error is a correct estimate. First, the input automatic frequency corrector is used to correct the carrier frequency. It is printed by the Central Laboratories of the Ministry of Economic Affairs and the Consumer Cooperatives (please read the ϊ ϊ 意事 # on the back page). The error rate is recorded, and the correction is recorded. The maximum output of the wave filter is considered. Amplitude (^ 二). Then input the input automatic frequency corrector to correct the frequency, and record the maximum amplitude (required) of the matched filter output after the correction. Compare ^ «and ^, choose the larger one and use its corresponding non-integer carrier frequency error as the correct estimate. Output the non-integer carrier frequency error control signal to the automatic frequency corrector to correct the non-integer carrier frequency error. The output start signal is used to start clock correction. e. The second clock correction. During the first clock correction, the SNR (signal-to-noise ratio) of the matched filter output signal may be attenuated because the carrier frequency error has not been corrected. Therefore, the clock correction is incorrect. Second clock correction. After step d, the carrier frequency error L: \ EXT \ 52 \ S2160.DOC \ 4 ~ 24 ~ This paper size is applicable to the Chinese National Standard (CNS) A4 specification (210X297 mm) 5. The description of the invention (22 A7 B7 is as good as敎示较. I show and Fan Da and the disclosed have revealed that the step can be made with a sudden charge, and the content can be repaired h. The internal basic pulse method is changed at the same time. The scholars repaired the technique and improved the technique of the pulse special technique. The types of the technical items are being estimated in the training of the pulse m. As the script of the time, what they have shown above; --- -一 ------------ ir ----- om. · (Please read the note on the back · Issue # 'to complete this page) Paper size applies to China National Standard (CNS) A4 (210X297 mm)

Claims

經濟部中央揉準局貝工消背合作社印策 AS B8 C8 D8 々、申請專利範圍 1. 一種於直接序列展頻通信系統中修正Hg 率誤差之方法，包括―： ^ 將接收信號匹配中心頻率爲一個展頻碼序列週期倒數之整數倍的該展頻碼序列； .搜尋匹配結果中振輻最大者，以得出..太小爲該展頻碼序列週期倒數之某一整數倍的載波頻率誤差；將搜尋得出之該·整數的匹配結果與其前後整 r 數之匹配結果作比較，以定出方向旗標；依據該方向旗標計算出相鄰兩個匹配結果之旋轉角度差；以該旋轉角度差計算出非整數載波頻率誤差 ;以及_ _ 修正該4數倍及非整數之載波頻率誤差。 2. 如申請專利範圍第1項之方法，其中該展頻碼序列爲一已知之領航展頻碼序列。 3. 如申請專利範圍第1項之汸法，其中該展頻碼序列爲資訊展頻碼序列，且該匹配步驟之前更包括：搜尋該接收信號與該資訊展頻碼序列匹配得出，之尖峰；以及以具有最小延遲分散範圍之該尖峰修正該資訊展頻碼序列時脈β 4. 如申請專利範圍第1項之方法，其中該匹配係 L:\EXT\52\52160.DOC\S — 26 — 本紙張Λ度適用中國國家揉率（CNS ) Α4规格（210Χ297公釐） (婧先W讀背面之注意事項再填寫本頁) 訂 .-ο 六、申請專利範圍利用一快速傅利葉轉換匹配滤波器來處理。 5. 如申請專利範圍第1或3項之方法，其中該定出方向旗標之步驟中，當於該整數之匹配結果接近於大於該整數之相鄰整數的匹配結果 .時，該方向旗標係被設，定表示逆時鐘方向；/ 當於該整數之匹配結果較接近於小於該整數之相鄰整數的匹配結果時，該方向旗標係被設定表示順時鐘方向；若非上述兩種狀況時，則，設定方向旗標爲零。 6. 如申請專利範圍第5項之方砝，其中、該y計算旋轉角度差之步驟中，當該方向旗標爲零時，可依逆時鐘方向或順時鐘方向計算旋轉角度差，但計算所得之旋轉角度差須介於0度至180 度或介於0度至-180度。 7. 如申請專利範圍籮4或3硬鈦方祛，其中該計算非整數之載波頻率誤差的步驟更包括：將該旋轉角度差除以該展頻碼序列之長度。經濟部中央揉準局男工消費合作社印裝 (請先聞讀背面之注意事項再填寫本頁) 8. 如申請專利範圍第1或3項之方法，其中該計算非整數之載波頻率誤差的步驟更包括：定義匹配得出之振輻大於一預定門檻値者爲主要路徑；計算每一主要路徑之旋轉角度差；計算該等旋轉角度差的平均値；以及將該平均値除以該展頻碼序列之長度。 L:\EXT\52\52160.DOC\5 27 - 本紙張尺度適用中，國家標準（CNS ) A4规格( 210X297公釐）六、申請專利範圍 9. 如申請專利範圍第1或3項之方法，其中於該定出方向旗標之步驟後，即修正該整數倍之載波頻率誤差。 10. —種於直接序列展頻通信系，统中修正載波頻 .率誤差及展頻碼序列時脈之方法，包括：，將接收信號匹配中心頻率爲一個展頻碼序列週期倒數之整數倍的該展頻碼序列；搜尋匹配結果中振輻最大者，以得出大小爲該展頻碼序列週期倒數之某一整倍的載波頻率誤差；將搜尋得出之該整數的匹配結果與其前後整數之匹配結果比較，以定出方向旗標；依據該方向旗標計算出相鄰兩個匹配結果之旋轉角度差；以該旋轉角度差計算出非整數載波頻率誤差修正該整數倍及非整數之載波頻率誤差；以及以匹配得出之振輻的延遲分散範圍最小，者修正展頻碼序列時脈。 11. 如申請專利範圍第10項之方，其中該展頻碼序列爲一已知之領航展頻碼序列。， 12. 如申請專利範圍第10項之方法，其中該展頻碼序列爲資訊展頻碼序列，且於該匹配步驟 L:\EXT\52\S2160.DOC\S 本紙張尺度逋用中國國家揉準（CNS ) A4規格（？1〇Χ297公釐）請先閱讀背面之注意· 事項再- i 經濟部中央標準局爲工消費合作社印裝々、申請專利耗圍之前更包括：搜尋該接收信號與該資訊展頻碼序列匹配得出之尖峰；以及以具有最小延遲分散範圍之該尖峰修正該資 .訊展頻碼序列時脈。 13. 如申請專利範圍第10或12項之..方法，其…中該匹配係利用一快速傅利葉轉換匹配濾波器來處理。. 14. 如申請專利範圍第10或12項之方法，其中該定出方向旗標之步驟中，當於該整數之匹配結果較接近於大於該整數之相鄰整數的匹配結果時，該方向旗標係被設定表示逆時鐘方向 ;當於該整數之匹配結果較接近於小於該整數之相鄰聱數的匹配結果時，該方向旗標係被設定表示順時鐘方向；若非上述兩種狀況時，則設定方向旗標爲零。經濟部中央揉準局貝工消费合作社印装 (請先S讀背面之注意Ϋ項再填寫本頁) 15·如申請專利範圍第14項之方法，其中該計算旋轉角度差之步驟中，當該方向旗標爲零時，可依逆時鐘方向或順時鐘方向計算旋轉角度差，但計算所得之旋轉角度差須介於0度至 180度或介於0度至-180度。 16.如申請專利範圍第10或12項之方法，其中該計算非整數之載波頻率誤差的步驟更包括：將該旋轉角度差除以該展頻碼序列之長度。 L:\EXT\52\52160.DOC\5 _ 29 — 本紙張尺度逋用中國國&準（CNS ) A4規格（2l〇X297公釐) 六、申請專利範圍一搜尋裝置，接收該匹配裝置輸出之振輻，並據以決定出大小爲該展頻碼序列週期倒數之某一整數倍的載波頻率誤差；一比較裝置，用以比較搜尋得出之該整數與 .其前後整數的匹配結果，以定出方向旗標；’ 一第一計算裝置，用以依據.該方向旗標計算出該匹配裝置兩個.相鄰輸出的旋轉角度差，並利用該旋轉角度差計算得出非整數載波頻率誤差；一第一修正裝置，用以將m搜尋裝置及該第一計算裝置之輸出反饋至該匹配裝置，以修正該整數倍及非整數載波頻率誤差；一第二計算裝置，接收已修正載波頻率誤差之該匹配裝置的輸出，並計算出輸出振輻之延遲分散範圍最小者以決定展瑪序列時脈 ;以及經濟部中央標隼局貞工消费合作社印裝 (請先W讀背面之注意事項再埃寫本頁) 一第二修正裝置，用以將該第二計算黎置之輸出反餚至該匹配裝置，以修正展頻碼序列時脈。 21. 如申請專利範圍第20項之接收機，其中該展 '頻碼序列爲一已知之領航.展頻碼序列。 22. 如申請專利範圍第20項之接收機，其中該展頻碼序列爲資訊展頻碼序列；該搜尋裝置亦用以搜尋該匹配裝置之輸出振輻中的尖峰； L:\EXT\52\52160.DOC\5 — 3 1 — 本紙張尺度適用中國國家標準（CNS ) A4«L格（210X297公釐）々、申請專利範圍 17. 如申請專利範圍第10或12項之方法，其中該計算非整數之載波頻率誤差的步骤更包括：定義匹配得出之振輻大於一預定門檻，値得主要路徑； ,計算每一主要路徑之旋轉角度差； ^計算該等旋轉角度差的平均値；以及將該平均値除以該展頻碼序列之長度。 18. 如申請專利範圍第10或12項之方法，其中該修正展頻碼序列時脈之步驟更包括定義匹配得出之振輻大於一預定門檻値者爲主要路徑；分別以各個主要路徑爲起始點計算全部主要路徑之延遲分散範圍；以及，以最小延遲分散範圍之起始主要路徑作爲該展頻碼序列時脈之起始點。 19. 如申請專利範圍第10或12項之方法，其中於該 -V ... I 經濟部中央標準局負工消费合作社印製 (請先聞讀背面之注意事項再填寫本頁) '定出方向旗標之步驟後，即修正該整數倍之載波頻率誤差。 20. —種用鲶i揍皋斗展逛通隹系ϋ接I邋， t 包括有天線及接收模组，其特徵在_於該接收機包括：一匹配裝置，用以將接收信/號_匹配中心頻率爲一個展頻碼序列週期倒數之整數/倍I的該展頻碼_^列； L:\EXT\52\52160.DOC\£ — 3 0 " 本紙張尺度適用中國國家揉準（CNS ) A4規格（2H)X297公釐） A8 B8 C8 D8 、申請專利範圍以及該第二計算裝置亦用以計算出該等尖峰之延遲分散範圍最小者以決定該資訊展頻碼序列時脈。 23. 如申請專利範園第20或益J之基验機，其中該 - — J J .匹釔裝置爲一快速傅利葉轉換匹配濾波器。 24. 如申請專利範圍第20或22項之接收機，其中該搜尋裝置包括：控制裝置，用以移動該展頻碼序列之中心頻率；以及選擇裝置，用以選出該匹配裝置輸出振輻中之最大者。 25. 如申請專利範圍第21項之接收機，其中該第一計算装置包括：決定主要路徑之裝置，用以判定該匹配裝置之輸出振輻大於一門檻値者爲主要路徑；旋轉角度差計算裝置，用以依據該方向旗標計算各主要路徑之旋轉角度差；平均裝置，用以計算該等旋轉角度差之平均値；以及非整數載波頻率誤差計算裝置，用以將該平均値除以該展頻碼序列之長度，以得出非整數載波頻率誤差。 _ 26. 如申請專利範圍第22項之接收機，其中該第一計算裝置包括： L:\EXT\52\S2160.DOC\5 本紙張尺度逍用中國國家揉準（CNS ) Α4规格（210X297公釐）請先 Η 背面之注項再- 填寫本頁經濟部中央標準局身工消費合作社印装 -32 - A8 B$ C8 D8 六、申請專利範圍決定主要路徑之裝置，用以判定該匹配裝置之輸出振輻大於一預定之門摇値者爲主要路徑；旋轉角度差計算裝置，用以依據該方向旗標計算各主要路徑之旋轉角度差； / 旋轉角度差修正裝置，用以-修正該計算，得出之旋轉角度差中因資訊調變所造成之模糊量 > 平均裝置，用以計算該等修正後之旋轉角度差的平均値；以及非整數載波頻率誤差計算裝置，用以將該平均値除以該展頻碼序列之長度，以得出非整數載波頻率誤差。 27·如申請專利範圍考趄4茲頊之揸氇，其中該第一修正裝置係於該第一計算裝置啓動前即修正該整數倍載波頻率誤差。經濟部中央標準局貞工消费合作社印製 (請先《讀背面之注t事項毛填寫本頁) 28.如申請專利範圍第20或22項之接收機，其中該第二計算裝置包括：決定主要路徑之裝置，用以判定該匹配裝置之輸出振輻大於一門檻値者爲主要路徑；計算最小延遲分散範圍之裝置，其分別以各主要路徑爲起始點計算全部主要路徑之延遲分散範圍；以及時脈決定裝置，用以將最小延遲分散範^圍之 L:\EXT\52\52160. DOC\5 — 33 — 本紙張尺度適用中國國家揉準（CNS ) A4规格（210X297公釐） A8 B8 C8 D8 申請專利範圍起始主要路徑作爲該展頻碼序列時脈之起始點。』請先閩之注事· 項裝訂經濟部中央標準局属工消費合作社印«. -34 - L:\EXT\52\S2160.DOC\5 本紙張尺度適用中國國家標準（CNS ) A4规格（2!〇X297公釐）The Central Bureau of the Ministry of Economic Affairs of the Central Government Bureau of Beige Consumers Cooperative Co., Ltd. AS B8 C8 D8 々, patent application scope 1. A method of correcting Hg rate error in direct sequence spread spectrum communication system, including: ^ matching the received signal to the center frequency The spreading code sequence which is an integer multiple of the period of the reciprocal of the spreading code sequence;. Search for the one with the largest amplitude in the matching result to get .. too small a carrier that is an integer multiple of the reciprocal of the period of the spreading code sequence Frequency error; Compare the matching result of the · integer obtained by searching with the matching result of the integer r number before and after to determine the direction flag; calculate the difference in the rotation angle between the two adjacent matching results based on the direction flag; A non-integer carrier frequency error is calculated based on the rotation angle difference; and _ _ corrects the 4 times and non-integer carrier frequency error. 2. The method according to item 1 of the patent application range, wherein the spreading code sequence is a known pilot spreading code sequence. 3. For example, the method of item 1 of the scope of patent application, wherein the spreading code sequence is an information spreading code sequence, and the matching step further includes: searching for the received signal and matching the information spreading code sequence, and Spikes; and the spikes with the smallest delay dispersion to modify the information spreading code sequence clock β 4. The method of item 1 in the patent application range, wherein the matching is L: \ EXT \ 52 \ 52160.DOC \ S — 26 — This paper Λ degree is applicable to the Chinese National Kneading Rate (CNS) Α4 size (210 × 297 mm) (Jing first read the notes on the back and then fill out this page). -Ο 6. The scope of patent application uses a fast Fourier transform Matching filters to handle. 5. For the method of claim 1 or 3 in the scope of patent application, wherein in the step of determining the direction flag, when the matching result of the integer is close to the matching result of an adjacent integer greater than the integer, the direction flag The standard system is set to indicate the counterclockwise direction; / When the matching result of the integer is closer to the matching result of the adjacent integer less than the integer, the direction flag system is set to indicate the clockwise direction; if it is not the above two, In this case, the direction flag is set to zero. 6. For the square weight of item 5 in the scope of patent application, in the step of calculating the rotation angle difference in y, when the direction flag is zero, the rotation angle difference can be calculated according to the counterclockwise direction or clockwise direction, but The resulting rotation angle difference must be between 0 degrees and 180 degrees or between 0 degrees and -180 degrees. 7. If the patent application range is 4 or 3 hard titanium squares, wherein the step of calculating a non-integer carrier frequency error further comprises: dividing the rotation angle difference by the length of the spreading code sequence. Printed by the Central Labor Bureau of the Ministry of Economic Affairs of the Male Workers Consumer Cooperatives (please read the precautions on the back before filling out this page) 8. If the method of item 1 or 3 of the patent scope is applied, where the non-integer carrier frequency error is calculated The steps further include: defining a matching path with a vibration amplitude greater than a predetermined threshold, as the main path; calculating the rotation angle difference of each main path; calculating the average 値 of the rotation angle differences; and dividing the average 値 by the exhibition. The length of the frequency code sequence. L: \ EXT \ 52 \ 52160.DOC \ 5 27-National paper (CNS) A4 specification (210X297 mm) in the application of this paper size 6. Application scope of patent 9. Method of applying for item 1 or 3 of patent scope In which, after the step of determining the direction flag, the carrier frequency error of the integer multiple is corrected. 10. —A method for correcting the carrier frequency, rate error and clock of the spreading code sequence in the direct-sequence spread spectrum communication system, including:, matching the received signal center frequency to an integer multiple of the reciprocal of the spreading code sequence period The spread spectrum code sequence is searched; the one with the largest amplitude in the matching result is searched to obtain a carrier frequency error whose size is a certain multiple of the period reciprocal of the spread spectrum code sequence; the matching result of the integer obtained before and after the search Compare the matching results of integers to determine the direction flag. Calculate the rotation angle difference between two adjacent matching results based on the direction flag. Use the rotation angle difference to calculate the non-integer carrier frequency error and correct the integer multiple and non-integer. The carrier frequency error; and the minimum delay dispersion range of the radiated amplitude obtained by the matching, which corrects the spreading code sequence clock. 11. The party applying for item 10 of the patent scope, wherein the spread spectrum code sequence is a known pilot spread spectrum code sequence. 12. If the method of item 10 of the scope of patent application is applied, wherein the spreading code sequence is an information spreading code sequence, and in the matching step L: \ EXT \ 52 \ S2160.DOC \ S, this paper standard uses the Chinese country CNS A4 specification (? 10 × 297 mm) Please read the notes and notes on the back first-i The Central Standards Bureau of the Ministry of Economic Affairs has printed and decorated the industrial and consumer cooperatives, and before applying for a patent, it also includes: A spike generated by the signal matching the information spread spectrum code sequence; and the spike with the minimum delay dispersion range to modify the data spread spectrum code sequence clock. 13. The method according to item 10 or 12 of the scope of patent application, wherein the matching system is processed by a fast Fourier transform matched filter. 14. As in the method of claim 10 or 12, in the method of determining the direction flag, when the matching result of the integer is closer to the matching result of an adjacent integer greater than the integer, the direction The flag system is set to indicate the counterclockwise direction; when the matching result of the integer is closer to the matching result of the adjacent unit number less than the integer, the direction flag system is set to indicate the clockwise direction; if it is not the above two conditions, , The direction flag is set to zero. Printed by the Central Government Bureau of the Ministry of Economic Affairs, Shellfish Consumer Cooperatives (please read the note on the back, and then fill out this page) 15 · If the method of applying for the scope of patent No.14, in the step of calculating the difference in rotation angle, when When the direction flag is zero, the rotation angle difference can be calculated in the counterclockwise or clockwise direction, but the calculated rotation angle difference must be between 0 degrees and 180 degrees or between 0 degrees and -180 degrees. 16. The method of claim 10 or 12, wherein the step of calculating a non-integer carrier frequency error further comprises: dividing the rotation angle difference by the length of the spreading code sequence. L: \ EXT \ 52 \ 52160.DOC \ 5 _ 29 — This paper uses China & quasi (CNS) A4 specifications (210 × 297 mm) 6. Application scope for a patent search device, receiving the matching device The output amplitude is used to determine the carrier frequency error whose size is an integer multiple of the reciprocal of the period of the spreading code sequence. A comparison device is used to compare the matching result between the integer obtained by searching and the integer before and after. To determine the direction flag; 'a first calculation device to calculate two matching devices based on the direction flag. The rotation angle difference between adjacent outputs, and use the rotation angle difference to calculate a non-integer Carrier frequency error; a first correction device for feeding back the output of the m search device and the first calculation device to the matching device to correct the integer multiple and non-integer carrier frequency error; a second calculation device that receives the Correct the carrier frequency error of the output of the matching device, and calculate the minimum delay dispersion range of the output vibration to determine the Zhanma sequence clock; and printed by Zhengong Consumer Cooperative, Central Bureau of Standards, Ministry of Economic Affairs (Please read the notes on the back before writing this page) A second correction device is used to reverse the output of the second calculation to the matching device to correct the clock of the spreading code sequence. 21. The receiver of claim 20, wherein the spreading code sequence is a known pilot. Spreading code sequence. 22. If the receiver of the patent application item 20, wherein the spread spectrum code sequence is an information spread spectrum code sequence; the search device is also used to search for spikes in the output vibration of the matching device; L: \ EXT \ 52 \ 52160.DOC \ 5 — 3 1 — This paper size is applicable to Chinese National Standard (CNS) A4 «L grid (210X297 mm) 々, patent application scope 17. If the method of patent application scope item 10 or 12, where The step of calculating the non-integer carrier frequency error further includes: defining a matching radiation amplitude greater than a predetermined threshold to obtain a main path; calculating a rotation angle difference of each main path; ^ calculating an average of the rotation angle differences 値; And dividing the average chirp by the length of the spreading code sequence. 18. If the method of claim 10 or 12 is applied for, the step of modifying the clock of the spreading code sequence further includes defining the main path that the vibration amplitude obtained by the matching is greater than a predetermined threshold; each main path is The starting point calculates the delay dispersion range of all the main paths; and the starting main path of the minimum delay dispersion range is used as the starting point of the clock of the spreading code sequence. 19. If you apply for the method of item 10 or 12 of the patent scope, which is printed in the -V ... I Central Labor Bureau of Ministry of Economic Affairs Consumer Cooperatives (please read the precautions on the back before filling this page) After the step of the outbound flag, the carrier frequency error of the integer multiple is corrected. 20. —A kind of communication system is used to connect to I, t includes an antenna and a receiving module, which are characterized in that the receiver includes: a matching device for receiving signals / signals _Matches the spreading code whose center frequency is an integer / multiple of the period reciprocal of the spreading code sequence. ^^; L: \ EXT \ 52 \ 52160.DOC \ £ — 3 0 " This paper size is applicable to Chinese countries. Standard (CNS) A4 specification (2H) X297 mm) A8 B8 C8 D8, the scope of patent application and the second computing device is also used to calculate the smallest delay dispersion range of these spikes to determine the information spreading code sequence pulse. 23. For example, the patented Fanyuan No. 20 or Yi J-based inspection machine, wherein the-J J. Pi Yt device is a fast Fourier transform matched filter. 24. The receiver of claim 20 or 22, wherein the search device includes: a control device for moving the center frequency of the spread spectrum code sequence; and a selection device for selecting the output device of the matching device The biggest one. 25. The receiver of item 21 of the patent application, wherein the first computing device includes: a device for determining a main path, which is used to determine that the output amplitude of the matching device is greater than a threshold value; the main path is calculated; A device for calculating the rotation angle difference of each main path according to the direction flag; an averaging device for calculating the average 値 of these rotation angle differences; and a non-integer carrier frequency error calculation device for dividing the average 値 by The length of the spreading code sequence to obtain a non-integer carrier frequency error. _ 26. If the receiver of the 22nd scope of the patent application, the first computing device includes: L: \ EXT \ 52 \ S2160.DOC \ 5 This paper size is used in China National Standards (CNS) Α4 specification (210X297 (Mm) Please fill in the notes on the back first-fill in this page Printed by the Central Standards Bureau of the Ministry of Economic Affairs and printed by the Industrial and Commercial Cooperatives-32-A8 B $ C8 D8 6. The device for determining the main route for the scope of patent application to determine the match The output path of the device is greater than a predetermined door shaker as the main path; the rotation angle difference calculation device is used to calculate the rotation angle difference of each main path according to the direction flag; / the rotation angle difference correction device is used to-correct The calculation, the amount of blur caused by the information adjustment in the rotation angle difference > averaging device is used to calculate the average 値 of the corrected rotation angle difference; and a non-integer carrier frequency error calculation device is used to Divide the average chirp by the length of the spreading code sequence to obtain a non-integer carrier frequency error. 27. If the scope of the patent application is in the range of 4 揸氇, the first correction device corrects the integer multiple carrier frequency error before the first calculation device is started. Printed by the Zhengong Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economics (please read "Notes on the back" and fill in this page first) 28. If you apply for a receiver with the scope of patent 20 or 22, the second computing device includes: Decision The device of the main path is used to determine that the output amplitude of the matching device is greater than a threshold. The device is a main path. The device of calculating the minimum delay dispersion range calculates the delay dispersion range of all the main paths with each main path as the starting point. ; And a clock determination device, which is used to spread the range of minimum delay L: \ EXT \ 52 \ 52160. DOC \ 5 — 33 — This paper size is applicable to China National Standard (CNS) A4 (210X297 mm) A8 B8 C8 D8 The main path of the patent application scope is the starting point of the clock of the spread spectrum code sequence. 『Please note from Min · Item Binding Printed by the Industrial and Consumer Cooperatives of the Central Standards Bureau of the Ministry of Economic Affairs«. -34-L: \ EXT \ 52 \ S2160.DOC \ 5 This paper size applies the Chinese National Standard (CNS) A4 specification ( 2! 〇X297 mm)