1229512 玖、發明說明： 【發明所屬之技術領域】 本發明係關於一種通道連線品質之估算方法及裝置， 尤指一種以衰耗（fading)值修正一通道品質之估算值，據此 獲得一更為接近該通道實際連線品質之估算量。 【先前技術】 在無線傳輸的系統中，無論是對訊號發射或接收的一 方而言，通道（channel)的連線品質係為一重要參數，舉例 而e，若一接取點（access p〇int)與其有效涵蓋範圍内的複 數節點（node)之連線品質能精確量測得到，則該接取點之 媒介存取控制層（media access c〇ntr〇1 layer，MAC)的通訊 協定中便能於眾多節點之間進行適當切換，據以適切地降 低/提南訊號之傳輸率。 基本上，連線品質係取決於量測所得之訊雜比（snr) ，雖然該參數在定義連線品質方面具有一定的代表性，但 對於要求更為精準的系統而言，仍難以利用單一的訊雜比 代表整體連線品質。例如因為多路徑（muhipath)所導致的 衰耗現象（fading)以及其它眾多因素，均有可能影響量測到 的連線品質。 由此可見，有關連線品質的估算仍有待進一步提出具 體可行的解決方案。 【發明内容】 有鏜於此，本發明之主要目的係提供一種可提高精準 度的通道連線品質估算方法，其中因多路徑（muhipath)所 導致的衰耗（fading)係加入於通道品質之估算過程中，令通 1229512 道品質之估計更為精準。 為達成前述目的 包含有： 本發明通道連線品質之估算方法係 ^通道雜訊量(B)估算，係利用在一通道中傳輸之各個正 X ^ ^ ^ (orthogonal frequency division multiplexing OFDM)封包内所含之長 g’ 我 J丨練符 70 (long training symbol)以 估鼻出該通道雜訊量（B); 通道增益值（A)計算 (subcairier)增益的絕對值， 值（A); ’係累積該通道之次載波 以計算出代表整體通道之增益 一通道衰耗值（F)估算，係依據該it道其次載波之增益值 估算出該通道之衰耗值（F); 通道連線品質計算，係計算前述通道增益值（a)與衰耗 值(F)兩者之差值(A_F)，定義該差值與雜訊量⑻之比值為 通道之連線品質。 由前述詳細步驟中係可得知該通道之衰耗因素（f)係已 反應於通道增益值⑷上’如此—來將令計算得到的連線品 質更為精確。 【實施方式】 本發明係為一種通道連線品質之估算方法及裝置，係 適用於-傳輸有正交多工分頻（GFDM)封包之無線通道。就 每-個正交多工分頻封包而言’冑包的前端係定義有一實 體層整合協定(PLCP)節段，於該節段中具有十個已知的短 符元（short symbol)及兩個長訓練符元（1〇ng打心― 1229512 symb〇i)，根據正交多工分頻之設定襟 其特定音羞 旱各個符元係具有 〜、義，在此不加贅述。在本發 ^ 練符元係用來計算出-些適當參數德=述兩個長訓 估外之田+ 乂供後續的連線品質 ° °在須聲明的一點是：就單-個通、首夾县 係視為由N個次載波(―) ：通道來看’ ^ - 44 ^ ^ ^ 、、’成，各次載波即代 =值。以，.lla通道為例，共計有“個次載 波而备中的52個次載波則較常被使用。 請參閱第一、二圖所示，當一個 中的笫一墙, 個正乂夕工分頻封包當 兴值# 練符元依序傳輸時，各個:欠載波之增 可測量獲得。於圖式中’次載波其增 號為^⑽^為少下標中的第—字元由^^ 即表不具有1^個次載波（即次載波之序號），而第二字元 僅1或2則表示該增益值是根據第一或第二訓 獲得。 2於-個理想的通道而言，其内部係無任何雜訊存在 ，故前後利用第一、第二長訓練符元所測得之次載波增益 值旧1，1."1^，1)(111，2〜1^，2)，應該是相等的。因此由第一、 :圖所示各次載波之增益值來看，若同一次載波前後兩次 增益值存在有差異，即顯示該通道内部存在雜訊，差值愈 大’即代表當中所含的雜訊強度愈強。 因此，本發明首先即估算該通道之雜訊量（Β)，其估算 方式係先計算各次載波其前後兩次增益值（Ηκ ι、2)之差 值1再對該差值取絕對值，並將各絕對值累積相加，藉其 表示為整體雜訊量（Β)，表示式為：万=I;7^ 一仏2。 k-\ ’ 6 1229512 ^ 請參閱第三圖，以各個次載波來看，為更能確切表示 5亥次載波之增益，係將前後兩次增益值（Hq、Ηκ，2)相加再 取其平均值’利用此平均值表示該次載波之增益估計值Ηκ ，如方程式所示。 右將則述所有次載波的增益估計值Ηκ進一步累積總 =，則$體通道增益（A)係可計算出來，整體增益值係表示 為：Σ|Α| 〇 k=l 為了反應出多路徑傳輸所引起的衰耗現象（fading)，本 發明係特地將此因素納入品質估算過程中。通道之整體衰 耗值（F)係以下式計算：F=$圮卜差。1229512 发明 Description of the invention: [Technical field to which the invention belongs] The present invention relates to a method and device for estimating channel connection quality, and in particular to a method for correcting an estimated value of a channel quality by a faded value, and obtaining a It is closer to the estimated quantity of the actual connection quality of the channel. [Prior art] In a wireless transmission system, whether it is a signal transmitting or receiving party, the channel connection quality is an important parameter. For example, e, if an access point (access p) int) can accurately measure the connection quality with the plurality of nodes within its effective coverage range, then the communication protocol of the access point's media access control layer (MAC) layer It is then possible to switch between many nodes appropriately, so as to appropriately reduce / provide the transmission rate of the South signal. Basically, the connection quality depends on the measured signal-to-noise ratio (snr). Although this parameter is representative in defining the connection quality, it is still difficult to use a single unit for a system that requires more precision. The noise-to-noise ratio represents the overall connection quality. For example, the fading caused by muhipath and many other factors may affect the measured connection quality. It can be seen that the estimation of connection quality still needs to further propose concrete feasible solutions. [Summary] With this in mind, the main purpose of the present invention is to provide a method for estimating channel connection quality that can improve accuracy, in which the fading caused by muhipath is added to the channel quality. During the estimation process, the estimation of the quality of 1229512 channels is more accurate. In order to achieve the foregoing objectives, the method for estimating channel connection quality of the present invention is ^ channel noise amount (B) estimation, which uses each positive X ^ ^ (orthogonal frequency division multiplexing OFDM) packet transmitted in a channel. The long g 'I contains training symbols 70 (long training symbol) to estimate the noise of the channel (B); the channel gain value (A) calculates the absolute value of the subcairier gain, the value (A); 'It accumulates the secondary carriers of the channel to calculate the gain representing the overall channel. A channel attenuation value (F) is estimated. It is based on the gain value of the secondary carrier of the it channel to estimate the attenuation value (F) of the channel. The line quality calculation is to calculate the difference (A_F) between the channel gain value (a) and the attenuation value (F), and define the ratio of the difference to the noise amount ⑻ as the connection quality of the channel. It can be known from the foregoing detailed steps that the attenuation factor (f) of the channel has been reflected on the channel gain value ⑷ 'so as to make the calculated connection quality more accurate. [Embodiment] The present invention is a method and device for estimating channel connection quality, and is suitable for-transmitting a wireless channel with orthogonal multiplexed frequency division (GFDM) packets. For each orthogonal multiplexed frequency division packet, the front end of the packet is defined with a physical layer integration protocol (PLCP) segment, which has ten known short symbols and The two long training symbols (10ng heartbeat-1229512 symb〇i), according to the orthogonal multiplexing frequency setting, its specific sounds and symbols have the meanings of ~, meaning, and will not be repeated here. In this book, the practice element system is used to calculate some appropriate parameters. Describe the two long training estimates outside the field + 乂 for subsequent connection quality ° ° One point to be declared is: on the single-pass, The first folder county is considered to be composed of N sub-carriers (-): from the channel's perspective, ^-44 ^ ^ ^,, and 成, and each sub-carrier is equal to the value. Taking the .lla channel as an example, there are "sub-carriers in total and 52 sub-carriers in standby are more commonly used. Please refer to the first and second diagrams. When the frequency-dividing packet when the value ## symbol is transmitted in sequence, each: the increase of the under-carrier can be measured. In the figure, the sub-carrier is incremented by ^ ⑽ ^ as the first character in the subscript. ^^ means that there are no 1 ^ subcarriers (that is, the number of the subcarriers), and the second character is only 1 or 2 to indicate that the gain value is obtained according to the first or second training. 2 is an ideal As far as the channel is concerned, there is no noise in its internal system, so the subcarrier gain values measured before and after using the first and second long training symbols are old 1,1. &Quot; 1 ^, 1) (111, 2 ~ 1 ^, 2) should be equal. Therefore, from the gain values of the carrier waves shown in the first and second figures, if there is a difference in the gain value between the two times before and after the same carrier wave, it indicates that there is noise in the channel. The larger the difference, the stronger the noise intensity contained in it. Therefore, the present invention first estimates the noise amount (B) of the channel, and the estimation method is first Calculate the difference between the two gains (Ηκ ι, 2) of each carrier, and then take the absolute value of the difference, and add up the absolute values, and use it as the overall noise (B), The expression is: 10,000 = I; 7 ^ 1 仏 2. k- \ '6 1229512 ^ Please refer to the third figure. In view of each subcarrier, in order to more accurately represent the gain of the 5th subcarrier, two Add the sub-gain values (Hq, Ηκ, 2) and take the average value. Use this average value to represent the estimated gain of the sub-carrier Ηκ as shown in the equation. The right-hand side will describe the gain estimates of all sub-carriers Ηκ further. Cumulative total =, the $ body channel gain (A) can be calculated, and the overall gain value is expressed as: Σ | Α | 〇k = l In order to reflect the fading caused by multipath transmission, the present invention This factor is specifically included in the quality estimation process. The overall attenuation value (F) of the channel is calculated by the following formula: F = $ 圮 卜 差.
女=ι N 差，在前式中，整體通道增益（A)與次載波數目（N)之比值 #，其代表涵義係指所有次載波之平均增益值。因此若將 各-人載波之增盈估計值（Ηκ)與該平均增益值芷相減，該差 Ν 值所代表的即視為是次載波之變動程度。當各個次載波的 菱動程度加以累積後，則通道整體之衰耗值（F)即可獲得。 、、根據刚述說明，有關該通道的各項參數，包含通道增 益值（Α)、通道雜訊量、及通道衰耗值等均可由次載 =之增益值（HK1)(HK，2)計算得到。在具備前述三項參數的 條件下，該通道之連線品質（LQ)係可定義為^ = 由Female = ι N difference. In the previous formula, the ratio of the overall channel gain (A) to the number of subcarriers (N) #, which means the average gain value of all subcarriers. Therefore, if the estimated gain (Ηκ) of each individual carrier is subtracted from the average gain value, the difference N represents the degree of change of the secondary carrier. When the degree of jitter of each sub-carrier is accumulated, the overall channel attenuation value (F) can be obtained. According to the description just now, the parameters of the channel, including the channel gain value (Α), the channel noise level, and the channel attenuation value, can be changed by the gain value of the secondary load = (HK1) (HK, 2) Calculated. With the above three parameters, the connection quality (LQ) of the channel can be defined as ^ = by
B 則述式子中不難看出，本發明所定義之連線品質不再只是 單純的訊雜比（即A/B)，而是將通道衰耗值（1?)考量進去 ，因此特別是對一個增益變化幅度大的通道而言，其連線 品質更可確切地反映出來。 叫參閱第四圖所示，為本發明通道連線品質之估算裝 1229512 塊圖，包含有-通道增益估計單元。〇)、 。/數U單元（20)及-連線品f計算單元（3〇) 中的=述通道增益估計單元（1〇)係依據每個資料封包 增益值第二長訓練符疋，估算出通道中各個次載波其 (上數？單元*(2〇)獲得各個次載波之增益值 通道雜即可依據前述各公式計算出通道增益值⑷、 步輸及二耗值(F)等，這些參數係 异早，依據如四一式 °十异出通道之連線品質。 β 際，亡詳細說明’本發明係當每次接收資料封包之 、 了估鼻出當時的通道連線α暂丄. 道目前之資料傳輸狀態，對於：：°"P時地呈現通 機動地調整其資料傳輸率；心=的一方而言，係可 考量.;估算所得之連㈣:更^^ 從而符合:ί專:::確爰已:法具:著的新穎性及進步性， f圖式簡單說明】 提出申請。 (~ )圖式部分 第圖·係本發明於一通谨由+ 練符元挪量獲得之增益示意圖道"各次載波根據第-長訓 第二圖··係本發明於一 練符元挪量獲得之增益示意圓；^各次載波根據第二長訓 1229512 第三圖··係本發明於一通道中各次載波之平均增益示 意圖。 第四圖：係本發明通道連線品質估算裝置之結構方塊 圖。 (二）元件代表符號 (A) 通道增益值 (B) 通道雜訊量 (Ηκ)次載波之增益估計值 (F)通道衰耗值 (LQ)連線品質 (1 0 )通道增益估計單元 (20)參數計算單元 (3 0 )連線品質計算單元 參 9It is not difficult to see in the formula that the connection quality defined by the present invention is no longer just a simple signal-to-noise ratio (ie, A / B), but a channel attenuation value (1?) Is taken into account, so in particular For a channel with a large gain change, its connection quality can be more accurately reflected. Please refer to the fourth figure, which is a 1229512 block diagram for estimating the channel connection quality of the present invention, including a -channel gain estimation unit. 〇),. The channel gain estimation unit (10) in the / count U unit (20) and the -connected product f calculation unit (30) estimates the channel length based on the second long training symbol 疋 of each data packet gain value. For each sub-carrier, (the number? Unit * (20)) can obtain the gain value of each sub-carrier. The channel gain can be calculated according to the foregoing formulas. Different early, based on the connection quality of the ten different outgoing channels such as the four-in-one formula. Β, the detailed description of the present invention 'Each time the data packet is received, the present invention estimates the channel connection at that time. For the current data transmission status, for: ° " P presents time to time to adjust its data transmission rate flexibly; for the party with heart =, it can be considered.; Estimated flail: more ^^ and thus meet: ί Dedicated ::: Confirmation: Method: Novelty and Progressiveness of the Book, f Schematic Simple Description] File an application. (~) Schematic part of the drawing. This invention is in one pass. Schematic diagram of the obtained gain " The carrier of each time according to the second-long training second picture ... This invention is obtained by the amount of symbols in a practice The obtained gain schematic circle; ^ each carrier according to the second long training 1229512 the third picture ... is a schematic diagram of the average gain of each carrier in a channel of the present invention. The fourth picture: the channel connection quality estimation device of the present invention Block diagram of the structure. (II) Symbols of component representation (A) Channel gain value (B) Channel noise (Ηκ) Subcarrier gain estimation value (F) Channel attenuation value (LQ) Connection quality (1 0) Channel Gain estimation unit (20) Parameter calculation unit (3 0) Connection quality calculation unit Reference 9