201105057 六、發明說明： 【發明所屬之技術領域】 本伽系指-種用於無線通訊系統之訓練序列產生方法及訓練 序列產线置’尤指-種可有效減少通道估_需之資料傳輸量的 訓練序列產生方法及訓練序列產生裝置。 【先前技術】201105057 VI. Description of the invention: [Technical field to which the invention pertains] This gamma system refers to a training sequence generation method for a wireless communication system and a training sequence production line setting, in particular, a type of data transmission that can effectively reduce channel estimation. A training sequence generation method and a training sequence generating device. [Prior Art]
正交分頻多工（Orthogonal Frequency Division Multiplexing， OFDM)調憂技術是種「多載波調變」（MuitiCarrierM^uiatiQn， MCM)傳輸方法，其基本的概念是把一個高速傳輸速率的數據串 流，切割成許多平行且較低速的傳輸速率串流，並且把每一個次串. 流調變到不同的次載波（Subcarrier)上。在此情形下，符元（Symb〇1) 時間變得夠長，所以通道導致的延遲變得只是符碼時間的一小部 伤’因而可消除或減少符碼間干擾（Inter Symb〇i jnterference )，有 效提昇頻譜利用率（Spectrum efflciency)，增加系統的資料傳輸量。 因此’正交分頻多工調變技術已廣泛地用於許多無線通訊系統中， 例如無線區域網路（WJreless Local Area Network，WLAN )就是其 中之一’相關無線區域網路通訊協定標準包含！ΕΕΕ8〇2.11&、正EE 802.11b、IEEE 802.11g，至IEEE 802.11η等，皆採用正交分頻多工 調變技術。其中，與IEEE802.11a/g標準不同的是，正ΕΕ802.11η 201105057 標準使用多輸入多輸出⑽物nputMuU__，m_ & 術及其它新功能，大幅改善了資料速率及傳輸吞吐量 (Throughput)，同時，通道頻寬由2〇MHz增加為4〇MHz。< 多輸入多輸出正交分頻多工系統的典型架構係如第ι圖所示， 其傳輸端與接㈣間透過多個收發天線進行無線峨傳遞，且在進 行傳輸前’傳输端與接收端無法取得兩者間通道狀態的相關資訊。 _ 因此，通道狀態的估測就顯得特別重要。 、-般而言，在第1騎類多重路徑之正交分頻多^统下，通 道估測（ChannelEstimation)係透過訓練序列（TrainingSequence) 2完成。魏’由傳輸端發送訓練序列或產生訓練搁做帶於封包， 當接收端收到訓練序列或訓練欄位時，接收端可根據其内容，進行 通道估測運算，以判斷通道狀態。 ® 舉例來s兒’請參考第2圖’第2圖為習知ffiEE8〇2如標準之 封包格式示意圖。如第2圖所示’正EE8〇2 lln標準之封包係由一 月’J置資料（Preamble)與待傳輸之資料組合而成，前置資料位於每 -封包的最前端，接續為待傳輸之資料。另外，前置資料為混合格 式，可向下相容於IEEE802.11a/g標準之無線區域網路裝置，所包 含的欄位依序為傳統短訓練欄位L-STF (Legacy Short Training Fidd)、傳統長訓練欄位 L-LTF (Legacy Long Training Field)、傳統Orthogonal Frequency Division Multiplexing (OFDM) is a kind of "multi-carrier modulation" (MuitiCarrierM^uiatiQn, MCM) transmission method. The basic concept is to stream data at a high-speed transmission rate. It is cut into a number of parallel and lower speed transmission rate streams, and each substring stream is modulated onto a different subcarrier. In this case, the symbol (Symb〇1) time becomes long enough, so the delay caused by the channel becomes only a small part of the symbol time', thus eliminating or reducing inter-symbol interference (Inter Symb〇i jnterference ), effectively improve spectrum utilization (Spectrum efflciency), increase the amount of data transmission of the system. Therefore, the 'orthogonal frequency division multiplexing modulation technology has been widely used in many wireless communication systems. For example, the WJreless Local Area Network (WLAN) is one of them. The related wireless local area network protocol standard is included! ΕΕΕ8〇2.11&, EE 802.11b, IEEE 802.11g, IEEE 802.11η, etc., all adopt orthogonal frequency division multiplexing modulation technology. Among them, unlike the IEEE802.11a/g standard, the 802.11n 201105057 standard uses the multi-input multi-output (10) object nputMuU__, m_ & and other new functions to greatly improve the data rate and throughput (Throughput). At the same time, the channel bandwidth is increased from 2 〇 MHz to 4 〇 MHz. < Typical architecture of multi-input multi-output orthogonal frequency division multiplexing system is shown in Figure ι. The transmission end and the connection (4) are wirelessly transmitted through multiple transceiver antennas, and the transmission end is transmitted before transmission. Information about the channel status between the two cannot be obtained from the receiving end. _ Therefore, the estimation of the channel state is particularly important. In general, under the orthogonal frequency division of the first riding multiple path, the channel estimation (ChannelEstimation) is completed by the training sequence (TrainingSequence) 2. Wei's sends a training sequence from the transmitting end or generates training to carry the packet. When the receiving end receives the training sequence or the training field, the receiving end can perform channel estimation calculation according to its content to determine the channel status. ® For example, please refer to Figure 2'. Figure 2 is a schematic diagram of the conventional ffiEE8〇2 standard package format. As shown in Figure 2, the package of the EE8〇2 lln standard is composed of the January 'Ja data (Preamble) and the data to be transmitted. The pre-data is located at the forefront of each packet, and is connected to be transmitted. Information. In addition, the pre-data is a mixed format, which can be compatible with the IEEE802.11a/g standard wireless local area network device. The included fields are in the traditional short training field L-STF (Legacy Short Training Fidd). , traditional long training field L-LTF (Legacy Long Training Field), tradition
凡號搁位L-SIG (Legacy Signal Field)、高吞吐量訊號欄位ht-SIG 201105057L-SIG (Legacy Signal Field), high throughput signal field ht-SIG 201105057
(High-Throughput Signal Field)、高吞吐量短訓練攔位 fjT-STF (High-Throughput Short Training Field )以及 N 個高吞吐量長訓練 攔位 HT-LTF (High-Throughput Long Training Field)。傳統短訓練攔 位L-STF用於封包起始偵測（start-of-packet Detection)、自動增益 控制（Automatic Gain Control，AGC )、初始頻率偏移估測（Frequency(High-Throughput Signal Field), High-Throughput Short Training Field, and High-Throughput Long Training Field (HT-LTF). Traditional short training block L-STF is used for start-of-packet detection, automatic gain control (AGC), initial frequency offset estimation (Frequency
Offset Estimation)及初始時間同步（Time Synchronization);傳統長 訓練欄位L-LTF用於精密之頻率偏移估測及時間同步；傳統訊號攔 位L-SIG攜帶資料速率及封包長度之資訊。高吞吐量訊號攔位 HT-SIG攜帶資料速率之資訊，並且用於自動偵測封包屬於混合格式 或傳統格式；高呑吐量短訓練欄位HT-STF用於自動增益控制；以 及高吞吐量長訓練欄位HT-LTF用於多輸入多輸出之通道估測，使 接收端可據以判斷通道狀態。 高吞吐量長訓練攔位HT-LTF的態樣（Pattem)係為業界所熟 知在此不贅述，而根據其功能，咼吞吐量長訓練欄位Ht ltf可 進-步分為兩類。第-類為資料高吞吐量長鱗欄位，用來估測當 前（資料)所使用之通道的狀態，其數量U由空間時間束(Space Time Stream)的數量1^阳所決定，如第3圖所示。第二類為延伸 (Extension)高吞吐量長訓練攔位’用來偵測未使用之通道的額外 空間維度（SpatialDimension)，其數量Neltj^由待偵測之額外空 間維度的數量ness所決^ ’且兩者關係相同於第3圖所示之數量 NDLTF與數量NSTS之關係。此外，由於^標準至多支援 四個天線，因此，NDLTF、neltf皆小於等於4。 201105057 另-方面，為了降低通道估測的複雜度，習知技術係將高吞吐 量長訓練欄位HT-LTF設計為由單一符元（Symb〇1)賊予不同之權 重與延遲而產主，其作法係將高呑吐量長訓練搁位抓抓與一延 伸碼（SpreadingCode)矩陣相乘，以產生複數個獨立的相位，使接 收端可據以進行通道估測。隨著超高吞吐量（Veryffigh ,Throughput，VHT)無線通訊技術的不斷發展，新一代的無線區域 •網路標準臓8〇2.llac提供較高的通道頻寬（80MHZ)，且支援更 夕（四支以上）的天線。在此情形下，高吞吐量長訓練搁位的設計 就變得更為關鍵。然而’習知使用延伸碼矩陣轉換高吞吐量長訓練 攔位的機齡彡天線（四支社）_下會大幅增加傳輸資料量。 有鑑於此，實有必要設計一種新的長訓練搁位產生方式，以利新一 代的無線區域網路標準之實現。 【發明内容】 因此，本發明之主要目的即在於提供用於無線通訊系統之訓練 序列產生方法及訓練序列產生裝置。 本發明揭露—種訓練序列產生方法，用於-無線通訊系統之-傳輸端衫生_序列’該傳輸端包含複數健射天線，該訓練序 列產生方法包含有將複數爾元之每一符元分割為複數個子符元， 该複數個子槪之數量與該複數個發射天線之數量_ ;依序轉換 201105057 對應於每一符元之相同位置的複數個子符元，以產生複數個訓練資 料；以及根據該複數個訓練資料，產生頻率互為獨立之複數個訓練 序列’分別承載於該複數個發射天線之一發射天線所發送之封包中。 本發明揭露一種訓練序列產生裝置，用於一無線通訊系統孓一 傳輸端中產生訓練序列，該傳輸端包含複數個發射天線，該訓練序 列產生裝置包含有一微處理器；以及一記憶體，用來儲存一程式， 該程式用來指示該微處理器執行以下步驟：將複數個符元之每一符 元为割為複數個子符元，該複數個子符元之數量與該複數個發射天 線之數量相同；依序轉換對應於每一符元之相同位置的複數個子符 元，以產生複數個訓練資料；以及根據該複數個訓練資料，產生頻 率互為獨立之複數個訓練序列，分別承載於該複數個發射天線之一 發射天線所發送之封包中。 【實施方式】 呀參考第4圖，第4圖為本發明實施例用於一無線通訊系統之 一傳輸端中的一訓練序列產生流程4〇之示意圖。該無線通訊系統為 一多輸入多輸出正交分頻多工系統，亦即該傳輸端包含有複數個發 射天線。靖序随纽程4〇用來產生娜序列，包含以下步驟： 步驟400 :開始。 步驟402 :將複數個符元之每一符元分割為複數個子符元，該 複數個子符元之數量與該複數個發射天線之數量相 201105057 同。 步驟404 .依序轉換對應於每一符元之相同位置的複數個子符 元，以產生複數個訓練資料。 步驟概：根_複_辑_，產生解互_立之複數 偏1丨練序列’分別承載於該複數個發射天線之-發 射天線所發送之封包中。 步驟408:結束。 樹康訓練序列產生流牙呈40 ’本發明係根據傳輸端所包含的發射 天線數’將複數個特定符元的每一符元分割為與發射天線數相同數 量的子符元。接著’依序對所有符元中對應於相同位置的子符元進 行轉換，如逆向離散傅利葉轉換（InverseDiscreteFmirierOffset Estimation) and Time Synchronization; the traditional long training field L-LTF is used for precise frequency offset estimation and time synchronization; the traditional signal intercept L-SIG carries information on data rate and packet length. High-throughput signal interception HT-SIG carries data rate information and is used to automatically detect packets in mixed or traditional formats; high throughput short training field HT-STF for automatic gain control; and high throughput long The training field HT-LTF is used for multi-input and multi-output channel estimation, so that the receiving end can judge the channel status. The high-throughput long training block HT-LTF (Pattem) is well known in the industry and will not be described here. According to its function, the long-running training field Ht ltf can be further divided into two categories. The first type is the high-throughput long-scale field of the data, which is used to estimate the state of the channel used by the current (data). The number U is determined by the number of space time streams (1) Yang, such as Figure 3 shows. The second type is the Extension High-Performance Long Training Blocker's Spatial Dimension for detecting unused channels. The number of Neltj^ is determined by the number of additional spatial dimensions to be detected. 'The relationship between the two is the same as the relationship between the number of NDLTFs shown in Figure 3 and the number NSTS. In addition, since the ^ standard supports up to four antennas, both NDLTF and neltf are less than or equal to four. 201105057 On the other hand, in order to reduce the complexity of channel estimation, the conventional technology designed the high-throughput long training field HT-LTF as a single symbol (Symb〇1) thief with different weights and delays. The method is to multiply the high sputum long training position by a spreading code (Spreading Code) matrix to generate a plurality of independent phases, so that the receiving end can perform channel estimation. With the continuous development of ultra-high-throughput (Veryffigh, Throughput, VHT) wireless communication technology, the new generation of wireless area and network standard 臓8〇2.llac provides higher channel bandwidth (80MHZ) and supports even more (four or more) antennas. In this case, the design of high throughput long training shelves becomes even more critical. However, the conventional use of extended code matrix to convert high-throughput long training blocks to the age of the antenna (four branches) will greatly increase the amount of data transmitted. In view of this, it is necessary to design a new long training placement method to facilitate the implementation of the new generation of wireless local area network standards. SUMMARY OF THE INVENTION Accordingly, it is a primary object of the present invention to provide a training sequence generating method and a training sequence generating apparatus for a wireless communication system. The invention discloses a training sequence generating method, which is used for a wireless communication system, and includes a plurality of smart antennas, and the training sequence generating method includes each symbol of a plurality of complex elements. Dividing into a plurality of sub-symbols, the number of the plurality of sub-frames and the number of the plurality of transmitting antennas; sequentially converting 201105057 to a plurality of sub-symbols corresponding to the same position of each symbol to generate a plurality of training materials; According to the plurality of training data, a plurality of training sequences that generate independent frequencies are respectively carried in a packet sent by one of the plurality of transmitting antennas. The present invention discloses a training sequence generating apparatus for generating a training sequence in a transmission end of a wireless communication system, the transmission end comprising a plurality of transmitting antennas, the training sequence generating device comprising a microprocessor, and a memory for using And storing a program, the program is used to instruct the microprocessor to perform the following steps: cutting each symbol of the plurality of symbols into a plurality of sub-symbols, the number of the plurality of sub-symbols and the plurality of transmitting antennas The same number; the plurality of sub-symbols corresponding to the same position of each symbol are sequentially converted to generate a plurality of training materials; and the plurality of training sequences having independent frequencies are generated according to the plurality of training materials, respectively One of the plurality of transmit antennas is in a packet transmitted by the transmit antenna. [Embodiment] Referring to FIG. 4, FIG. 4 is a schematic diagram of a training sequence generation process in a transmission end of a wireless communication system according to an embodiment of the present invention. The wireless communication system is a multiple input multiple output orthogonal frequency division multiplexing system, that is, the transmission end includes a plurality of transmitting antennas. The sequence is used to generate the sequence of Na, which consists of the following steps: Step 400: Start. Step 402: Divide each symbol of the plurality of symbols into a plurality of sub-symbols, and the number of the plurality of sub-symbols is the same as the number of the plurality of transmitting antennas. Step 404. Convert a plurality of sub-symbols corresponding to the same position of each symbol to sequentially generate a plurality of training materials. The step is as follows: root_complex_series_, generating a complex _ 立 立 立 立 立 立 立 立 立 立 立 立 立 立 立 立 立 立 立 立 立Step 408: End. The tree training sequence generates a stream of 40'. The present invention divides each symbol of a plurality of specific symbols into the same number of sub-symbols as the number of transmitting antennas according to the number of transmitting antennas included in the transmitting end. Then, 'sub-symbols corresponding to the same position in all symbols are sequentially transformed, such as inverse discrete Fourier transform (InverseDiscreteFmirier)
Transform) ^ amm C Frequency Domain ) (TimeTransform) ^ amm C Frequency Domain ) (Time
Domain)資料，進而產生訓練資料。最後，將所得的訓練資料適當 排列，以產生頻率互為獨立的訓練序列，並將之分別承載於一發射 天線所發送之封包中。 為清楚說明訓練序列產生流程40，以下以包含有六個發送天線 (傳輸路徑）TX1〜TX6之傳輸端為例，配合第5圖及第6圖，說 明如何產生對應之訓練序列。在第5圖中，實線區塊表示符元，虛 線區塊表示子符元，斜線區域表示被選取之子符元，IFFT表示逆白. 離散傅利葉轉換，以及LTF1〜LTF6表示訓練資料。因此，由第5 圖可知，每一符元被分為六個子符元，且相同位置之子符元會透過 201105057 勒離散翻轉換IFFT，轉換為訓練資料 LTF1〜LTF6。也就是 +丨”東貝料LTF1係對所有符元的第一個子符元進行逆向離散傅 利葉轉換IFFT而仔’訓練資料咖係對所有符元的第二個子符元 進行逆向離散傅雜附而得，並以鋪推。 接下來，在第6圖中，訓練資料LTF1〜LTF6會被適當排列而 成為:種:練，’分別承載於發送天線m〜τχ6的輸出封包 //主思的疋’第6圖係顯示剌練序列.的内容，故省略了封包中 -匕攔位此外’由第6圖可知，發送天線在相同時間 所輸出的訓練胃料係為減，例如當發送天線τχι發送訓練資料 時，發送天線TX2〜TX6係發送訓練資料LTF2〜LTF6時。換 曰之’雖㈣6 ®所示的六個辑相皆是㈣練資料[顶〜取6 排列而成，但這六個訓練序列在相同時間的内容係為城，亦即類 懈刀頻多工（Frequency Divisi〇nMultiplexing)技術中的正交特 f。藉此，接收端可輕易透過反向運作，執行通道估測；更重要的 是，封包中所需的訓練欄錄即為天線數，可有效減少多天線系統 下的資料傳輸1 ’特別是新—代超高吞吐量的無線通訊系統。 為了驗證本發明可行性，可透過適當模擬方式，得出如第7圖 所不之模擬結果’用以表示—6送2收系統之通道估_果。其中， X軸表示訊雜比（Signal to Noise Ratio) SNR，y軸表示均方誤差 MSE (Mean Square Error) ° 201105057 需注意的是’第5圖及第6圖_來說明本發明之精神，本領 域具通常知識者當可據以做不同之修飾，而不限於此。例如，在; 割航時’可奸紐為單位，等分每—符元。細，若無法等分刀， 則使每一子符元的大小相近亦可。 驟，本發明中，苟丨丨練序列的排列或產生方式不限於特 :，，只要能確絲—訓練相皆包含所有訓練資料，且不同訓Domain) data to generate training materials. Finally, the obtained training materials are appropriately arranged to generate training sequences whose frequencies are independent of each other, and are respectively carried in the packets sent by a transmitting antenna. In order to clarify the training sequence generation process 40, the transmission terminal including six transmission antennas (transmission paths) TX1 to TX6 is taken as an example, and the fifth and sixth pictures are combined to illustrate how to generate a corresponding training sequence. In Fig. 5, the solid line block represents the symbol, the dotted line block represents the sub-symbol, the oblique line area represents the selected sub-symbol, the IFFT represents the inverse white. The discrete Fourier transform, and LTF1~LTF6 represent the training data. Therefore, as can be seen from Fig. 5, each symbol is divided into six sub-symbols, and the sub-symbols of the same position are converted into training materials LTF1 to LTF6 through the 201105057 Le discrete conversion IFFT. That is, +丨" East Baye LTF1 is the inverse discrete Fourier transform IFFT for the first sub-symbol of all symbols, and the training data is the inverse of the second sub-symbol of all symbols. Next, in the sixth figure, the training materials LTF1 to LTF6 will be properly arranged to become: kind: training, 'output packets respectively carried on the transmitting antenna m~τχ6//the main thinking 疋'The figure 6 shows the contents of the training sequence. Therefore, the packet is omitted. - The block is also blocked. In addition, as shown in Fig. 6, the transmission stomach output of the transmitting antenna at the same time is reduced, for example, when the transmitting antenna τχι When the training data is sent, the transmitting antennas TX2 to TX6 send the training materials LTF2 to LTF6. The six episodes shown in (4) 6 ® are (4) practice data [top ~ take 6 arrays, but these six The content of the training sequence at the same time is the city, which is the orthogonal special f in the Frequency Divisi〇n Multiplexing technology. Therefore, the receiving end can easily perform the channel estimation through the reverse operation. More importantly, what is needed in the package The training column is the number of antennas, which can effectively reduce the data transmission under the multi-antenna system. 1 'Specially new-generation ultra-high-throughput wireless communication system. In order to verify the feasibility of the present invention, it can be obtained through appropriate simulation. The simulation result shown in Figure 7 is used to represent the channel estimation of the 6-to-two-receive system. Among them, the X-axis represents the Signal to Noise Ratio SNR, and the y-axis represents the mean square error MSE (Mean Square Error) ° 201105057 It should be noted that '5th and 6th drawings _ to illustrate the spirit of the present invention, and those skilled in the art can make different modifications according to the present invention, and are not limited thereto. For example, in; In the voyage, the size of each sub-symbol is similar. The method is not limited to special:, as long as it can be confirmed - the training phase contains all training materials, and different training
二要ΓΓΐ同時間的内容係為相異即可。舉例來說，如第8圖所示， :要產生心圖之訓練序列，可將訓練資料刪〜㈣ =^^_A_deham) ’ _每—發射天線相對二 Γ各發射天軸環鏈她。例如，發射天線™ 鏈起點為訓練資料LTF1 (起點為訓練資料聰），發射天線 沿 ®之樹序歹’J。透過循環鏈架構，即使-訓練資 =w如_料LTF2與訓物_交 換）’仍可保證每一制綠皮π比 义 在相同時軸容係為減Ζ所 =_’綱訓練序列 2Γ紅—目㈣撕本發明， ’在硬體實現方面，可以軟 另一方面 序列產生流程40轉換為1 ^、幢等方式，將訓練 中，以指示微處理執行崎序線通訊裝置之一記憶體 、彳產生机私4〇之步驟。此等將訓練序 F 5： 1 11 201105057 置’應為本領域 歹i產生机程4〇轉換為適當料以實麟應之設定裝 具通常知識者所熟習之技藝。 曰^ ΓΓ絲_統，壯·延伸碼鱗轉換高吞吐 置長麟攔位的機制在多天線架構下會大幅增加傳輸資料量。相較 1下，本發騎產生御林顺包含與發送蝴目醜量之訓練 攔位，因此可降低雜雜》，提升魏辭。 ’ 综上所述，本發明係利用分頻多工技術中的正交特性，產生對 應，不同發送天_爾相，除了可有效執行通道估測，更重要 的是，封包所承載的訓練攔位數即為發送天線數，可有效減少資料 傳輸量’特财利於新-代超高吞吐量的無線通訊系統。^ 以上所述僅為本發明之較佳實施例，凡依本發明申請專利範圍 所做之均等變化婦飾，皆闕本發明之涵蓋範圍。 【圖式簡單說明】 第1圖為習知-多輸人多輸出正交分頻多工系統之示意圖。 第2圖為習知IEEE 802_lln標準之封包格式示意圖/ 第3圖為習知IEEE 8〇2.11n標準之高吞吐量長訓練棚位之數量 示意圖。 第4圖為本發明實施例一訓練序列產生流程之示意圖。 12 201105057 =圖為根據第4圖之魏序列產生流程產生訓練資料之示意 厨.〇 第6圖為根據苐5圖之訓練資料產生訓練序列之示意圖。 第7圖為本發明實制之通道估測結果示意圖。 第8圖為產生钏練序列之一實施例之示意圖。 【主要元件符號說明】Second, the content at the same time is different. For example, as shown in Fig. 8, to generate a training sequence of the heart map, the training data can be deleted ~ (4) = ^ ^ _A_deham) _ _ per transmitting antenna relative to each of the transmitting antenna axis chain. For example, the starting point of the transmitting antenna TM chain is the training data LTF1 (starting point is training data), and the transmitting antenna is along the tree sequence 歹'J. Through the cyclic chain architecture, even if the training resource = w, such as LTF2 and training material exchange, can still ensure that the π 比 每一 每一 每一 每一 每一 每一 = = = = = = = = = = = = = = = = = = Red-mesh (four) tears the invention, 'in terms of hardware implementation, it can be soft, on the other hand, the sequence generation process 40 is converted into 1 ^, building, etc., and will be trained to indicate that the micro-processing performs a memory of the S-sequence communication device. The body and the sputum produce the steps of the machine. These training orders F 5: 1 11 201105057 should be set in the field 歹i generating machine 4 〇 conversion to the appropriate material to meet the requirements of the general knowledge of the general knowledge of the skilled person.曰^ ΓΓ丝_统, Zhuang·Extended code scale conversion high throughput The mechanism of Changlin blocking will greatly increase the amount of data transmitted under the multi-antenna architecture. Compared with 1 time, this riding ride produces Yulin Shun and sends the butterfly's ugly training block, so it can reduce the miscellaneous and enhance Wei. In summary, the present invention utilizes the orthogonal characteristics in the frequency division multiplexing technique to generate correspondences, different transmission days, in addition to effectively performing channel estimation, and more importantly, the training blocks carried by the packets. The number of bits is the number of transmitting antennas, which can effectively reduce the amount of data transmission. The special wealth is beneficial to the new-generation ultra-high-throughput wireless communication system. The above is only a preferred embodiment of the present invention, and the equivalent variations of the present invention in accordance with the scope of the present invention are within the scope of the present invention. [Simple Description of the Drawing] Fig. 1 is a schematic diagram of a conventional-multiple input multi-output orthogonal frequency division multiplexing system. Figure 2 is a schematic diagram of the packet format of the conventional IEEE 802_lln standard / Figure 3 is a schematic diagram of the number of high throughput long training booths of the conventional IEEE 8〇2.11n standard. FIG. 4 is a schematic diagram of a training sequence generation process according to an embodiment of the present invention. 12 201105057 = The figure shows the training data generated according to the Wei sequence generation process of Fig. 4. Kitchen Fig. 6 is a schematic diagram of the training sequence generated based on the training data of Fig. 5. Figure 7 is a schematic diagram of the channel estimation results of the actual implementation of the present invention. Figure 8 is a schematic diagram of one embodiment of generating a training sequence. [Main component symbol description]
L-STF L-LTF L-SIG HT-SIG HT-STF HT-LTF Nsts Ndltf TX1 〜TX6 40 400、402、404、406、408 MSE SNR 傳統短訓練欄位 傳統長訓練欄位 傳統訊號攔位 高吞吐量之訊號攔位 高吞吐量之短訓練攔位 尚吞吐量之長訓練攔位 空間時間束的數量 資料高吞吐量長訓練攔位的數量 發送天線 訓練序列產生流程 步驟 均方誤差 訊雜比L-STF L-LTF L-SIG HT-SIG HT-STF HT-LTF Nsts Ndltf TX1 ~ TX6 40 400, 402, 404, 406, 408 MSE SNR Traditional short training field Traditional long training field Traditional signal block high Throughput signal interception high throughput short training block still throughput long training intercept space time beam quantity data high throughput long training block number of transmission antenna training sequence generation process step mean square error signal ratio
1313