I 懸1421 、 九、發明說明: [發明所屬之技術領域] 本發明係有關於一種偵測既定同步數據之同步數據偵 測單元以及方法,且特別是有關於一種在一無線區域網路 (WLAN,Wireless local areanetw〇rk)通訊系統内之接 器與接收方法。 [先前技術] 在比如為無線區域網路(WLAN)系統之通訊系統内,崎 接收器同步於傳輸器是重要的,這使得訊息可成功地交^ 於該傳輸器與該接收器之間。無線區域網路系統是一種具 有弹性之數據通訊系統,可實施為有線區域網路之延伸或 另種擇。WLAN系統利用射頻或紅外線技術而透過空氣 來傳輸以及接收數據以將連接線之需求降至最低。因此, 該WLAN系統兼有數據連結以及使用者移動性。 大部份的WLAN系統係使用展頻技術,該展頻技術為使 用於可靠與安全性通訊系統中所發展出之寬頻射頻技術。 該展頻技術係設計成能在可靠性、整合性與安全性之間取 得頻寬使用效率之平衡點。經常使用之兩種展頻無線系統 係··跳頻(frequency hoping)與直接序列(Direc1: Seq[uence) 糸統。 在直接序列展頻技術系統中,利用具有較高頻率與較 咼資訊位元率之字元碼與字符(symbol)來對各數據位元進 行編碼而完成展頻。所得之”展頻”後信號係分佈於較大頻 I上’這將導致相當低之功率頻譜密度,使得其他通訊系 92326倐正太 1^1421 ' . l 統較小會受到傳輸此直接序列展頻信號之裝置之干擾。直. 接序列展頻技術應用對該傳輸器與接收器而言皆為^之 模擬(pseudo)隨機雜訊字元碼以展頻該數據。該字元碼包 括由待傳輸之該資訊位元乘上(或進行,,互斥或”邏輯運算) 之^串之+脈衝(chlp)”。許多無線網路符合應用已知 巴克(Barker)字元碼uEEE〇nstitute 〇f and Electronics Engineers,電子電機工程師協會)之 802. 11規格來編碼與展頻該數據。該巴克字元碼包括一串籲 既定之11個小脈衝。在包含資訊之字符所佔之時間週期 内,係傳輸一整串巴克字元碼。 為允許更高之數據傳輸率,該IEEE8〇2 u規格延伸至 IEEE802· 1 lb。除了該11位元之巴克小脈衝外,該 IEEE802· Ub規格使用高數據傳輸率之8位元互補碼移位鍵 (Complementary Code Keying;簡稱CCK)演算法。 應用更咼階之調變技術,包括正交移相鍵控調變 (Quadrature Phase Shift Keying,QPSK),該數據傳輸率春 可改善成高於該字符率。根據此種調變技術,各位元由可 月匕之相位之較高號碼來代表。因而,該傳輸器產生兩個信 號,第一個信號稱為”同相位(丨)信號或”丨通道,,,而第二個 ^號稱為在相同頻率下之9〇度移相之正弦載波之"正交相 位”(0)信號或nQ通道,,。 使用直接序列展頻技術之無線區域網路之該 IEEE802. 1 1規格應用訓練前置碼(trai η 1 ng preambl e)以 訓練接收器對傳輸器之傳輸。各傳輸數據訊息包括接續著 92326修正本 ΡΚΪ4ΙΪ 襴=二:=置碼。該前置碼係包括-同步 班 玄设收裔可進仃必須之同步操作。對二 :碼長度:已定義兩種選擇,亦即一長前置碼與—短 :短:有二7'm.1㈣格之系統必須支援該長前置:。 音或1:傻 擇性提供於該規格内以改善當傳輸比如聲 〜之特殊數據時之網路總處理能力之效率。長前 且馬之同步攔位包括128個為!之位 攔位包括56個為〇之位元。 心❹碼之同步 測該同步符號並將該接㈣之内部時脈| 、邊同”位内之該符號以建立一固定之參考時間框 此广)以解譯接續該前置碼後之該傳輸框訊結構中之节 二包置括:)同步搁位之該前置碼係傳輸於各訊㈣ 作號前單7"之目的是持續地監測該前置碼之輸入I hang 1421, ninth, invention description: [Technical field of the invention] The present invention relates to a synchronous data detecting unit and method for detecting a predetermined synchronization data, and in particular to a wireless local area network (WLAN) , Wireless local areanetw〇rk) The connector and receiving method in the communication system. [Prior Art] In a communication system such as a wireless local area network (WLAN) system, it is important that the receiver is synchronized with the transmitter, so that the message can be successfully transferred between the transmitter and the receiver. A wireless local area network system is a flexible data communication system that can be implemented as an extension or alternative to a wired area network. WLAN systems use radio or infrared technology to transmit and receive data through the air to minimize the need for cables. Therefore, the WLAN system combines data connectivity and user mobility. Most of the WLAN systems use spread spectrum technology, which is used to make broadband RF technology developed in reliable and secure communication systems. The spread spectrum technology is designed to achieve a balance between bandwidth efficiency and reliability, integration and security. Two types of spread spectrum wireless systems that are frequently used are frequency hoping and direct sequence (Direc1: Seq[uence). In the direct sequence spread spectrum technology system, the spread spectrum is completed by encoding each data bit with a character code and a symbol having a higher frequency and a higher information bit rate. The resulting "spreading frequency" signal is distributed over a large frequency I' which will result in a relatively low power spectral density, making other communication systems 92326 倐正1^1421 '. l system will be transmitted this direct sequence show Interference from the device of the frequency signal. Straight. The sequence spread spectrum technology application is a pseudo random noise character code for both the transmitter and the receiver to spread the data. The character code includes a +ch (chlp) of the string of the information bit to be transmitted multiplied (or performed, mutually exclusive or "logical"). Many wireless networks conform to the application's known Barker character code uEEE〇nstitute 〇f and Electronics Engineers, 802.11 specification to encode and spread the data. The Barker character code includes a series of 11 small pulses that are predetermined. A full list of Barker character codes is transmitted during the time period in which the characters containing the information are taken. To allow for higher data transfer rates, the IEEE 8〇2 u specification extends to IEEE802·1 lb. In addition to the 11-bit Barkerian small pulse, the IEEE802.Ub specification uses an 8-bit Complementary Code Keying (CCK) algorithm with a high data rate. Applying a more sophisticated modulation technique, including Quadrature Phase Shift Keying (QPSK), the data transmission rate can be improved to be higher than the character rate. According to this modulation technique, each element is represented by a higher number of the phase of the moon. Thus, the transmitter produces two signals, the first signal being called the "in-phase (丨) signal or the "丨 channel", and the second ^ is called the sinusoidal carrier of the 9-degree phase shift at the same frequency. The "orthogonal phase" (0) signal or nQ channel, the IEEE802.11 specification using the direct sequence spread spectrum technique applies the training preamble (trai η 1 ng preambl e) to train Receiver-to-transmitter transmission. Each transmitted data message includes a subsequent correction of 92326 ΡΚΪ4ΙΪ 襕=2:=code. The pre-code includes - synchronous class 设 设 设 收 收 收 收 同步 同步 同步 同步 。 。 。 。 : Code length: Two options have been defined, namely a long preamble and a short: short: a system with two 7'm.1 (four) grids must support the long preposition: tone or 1: silly This specification is used to improve the efficiency of the network's total processing power when transmitting special data such as sounds. The long-term and the horse's synchronous block includes 128 bits! The block block includes 56 bits. The synchronization symbol is measured synchronously and the internal clock of the connection (4) is The symbol in the bit is set up to establish a fixed reference time frame to decode the section of the transmission frame structure after the preamble is connected: the preamble transmission of the synchronous shelf The purpose of the pre-order 7" in each news (4) is to continuously monitor the input of the preamble
Heed ^已_到該前置碼。決定連續之巴克符 之邊界,且該符號之往前傳送時序係同步於 。士亥接收A之處理時間表。根據該前置碼_以及符號 =與1續模組之處理時間表間之時序偏差,該輪入信號 '丁、同步於該接收器之處理時間表。 :在請參考第!圖,其繪示在一通訊信號内镇測一前置 2[測處理。在接收通訊信號之步驟⑽之後以及在將所 之该通訊信號進行更進一步處理,特別是譯碼步驟⑽ 之珂係進行前置碼偵測步驟101。 w 呈馬债測益2〇〇之架構係繪示於第2圖。包括同 92326修正本 7 1¾ 補:::::::二:: · : .丨‘、 "'>"·. - - V: : 外…?.........................- ,, 相位只i又相位成份之該接收之通訊信號201係輸入至該· 則且碼偵測為200。纟該前置碼债測器2〇〇内,該接收之通 訊信號2G1首先輪人至解展頻器㈣m,立特別 為巴克匹配濾波器⑽F,Barkermatchedfilter)。經解 展頻之逋汛信號係輸入至解調變器(丨D腿)2〇5 以對經解展頻之通訊信號進行解調變。解調變後之信號包 括-連串之所接收之位元順序之"明確"決定順序,也就是 說,解調變後之信號之各數據值為兩個可能之二進位值之· |測解凋’交後之位元流以偵測該預先定義之前置碼數 據 般而3,使用一相關器(correiat〇r)以偵測該前置 碼。亥相關為本質上為該前置碼順序之一匹配渡波器。當 β刖置碼存在時’該相關器產生具有大數值之輸出。當該 相關性之數值超過一預先定義之臨限值時,可宣稱正式地 偵測到前置碼。 在偵測到前置碼之後,經解調變之該通訊信號係輸入 至一(數位)譯碼器(descrambler,DDS)。習知技術所用之馨 澤碼為300之一例係繪示於第3圖中。根據預先定義之譯碼 原則,輸入k號3 〇 1 a係輸入至代表將時間延遲數個單位之 延遲方塊304與305。經延遲之該信號係回授並利用乘法器 或EXOR閘306而合併。該輸出信號係回授至該輸入信號3〇lb 並利用宋法為或EXqr閘303而合併,以產生經譯碼之輸出信 號302。 " 同步數據偵測單元仍然有數個問題。問題之一就是, 雜Λ 了 曾卩牛低该彳έ號品質,使得該同步單元,特別是該 92326修正本 1251421 前置碼偵測器,無法正確辨認前置碼,即使有前置碼存在 1所接收之該通訊信號内。當並沒有真實的前置碼存在 B于,雜訊也可能會產生超過該臨限值之一輪出传號。 [發明内容] J θ ^ ° ^本發明提供-種經改良之同步_單元以及方法,其 較不易在偵測預先定義之同步數據上產生錯誤。 在λ知例中,在通訊糸統内提供—種用以偵、測通訊 信號内之傳輸框内之既定同步數據之同步數據侦測單元。( 该同步數據包括連串之先編碼後傳輸之相同之二進位符 號。該同步數據偵測單元包括繹 匕枯存碼裔,以譯碼所接收之 石亥通訊#號以及產生且右客 〃 有^路“唬值之輸出數據串。該多 路仏諕值係輸入至濾波裝置以平^ ^ ^ ^ ^ ^ ^ T w T月1C< "茨洋碼态之輸出。該 =輸入至臨限裝置。該臨限裝置比較該繼 ==出與既定臨限值。如果該遽波裝置之該輸出 過姐疋臨限值’則該臨限裝置指示價測到該同步數 κ &例中,在通訊系統内提供-種用以摘測於 通訊k號内之傳輸框内之既定 、、 一 ^ 既疋同步數據之同步數據偵測單 兀。该同步數據包括一連串 甲又无'、扁碼俊傳輸之相同之二進 收之該通訊信號以及產生且 妾 ^ ^ ^ 八有夕路k唬值之輸出數據串。 μ夕路k 5虎值係輸入至一滹 # 一 芯波态以平滑化該譯碼器之輪 。咕平滑後信號係輸入至— Dn 、、古壯罢+、τ,、, 比1乂态。忒比較态比較該渡 衣、,錢輸出與—既定臨限值。如果該遽波裝置 9232()修正本 9Heed ^ has _ to the preamble. The boundary of the continuous Bark symbol is determined, and the forward transmission timing of the symbol is synchronized with . Shi Hai receives the processing schedule of A. According to the timing deviation between the preamble _ and the symbol = and the processing schedule of the 1 continuation module, the rounding signal 'synchronizes with the processing schedule of the receiver. : Please refer to the figure!, which shows a pre-measurement in a communication signal. The preamble detection step 101 is performed after the step (10) of receiving the communication signal and after further processing the communication signal, particularly the decoding step (10). w The structure of the Ma Yuan Debt 2 is shown in Figure 2. Including the same as the 92326 amendment 7 13⁄4 Supplement:::::::2:: · : .丨 ‘, "'>"·. - - V: : Outside...? .........................-,, the phase only i and the phase component of the received communication signal 201 is input to the code and code detection The test is 200.纟 Within the preamble debt detector 2, the received communication signal 2G1 first turns to the despreader (4) m, and is specifically a Barker matched filter (10)F, Barkermatched filter). The signal of the spread spectrum is input to the demodulation transformer (丨D leg) 2〇5 to demodulate the despread communication signal. The demodulated signal includes a sequence of the received bit sequence "clear" decision order, that is, each data value of the demodulated signal is two possible binary values. | Measure the bit stream after the intersection to detect the pre-defined preamble data. 3, use a correlator (correiat〇r) to detect the preamble. The Hai correlation is essentially one of the preamble sequences that match the ferris. When the β刖 code is present, the correlator produces an output with a large value. When the value of the correlation exceeds a predefined threshold, it can be declared that the preamble is formally detected. After detecting the preamble, the communication signal is demodulated and input to a (descrambler, DDS). One example of the stencil code used in the prior art is shown in Figure 3. According to a predefined decoding principle, the input k number 3 〇 1 a is input to delay blocks 304 and 305 representing delays in units of several units. The delayed signal is fed back and combined using a multiplier or EXOR gate 306. The output signal is fed back to the input signal 3 〇 lb and combined using the Song method or the EXqr gate 303 to produce a decoded output signal 302. " Synchronous data detection unit still has several problems. One of the problems is that the Λ 卩 卩 卩 低 低 低 , , , 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 同步 923 923 923 923 1 received within the communication signal. When there is no real preamble present, the noise may also produce a turn-out mark that exceeds one of the thresholds. SUMMARY OF THE INVENTION J θ ^ ° ^ The present invention provides an improved sync_unit and method that is less prone to generate errors in detecting pre-defined sync data. In the λ-known example, a synchronous data detecting unit for detecting and measuring the predetermined synchronization data in the transmission frame in the communication signal is provided in the communication system. (The synchronization data includes a series of identical binary symbols that are encoded and transmitted. The synchronous data detection unit includes a coded data source to decode the received Shihai communication # number and generate and right clients. There is an output data string of the "depreciation value". The multi-channel threshold value is input to the filtering device to output the output of ^ ^ ^ ^ ^ ^ ^ T w T 1 <"" a threshold device. The threshold device compares the subsequent == and a predetermined threshold. If the output of the chopper device exceeds the threshold, the threshold device indicates the synchronization number κ & In the example, a synchronous data detection unit for extracting the predetermined data in the transmission frame in the communication k number, and the synchronization data is included in the communication system. The synchronization data includes a series of A and no. The same two transmissions of the flat code Jun transmission and the output data string of the 妾^ ^ ^ 八有夕路k唬 value. μ 夕路 k 5虎值系 input to a 滹# a core wave The state is to smooth the wheel of the decoder. After smoothing, the signal is input to - Dn, Gu Zhuang +, τ, ,, 比1乂. 忒Comparative state compares the ferry, the money output and the established threshold. If the chopper device 9232() amends this 9
“出超過該既定臨限值,則該比較器指示债、測到該同步 卉在又另一貫施例中,提供一種用於通訊系統内接收同 二數據之方法。該既定同步數據係包括於通訊信號内之傳 .私内斤W同步數據包括一連串之先編碼後傳輸之相同之 一進位付戒。譯碼所接收之該通訊信號以產生一連串之多"After exceeding the established threshold, the comparator indicates the debt, and the synchronization is detected. In yet another embodiment, a method for receiving the same data in the communication system is provided. The predetermined synchronization data is included in The communication signal is transmitted within the communication signal. The synchronization data includes a series of the same one of the first coded transmissions. The decoded communication signal is generated to generate a series of signals.
路4¾號值。巫、'取二士々A I /月化该多路輸出信號值;並且比較該平 信號與既定⑪PP # °。限值。如不該平滑後信號超過該既定臨限 值,則代表偵測到該同步數據。 j "t本毛明之上述和其他目的、特徵、和優點能更明 顯易丨重,下文特舉一較佳實施例,並配合所附圖式,作詳 細說明如下。 ϋ [實施方式] 現在將苓考所附圖式來描述本發明之實施例。 現在請參考附圖,特別是參考第4圖,第4圖繪示一種 同^據_單元,用於_同步數據,特別是預先定義 之g置馬,如本文所描述。如第4圖所顯示之架構包括同步 數據備測模組4〇〇以及同步模組4()3。該同步數據横測模組 400偵測一别置碼’且也提供一符號到達時間與該後續模組 之該處理時程間之時序偏差。該符號同步器4 G 3將利用該時 序偏差資訊來將該數據流同步於該處理時程。 該同步數據偵測模組4 〇 〇包括下列模組:巴克匹配濾波 。。(BMF,Barker matched fi Iter)模組4〇4、差分BPSK解調 變器(DEM)模組405以及軟體譯碼器(s〇n descramb】er, 92326修正本 10 通眺錄 . '" ^ - 4Road number 43⁄4. Witch, 'take two gentry A I / monthlyize the multi-output signal value; and compare the flat signal with the established 11PP # °. Limit. If the smoothed signal does not exceed the predetermined threshold value, it means that the synchronization data is detected. The above and other objects, features, and advantages of the present invention will become more apparent and obvious. The following detailed description of the preferred embodiments and the accompanying drawings are set forth below. [Embodiment] Embodiments of the present invention will now be described with reference to the drawings. Referring now to the drawings, and in particular to FIG. 4, FIG. 4 illustrates a synchronizing unit for _synchronizing data, particularly a pre-defined g-horse, as described herein. The architecture shown in FIG. 4 includes a synchronous data preparation module 4A and a synchronization module 4()3. The synchronous data transversal module 400 detects a unique code' and also provides a timing offset between a symbol arrival time and the processing time of the subsequent module. The symbol synchronizer 4 G 3 will use the timing offset information to synchronize the data stream to the processing time. The synchronous data detection module 4 〇 〇 includes the following modules: Barker matched filtering. . (BMF, Barker matched fi Iter) module 4〇4, differential BPSK demodulation transformer (DEM) module 405 and software decoder (s〇n descramb) er, 92326 amend this 10 transcript. '" ^ - 4
.—.·.....… ·........ ' ‘ I SDS)模組406。該些模組,也就是說MF、DM以及S])S形成 非協調性(iicm-coherent)接收器。甚至,該同步數據偵測 模組400包括梳形濾波器(c⑽b fUter,⑶F)模組4〇7以平 滑該譯碼器之輸出。為評估該平滑後數據以偵測該預先定 我之鈾且碼數據,该同步數據偵測模組4⑽包含臨限控制器 408。上述模組之細節將於下文中描述。 該巴克匹配濾波器模組4〇4接收輸入至該同步數據偵 測模組400之該通訊信號4〇1,並計算該巴克順序與該輸入_ 信號之取樣值間之相關性。該巴克字元碼包括u個小脈 衝,該11個小脈衝之順序為”〇1〇〇1〇〇〇111”或,,+1,—1,+1, + 1 ? -1 5 +1 ^ +1 ^ +1 5 , -lf, (non-return-to~zero ^ 不歸零,NRZ),其中最左邊之小脈衝最先輸出。一個完整 的巴克字兀碼順序係接收於由含資訊之符號所佔據之時期 内。因此,如果該符號率是IMbaud,該巴克順序之該1H@ 小脈衝之基本小脈衝率為11MHz。藉由使用該UMHz之小脈 衝率信號,該傳輸信號所佔據之頻譜會變成十一倍大。在馨 —實施例中,因為輸入取樣率為22MSPS的關係,該巴克順 序係從11個取樣延伸成22個取樣。這可藉由將〇加入於該巴 克順序之原始元件之間而達成。 該I通道與该Q通道之該輸入取樣係分別相關於該巴克 順序。複合相關取樣係計算於各複合輸入取樣。這^藉由 ,知技術已知之”滑動-視窗(slldingiind〇w)”演算法而 實施。 八… 解展頻(despread)之通訊信號輸入至該解調變器模組 92326修正本 1¾賴1 4〇5〇 一 。在付疋實施例中,該解調變器模組405是一微分BPSK 解調變器。 之a I在明言1第5圖以及弟6圖,其繪示譯碼器模組4 〇 β 之討沐貫施方式。該譯碼器模組406可實施為能夠輸出多重 =取^之軟體譯碼器。相對於數位解調變器,在軟體譯碼 。。内σΡ,多重值之輸入信號係未受到"明確”二進位化之處 理’但經譯碼之該輸出值係沿著該預期之二進位值而排列。 士 =此處所描述之其他實施例中,"半軟體(half—sofd 用於對經解調變之信號進行譯碼j,半軟體"譯碼 W組500之架構之例子係顯示於第5圖與第6圖中。半軟體 ^码器與軟體譯碼器之不同處在於,輸人至該譯碼器内部 之/遲部份”5〇4至5()6之該輸人信㈣—被二㈣化」 7輸入信號5Glb之其他部份係維持成多重值信號。在第5 圖之譯碼器模組5〇〇中,所接收之該輸入信號心係輸入至 -進位化部份503以將多重值輸入取樣值轉換成二進位值。 该譯碼器模組500係根據預先定義之產生多項式而架( 構,该多項式比如為1 +xa + xb,i ^ ^ ^ 1 =。經延遲之信號係利用乘法器或職閘而回授並加入 二輪入信號以產生經譯碼之輪出。如第β圖中所顯示,該 =兀件504與505包括複數個1位元暫存器6〇ι至_,各暫 存為代表一單位的時間延遽。兮生 ,R 、社 之故5玄+軟體譯碼器輸出沿著該預 期二錢值而排列之-串軟符號(如卜咖。⑷。 、 弟5圖以及第6圖之該半軟體譯碼器配置之優點在於, 4目較於傳統之軟體譯碼器,提供多路輸出取樣所需之硬 92326修正本 12 1251421 體成本可獲得相當的節省。 當接收前置碼數據時,所有之軟體或半軟體譯碼器之 多路輸出取樣值將具有相同符號。為減少在該輪出取樣值 内之隨機誤差之影響,該輪出取樣值係藉由梳形濾波器模 ’’且4 0 7叩平均。木Tl形濾波益之實施例係顯示於第7圖以及第8 圖中。 /現在請參考第7圖,所輸入之實數與虛數數據符號701 係輸入至乘法器705而由權值趴加以加權。經加權之信號係籲 利=延遲元件7〇4以及加法器7〇3而加入至回授之延遲後輸 出信號702。在將該回授輸出信號與該輸入信號相加之前, 延遲後之回授信號也使用乘法器7〇6以由權值趴加以加權。 根據一特殊實施例’輸入之數據符號7〇1係由乘法器 705乘上0.2之固定點等效值,而延遲元件7〇4所提供之經延 遲之’’平均’’值係乘上〇. 8之固定點等效值。雖然此實施例之 該演算法係分別參考0.2以及0.8之權值而描述,但習知此 技者可了解該加權演算法可利用各種權值組合_W2而達籲 成相同效果。該輸入取樣值7〇1之雜訊降低效果可藉由減少 權,A之值以及增加I之值而增強。相對地,及時接收數 據符唬之正確性可藉由增加第一權值^之值以及減少第二 權值W2之值而增強。 曰一棱形濾波斋架構之另一種實施例係顯示於第8圖中。所 缔貝不出之該梳形濾波器之各輸出代表被u個或22個取樣 (取决於所應用之輸入取樣率)所隔開之η個小脈衝取樣之 平均饭。如同習知此技藝者所明瞭,平均取樣之數量可適 92326修正本 ]3 1251421 當設定以達到足狗之雜訊降低效果。當增力㈣數目時,將 更有效地減少雜訊。在此實施例之特殊例中,平均取樣之 鮮雜同相位通道1與正交相位通之平均振幅係獨 立計异。這可實施下列公式而在該特定實施例内達成: fQ(k)=缶Σ 冬(々-22/·) 取‘之巴克順序時之該譯碼器之輸出,如上述般。 波器之輸出fl以及fQ都將用於決1目前是否正在接 =碼。在_施例中’係將此兩輸出值相加以抵銷 6亥L號之頻率偏差之影響·· S(kT)-f,(kT) + fQ(kT) 表此兩輸出值之相加值’而fi以及“代表該各別通 道之杈形濾波器之輸出之平均值。 月回頭茶考第8圖’梳形濾、波器800包括複數個延 遲兀件804、加法器8〇3以及除法哭私认 除法M 805。所輸入之譯碼器輸 出8 01 {丁、輸入至該加法哭m q 忐為803亚且更輸入至該些彼此完全相 元侧之第一個元件。各延遲元件_之輸出係 =/加法⑤803,亚輸入至下—個延遲元件8G4之輸入 端。該加法器8〇3接收η個既玄叙旦+击入 謂既疋數里之輸入信號,該些輸入 、δ/Γ ή積並輸人至該除法器8Q5。輪人之數量^以及延 遲兀件804之相關數量lW係根據上述考量而設定。除法器 92326修正本 14 二:加《态8。3所提供之總和進行正規化。 子。==出如何在輸入通訊信號内_前置碼之例 體譯碼處理^嶋,她㈣係受到軟 以減少在該诵 处理而澤碼後輸出係受到平滑化 内之隨機誤差之影響(步驟9G2>在步 形漁波哭现置剔貞測係藉由比較該梳 權所執行之操作細節料示於第1G圖中。“u“ s(半+_示’所得到之各梳形;慮波器之總和 二)係在則置碼搜尋期間跟預先定義之臨限值T h =義=步驟1002所示。當該梳形遽波器之輸出大於該預 先疋義之臣品限值Th時,可視為已債測到前置碼(步驟 3) ’而該前置碼谓測操作進入到,,前置碼備測狀能 器彻將維持於該狀態中,但㈣取兩個小 決疋後縯之譯碼器之輸出是否會更大。 讀 在比較另外兩個取樣值之後,該臨限控制器將 =認該比較結果,也就是說,要決定在符號持續時間之 後是否:再度出現譯碼器之峰值。在明確宣稱已债測到前 置碼之前,該梳形渡波器之輸出大於該預先定義之臨限值 Th之次數必須超過既定次數(步驟1〇〇4,1〇〇5)。當步驟1⑽2 中之該比較器結果已確認過既定次數^之後,則進入該π 閉鎖狀態"(在已偵測到前置碼之情況中)。在一眚旆方卢 中,Τκ為15,也就是說,必須確認已偵測到前置碼丨5次。 幻326倐巴太 i Y : 習知此技藝者當可知也可使用任意的其他次數以達到相& 效果,比如說,介於1 0至20次之間。 同 根據無線LAN之該IEEE802. lib規格,該臨限控制哭4〇8 可用於偵測長與短前置碼。因為此兩種前置碼不尸在°°長产 上有差異,該前置碼順序之該二進位值也是不同,可.—...................... ' ‘ I SDS) module 406. The modules, that is to say MF, DM and S]) S form a non-coordinated (iicm-coherent) receiver. In even, the synchronous data detection module 400 includes a comb filter (c(10)b fUter, (3)F) module 4〇7 to smooth the output of the decoder. To evaluate the smoothed data to detect the pre-determined uranium code data, the synchronous data detection module 4 (10) includes a threshold controller 408. Details of the above modules will be described below. The Barker matched filter module 4〇4 receives the communication signal 4〇1 input to the synchronous data detection module 400, and calculates a correlation between the Barker sequence and the sampled value of the input_signal. The Bark character code includes u small pulses, and the order of the 11 small pulses is "〇1〇〇1〇〇〇111" or, +1, -1, +1, + 1 ? -1 5 +1 ^ +1 ^ +1 5 , -lf, (non-return-to~zero ^ does not return to zero, NRZ), where the leftmost small pulse is output first. A complete Bark code sequence is received during the period occupied by the symbol containing the information. Therefore, if the symbol rate is IMbaud, the basic small pulse rate of the 1H@ small pulse of the Barker sequence is 11 MHz. By using the UMHz small pulse rate signal, the spectrum occupied by the transmitted signal becomes eleven times larger. In the Xin-Example, the Barker sequence extends from 11 samples to 22 samples because of the input sampling rate of 22 MSPS. This can be achieved by adding 〇 to the original elements of the Barker sequence. The I channel and the input sampling system of the Q channel are associated with the Barker sequence, respectively. The composite correlation sampling is calculated at each composite input sample. This is implemented by the "slldingiind〇w" algorithm known in the art. Eight... The despread communication signal is input to the demodulator module 92326 to correct the original 1⁄4赖5〇 one. In the embodiment, the demodulator module 405 is a differential BPSK demodulation transformer. The a I is in Fig. 5 and Fig. 6 of Fig. 1, which shows the mode of the decoder module 4 〇 β. The decoder module 406 can be implemented as a software decoder capable of outputting multiple = fetches. Relative to the digital demodulator, it is decoded in software. . Internal σΡ, the multiple-valued input signal is not subject to "clear" binary processing' but the decoded output values are ranked along the expected binary value. 士=Other embodiments described herein In the middle, "half-software (half-sofd is used to decode the demodulated signal, j-software" is an example of the architecture of the W group 500. It is shown in Figures 5 and 6. The difference between the software coder and the software decoder is that the input/send part of the internal/lower part of the decoder "5 〇 4 to 5 () 6 of the input letter (four) - is two (four)" 7 input The other part of the signal 5Glb is maintained as a multi-value signal. In the decoder module 5A of Fig. 5, the received input signal is input to the carry-up portion 503 to input the multi-value input. The value is converted to a binary value. The decoder module 500 is configured according to a predefined polynomial, such as 1 + xa + xb, i ^ ^ ^ 1 =. The delayed signal is multiplied by a multiplication method. The device or the gate is fed back and adds a two-round signal to produce a decoded round. As shown in Figure β, the =5 04 and 505 include a plurality of 1-bit scratchpads 6〇ι to _, each temporary storage is a time delay representing one unit. Twins, R, and the company's 5 + + software decoder output along the expected The two-valued-column-soft symbol (such as Bucha. (4)., the 5th figure and the 6th figure of the semi-software decoder have the advantage that the 4 mesh provides more than the conventional software decoder. The hard output of the road output sampling requires a substantial savings of 12,125,421. When receiving preamble data, all software or semi-software decoders will have the same symbol for the multiple output samples. In the effect of the random error within the round-robin sample value, the round-robin sample value is averaged by the comb filter mode and 4 0 7 。. The embodiment of the wood T1-filter is shown in Figure 7 and In Fig. 8, / Now, referring to Fig. 7, the input real number and imaginary data symbol 701 are input to multiplier 705 and weighted by weight 。. The weighted signal is = = delay element 7 〇 4 and The adder 7〇3 is added to the delayed feedback output signal 702. The feedback is given The delayed feedback signal is also weighted by a weight 趴6 using a multiplier 7〇6 before the output signal is added to the input signal. According to a particular embodiment, the input data symbol 7〇1 is multiplied by a multiplier 705. The fixed point equivalent value of 0.2, and the delayed ''average'' value provided by the delay element 7〇4 is multiplied by the fixed point equivalent of 〇 8. Although the algorithm of this embodiment is respectively Referring to the weights of 0.2 and 0.8, it is known that the weighting algorithm can use the various weight combinations _W2 to achieve the same effect. The noise reduction effect of the input sample value 7〇1 can be achieved. It is enhanced by reducing the weight, the value of A, and increasing the value of I. In contrast, the correctness of receiving data symbols in time can be enhanced by increasing the value of the first weight ^ and decreasing the value of the second weight W2. Another embodiment of a prismatic filter architecture is shown in FIG. The outputs of the comb filter, which are not shown, represent the average of n small pulse samples separated by u or 22 samples (depending on the applied input sampling rate). As is known to those skilled in the art, the average number of samples can be adjusted to 92326.] 3 1251421 When set to achieve the noise reduction effect of the foot dog. When the number of (4) is increased, the noise will be reduced more effectively. In the particular example of this embodiment, the average amplitude of the equally sampled fresh phase in-phase channel 1 and the quadrature phase pass is independently calculated. This can be implemented in the specific embodiment by implementing the following formula: fQ(k) = 冬 Winter (々-22/·) The output of the decoder when the ‘Buck sequence is taken, as described above. The output of the waver fl and fQ will be used to determine whether the current code is currently being connected. In the _example, the relationship between the two output values is offset by the difference of the frequency deviation of the 6H L. · S(kT)-f, (kT) + fQ(kT) The values 'and fi and the average value of the output of the 杈-shaped filter representing the respective channels. Figure 8 of the retrospective tea test' comb filter, the wave device 800 includes a plurality of delay elements 804, an adder 8〇3 And the deciphering private recognition method M 805. The input decoder output 8 01 {d, input to the addition cry mq 忐 is 803 ya and is further input to the first component on the side of each complete phase. The output of the component_=addition 5803 is sub-inputted to the input of the lower delay component 8G4. The adder 8〇3 receives n input signals that are both meta-synchronized and hit into the number of turns, The input, δ/Γ is concatenated and input to the divider 8Q5. The number of rounds and the associated number lW of the delay element 804 are set according to the above considerations. The divider 92326 corrects the 14th: plus "state 8." The sum provided by 3 is normalized. Sub. == How to decode the internal code of the preamble in the input communication signal ^嶋, she (four) is soft In order to reduce the random error in the smoothing of the output system after the processing of the ( code (step 9G2), the details of the operation performed by comparing the combing rights are shown in the step-by-step method. In Fig. 1G, "u" s (half +_ shows the respective comb shape; the sum of the filter two) is in the code search period with a predefined threshold T h = meaning = step 1002. When the output of the comb chopper is greater than the pre-defective product limit value Th, it can be regarded as the pre-coded preamble (step 3)' and the preamble pre-measurement operation enters, The preamble ready tester will remain in this state, but (4) whether the output of the decoder will be larger after taking two small decisions. After reading the other two sample values, the The limit controller will = confirm the comparison result, that is, determine whether after the symbol duration: the peak of the decoder reappears. The output of the comb ferrite before explicitly claiming that the preamble has been measured. The number of times greater than the pre-defined threshold Th must exceed the specified number of times (steps 1〇〇4,1〇〇5) When the result of the comparator in step 1 (10) 2 has been confirmed for a predetermined number of times ^, then the π latched state is entered (in the case where the preamble has been detected). In a square, Τ κ It is 15, that is, it must be confirmed that the preamble has been detected 5 times. 幻326倐巴太 i Y : I know that this artist can use any other number of times to achieve phase & Said, between 10 and 20 times. According to the IEEE802. lib specification of wireless LAN, the threshold control cry 4〇8 can be used to detect long and short preambles. Because the two preambles are not The corpse has a difference in the long-term production of °°, and the binary value of the preamble sequence is also different.
對應的該二進位前置碼值之不同臨限值來分別該些前 碼。 Λ -月丨J 根據該無線L A Ν規格之能夠偵測長與短前置碼之臨限 比較#作係繪示於第11圖中。該前置碼偵測操作應用第— 臨限值T h 1以及第二臨限值τ h 2。根據所預期之譯碼“器輪出— 值+ 1/-1,此兩臨限值只在於正負符號上有所不同。 乂在步驟11〇2或步驟1107内先偵測到長或短前置碼之 後,對各前置碼重複地進行步驟11〇3_11〇6或步驟 1⑽-⑴k "可靠性檢查”。在所侧之長或短前置碼無法 確認於長前置碼確認操作之步驟1103_1106或者短前置碼 確認操作之步驟1108-1111所形成之確認迴圈之某一迴圈 内,該臨限比較器跳回至步驟11〇1。 在已偵測到前置碼並且確認,以及已知下一數據模組 内之付號到達時間以及符號處理時間之間之時序偏差之 後’孩付唬同步器模組4〇3將釋放出數據使得符號釋放能匹 配於符號處理。 根據上述之各實施例,所送出之各符號之決定結果係 解調變器之輸出移至該梳形濾波器之輸出。根據一特 歹不戸' 把例’係使用半軟體譯碼器,該半軟體譯碼器在包括 92326修正本 L. imm \ ..…7 ………,.. v 祓數個1位兀寬度暫存器之該譯碼器之"延遲路徑"中導入 二進位化操作。該半軟體譯碼器之輸出為根據所送出之I 置碼順序而排列於預期值+1(由所送出之】所组成之長前=Corresponding thresholds of the binary preamble values are respectively used to separate the preambles. Λ -月丨J According to the wireless L A Ν specification, the detection of the long and short preambles can be detected. The preamble detection operation applies a first threshold T h 1 and a second threshold τ h 2 . According to the expected decoding "vector round-value + 1/-1, the two thresholds differ only in the sign of positive and negative. 乂In step 11〇2 or step 1107, long or short is detected first. After the code is set, steps 11〇3_11〇6 or steps 1(10)-(1)k "reliability check" are repeatedly performed for each preamble. The long or short preamble on the side cannot be confirmed in a certain loop of the confirmation loop formed by the step 1103_1106 of the long preamble confirmation operation or the step 1108-1111 of the short preamble confirmation operation. The comparator jumps back to step 11〇1. After the preamble has been detected and confirmed, and the timing offset between the payment arrival time and the symbol processing time in the next data module is known, the child delivery synchronizer module 4〇3 will release the data. Enable symbol release to match symbol processing. According to the above embodiments, the decision result of each of the transmitted symbols is that the output of the demodulator is shifted to the output of the comb filter. According to a special trick, the 'example' uses a semi-soft decoder, which includes 92326 to correct the L. imm \ .....7 ....,.. v 祓 1 1 兀The binaryization operation is introduced into the "delay path" of the decoder of the width register. The output of the semi-soft decoder is arranged in the order of the I coded by the expected value of +1 (by the sent).
碼)或-1(由所送出之〇所組成之短前置碼)之一連串軟S 號。 ’ 上述之實施例可提供更可靠以及獲得改善之前置碼偵 測能力,但不會增加硬體複雜度。藉由將該前置碼 + 驟移至後續之處理階段,上述之各實施例之前置碼偵測^ 減少無法偵測到前置碼或錯誤地债測前置碼之出現率。 、2然本發明已以—較佳實施例揭露如上,然其並非用 以限定本發明,任何熟習此技藝者,在不脫離本發明之精 神t範圍内,當可作些許之更動與潤飾,因此本發明之保 護範圍當視後附之ψ請專利範圍所界定者為準。 [圖式簡單說明] f1圖是繪示前置碼偵測流程之流程圖; 弟2圖是繪示偵測在通訊信號内之前置碼之前置碼偵 測器之方塊圖; 、 第3圖〜不包含於如第2圖所顯示之前置碼债測器内 之澤碼器之架構之方塊圖; 第4圖是%示同步數據偵測單元之方塊圖; 弟;圖是繪示包含於如第4圖所顯示之該架構内之半軟 版1 ι〇ί1:)譯碼模組之架構之方塊圖; 弟6圖疋~不弟5圖所顯示之該譯碼模组之更詳細實施 例之方塊圖; 、 92326修正本 架構示第4圖所顯示之該架構中之遽波器模組之 第8圖是繪示第6圖所顯示之該架構中之濾波器模組之 另一種架構之方塊圖; 第9圖是繪示前置碼偵測處理流程之流程圖; 第10圖是繪示在通訊信號内偵測既定同步數據之陟 限值比較流程之流程圖;以及 ,第11圖是繪示在通訊信號内偵測兩種不同前置碼之更 詳細之臨限值比較流程之流程圖。 [主要元件符號說明] 2 0 0前置碼偵測器 204解展頻器 3 0 0譯碼器 302、702輸出信號 304、305延遲方塊 401通訊信號 201通訊信號 205解調變器 301、301a、301b 輸入信號 303 、 306 EXOR閘One of the code S) or -1 (the short preamble consisting of the sent 〇) is a series of soft S numbers. The above embodiments provide more reliable and improved preamble detection capabilities without increasing hardware complexity. By shifting the preamble + to the subsequent processing stage, the preamble detection of each of the above embodiments reduces the occurrence rate of the preamble or the erroneously measured preamble. The present invention has been disclosed in the above-described preferred embodiments, and is not intended to limit the present invention. Any one skilled in the art can make some modifications and refinements without departing from the spirit of the invention. Therefore, the scope of the invention is defined by the scope of the appended claims. [Simple diagram of the diagram] The f1 diagram is a flow chart showing the preamble detection process; the second diagram is a block diagram showing the preamble detector before detecting the preamble in the communication signal; 3 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图The block diagram of the architecture of the semi-soft version 1 ι〇ί1:) decoding module included in the architecture as shown in FIG. 4; the decoding module shown in the figure 6 Block diagram of a more detailed embodiment; 92326, the eighth embodiment of the chopper module in the architecture shown in FIG. 4 is a filter module in the architecture shown in FIG. Block diagram of another architecture of the group; FIG. 9 is a flow chart showing the processing flow of the preamble detection process; FIG. 10 is a flow chart showing the process of comparing the threshold value of the predetermined synchronization data in the communication signal And, Figure 11 is a flow chart showing a more detailed threshold comparison process for detecting two different preambles in a communication signal. [Main component symbol description] 2 0 0 preamble detector 204 despreader 3 0 0 decoder 302, 702 output signal 304, 305 delay block 401 communication signal 201 communication signal 205 demodulation transformer 301, 301a , 301b input signal 303, 306 EXOR gate
400同步數據偵測模組 4 0 3同步模組 404巴克匹配濾波器(BMF)模組 405差分BPSK解調變器(DEM)模組 406軟體譯碼器(SDS)模組4〇7梳形濾波器(c〇F)模組 408臨限控制器 500譯碼器模組 501a、501b輸入信號 503二進位化部份 504、505 延遲部份 601、602、603、604、605、606 暫存器 701數據符號 703、803加法器 92326修正本 18400 Synchronous Data Detection Module 4 0 3 Synchronization Module 404 Buck Matched Filter (BMF) Module 405 Differential BPSK Demodulation Transformer (DEM) Module 406 Software Decoder (SDS) Module 4〇7 Comb Filter (c〇F) module 408 threshold controller 500 decoder module 501a, 501b input signal 503 binary portion 504, 505 delay portion 601, 602, 603, 604, 605, 606 temporary storage 701 data symbol 703, 803 adder 92326 correction of this 18
一丄一一… Κ5Λ421 704、804 延遲元件 800梳形濾波器 8 0 5除法器 7 0 5、7 0 6 乘法器 801譯碼器輸出One-to-one... Κ5Λ421 704,804 delay element 800 comb filter 8 0 5 divider 7 0 5,7 0 6 multiplier 801 decoder output
92326 倐 iL太92326 倐 iL too