1335994 九、發明說明： 【發明所屬之技術領域】 本發明係有關於衛星通訊，尤其是一種能改善首次 定位時間（Time To First Fix，TTFF)之衛星搜尋方法。 【先前技術】 衛星定位技術被廣泛地應用於各個領域。在衛星通 訊系統中，例如全球導航衛星系統（Global Navigation Satellite System，GNSS )中，冷啓動狀態的含義係指沒 有可利用的資訊，如接收器位置 '觀察時間、衛星執道 資訊（如衛星粗略星曆資料（almanac)或衛星精確星曆 資料（ephemeris ))。因此，一可見衛星的識別碼（ID ) 將無法得知。除了衛星識別碼之外，肇因於衛星相對於 使用者的運動而定的都卜勒（Doppler )頻率亦為未知。 如果上述系統採用分碼多重接取（CDMA)訊號，以全 球定位系統（Global Positioning System, GPS)為例，亦 需要衛星所使用的偽隨機雜訊碼的碼相位以追蹤衛星。 如上所述，衛星訊號的特性可以由如下變量決定：衛星 識別碼、都卜勒頻率，以及偽隨機雜訊（pseud〇 Random Noise，PRN)碼相位。 定位一個三維位置，至少需要使用四顆衛星。所需 的首次定位時間取決於需要的四顆可見衛星多快能夠被 994 舉^説’假設從地面上能_察到十二顆GPS 可。那麽，傳統上，所有可 搜尋，以便朗可讀星。另外會依序的 勘另外由於每顆衛星的都卜 ^頻率和騎機㈣碼相位較未知的，因而，需要— 費大量，間去嘗試所有可能的值，以判定衛星的存在: 通*，-接收H藉㈣關聯性分析來搜尋可見衛 /關聯性分析騎慮_星識科（例如，全球定位 糸統（㈣、伽利略（Galile。）、廣域擴增系統（WA 同步衛星導航覆蓋服務（EGN〇S)、行動服務自動化系 統（MSAS)等系統的碼相位，以及都卜勒頻率。另外， 串行或並行搜尋均可_。壯，如果有四條可用通道 的話’可以同時搜尋四顆不同_星。爲了搜尋一顆衛 星’所有可能的碼相位和都卜勒頻率都應被掃描。 ▲这裡以-數字實例來説明。對於一接收器來説，假 设可見GPS衛星以列表方式表示為{5，9 , ΐ4，ΐ5，μ， 22，26，29，30} ’其可共用一個實體通道和十個通 逼，以分職尋並追縱衛星。掃描時間Ts係被定義為择 描所有可能的都卜勒頻率和碼相位所需的時間。如果一 候選衛星為不可見，則一關聯器會在經歷Ts搜尋時間之 後“去除’，（dismiss)此衛星。另一方面，關聯器會在 經歷一平均為l/2Ts的搜尋時間之後“命中，，（hit) 一 1335994 可見衛星。又假設資料解調變位元錯誤率為零。囚而， 在命中第-顆衛星之後需# 750秒的時間以接收衛星粗 略星曆資料（almanac)<3在命中一顆衛星後，還需要平 均27.6秒的時間去接收衛星精確星層資料 (ephemeds)。追縱Gps L! C/A碼訊號的接收器的掃描 時間Ts可以被計算出來，Gpsuc/A焉訊號的每個竭 週期具有H)23個碼片（ehip)。如果獲取程序中的石馬相 關性需要1/2 W之解析度，則碼相位的不確定範圍為 屬。通常，接收器中使用相干性與耗干性積分的結 合’以擴增獲取的靈敏度。因此，—組特定的都卜勒頻 率和碼相位的相關性週期ΔΤ表示）是相干性時間 (以仙表示）與非相干性計數（以奶表示）的乘積曰。 甚且’都卜勒頻率解析度dF通常被設定為咖，整個都 卜勒範圍以AF表示。基於上述假設，掃描時間Ts可以 根據下述公式（1)計算出來：1335994 IX. Description of the Invention: [Technical Field of the Invention] The present invention relates to satellite communication, and more particularly to a satellite search method capable of improving Time To First Fix (TTFF). [Prior Art] Satellite positioning technology is widely used in various fields. In satellite communication systems, such as the Global Navigation Satellite System (GNSS), the meaning of the cold-start state refers to information that is not available, such as the receiver location 'observation time, satellite obstetric information (such as satellite rough) Ephemeris data (almanac) or satellite precise ephemeris data (ephemeris). Therefore, the identification code (ID) of a visible satellite will not be known. In addition to the satellite identification code, the Doppler frequency due to the motion of the satellite relative to the user is also unknown. If the above system uses a code division multiple access (CDMA) signal, taking the Global Positioning System (GPS) as an example, the code phase of the pseudorandom noise code used by the satellite is also required to track the satellite. As mentioned above, the characteristics of the satellite signal can be determined by the following variables: satellite identification code, Doppler frequency, and pseudo-random noise (PRN) code phase. To locate a three-dimensional position, you need at least four satellites. The required first positioning time depends on how fast the four visible satellites needed can be detected by 994. Assume that 12 GPS can be detected from the ground. So, traditionally, all can be searched for readable stars. In addition, the sequence will be surveyed. Since the frequency of each satellite and the phase of the rider (four) code are relatively unknown, it is necessary to charge a large amount of time to try all possible values to determine the existence of the satellite: - Receiving H borrowing (4) Correlation analysis to search for visible guard/correlation analysis. _ Xingzhi (eg, Global Positioning System ((4), Galileo, Wide Area Augmentation System (WA Synchronous Satellite Navigation Coverage Service) (EGN〇S), code phase of systems such as the Service Automation System (MSAS), and Doppler frequency. In addition, serial or parallel search can be _. Strong, if there are four available channels, you can search for four at the same time. Different _ stars. In order to search for a satellite, all possible code phases and Doppler frequencies should be scanned. ▲This is illustrated by a digital example. For a receiver, it is assumed that visible GPS satellites are represented as a list. For {5,9,ΐ4,ΐ5,μ, 22,26,29,30} 'which can share a physical channel and ten tongs to separate and search for satellites. Scan time Ts is defined as Describe all possible The time required for the frequency and code phase. If a candidate satellite is invisible, a correlator will “dismiss” the satellite after experiencing the Ts seek time. On the other hand, the correlator will experience an average After the search time of l/2Ts, "hit, (hit) a 1335994 visible satellite. Also assume that the data demodulation bit error rate is zero. Prison, after hitting the first satellite, it takes #750 seconds to Receiving satellite rough ephemeris data (almanac) <3 after hitting a satellite, it takes an average of 27.6 seconds to receive satellite accurate star layer data (ephemeds). Tracking Gps L! C/A code signal receiver The scan time Ts can be calculated, and each exhaust cycle of the Gpsuc/A signal has H) 23 chips (ehip). If the stone horse correlation in the acquisition program requires 1/2 W resolution, then the code The uncertainty range of the phase is genus. Usually, the combination of coherence and depletion integral is used in the receiver to amplify the sensitivity of the acquisition. Therefore, the correlation period ΔΤ of the group-specific Doppler frequency and code phase is expressed. ) is the coherence time ( The prime is the product of the non-coherent count (indicated by milk). Even the 'Doppler frequency resolution dF is usually set to coffee, and the entire Doppler range is expressed in AF. Based on the above assumption, the scan time Ts can be Calculated according to the following formula (1):
Ts =2046χ-^χΔ7, dF = 2046x^x(7：.xrc) = 2046xAf χη xTc2 ( 1 ) 如果接收n使用1毫秒的Te以實行相干性積分，且 用非相干性積分，亦即非相干性數值可得出掃描一 千赫兹（kHz)的都卜勒範圍所需的掃描時間Ts為 1335994 . · - 20.46秒。需要找到四顆衛星才能定位一個位置點。假設 - 一 GPS衛星的依序搜尋順序為1，2，......，32，則在可 見衛星列表中的5，9，14和15號衛星會被依次命中。 下表1示出了依序搜尋的結果。 衛星識別碼 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 掃楛時間 (Ts) 1 2 3 4 4.5 5.5 6.5 7.5 8 9 10 11 12 12.5 13 14 被命中 0 0 0 0 1 0 0 0 2 0 0 0 0 3 4 0 衛星識別碼 17 18 19 20 21 22 23 24 25 26 27 28 29 30 掃描時間 (Ts) 15 15.5 16.5 17.5 18 18.5 19.5 20.5 21.5 22 23 24 24.5 25 被命中 0 5 0 0 6 7 0 0 0 8 0 0 9 10 表1依序搜尋衛星的命中結果Ts = 2046 χ - ^ χ Δ7, dF = 2046x^x (7:.xrc) = 2046xAf χ η xTc2 ( 1 ) If the reception n uses 1 millisecond of Te to perform coherence integration and is integrated with incoherence, ie, non-coherent The sex value gives the scan time Ts required to scan the Doppler range of one kilohertz (kHz) to be 1335994. · - 20.46 seconds. Four satellites need to be found to locate a location. Assume that - a GPS satellite's sequential search order is 1, 2, ..., 32, then satellites 5, 9, 14 and 15 in the visible satellite list will be hit in sequence. Table 1 below shows the results of the sequential search. Satellite identification code 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Broom time (Ts) 1 2 3 4 4.5 5.5 6.5 7.5 8 9 10 11 12 12.5 13 14 Hit 0 0 0 0 1 0 0 0 2 0 0 0 0 3 4 0 Satellite identification code 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Scan time (Ts) 15 15.5 16.5 17.5 18 18.5 19.5 20.5 21.5 22 23 24 24.5 25 Hit 0 5 0 0 6 7 0 0 0 8 0 0 9 10 Table 1 searches for the hit results of the satellites in order
各衛星的可見度可以根據同一固定位置在不同時間點的觀察 值的統計結果得出。第1圖係顯示針對一特定位置，在不同觀 察時間點觀察到的衛星可見度之觀察結杲示意圖。例如， 在圖中的觀察點“8”，可觀察到的可見衛星為SV5，9，14，15， 18，21，22，26，29 和 30。 從表1中可見，依序搜尋方案需要花費4.5Ts (=110.07秒） 的時間命中首顆衛星（命中衛星5)，並需要花費 13TS+27.6=345.58秒的時間來達成首次定位（衛星5, 9, 14和15被命中，收集衛星15之精確星曆資料需要27.6 秒）。爲了找到可見衛星列表中的所有衛星，需要的時間 為25Ts( 611.5秒）。這些命中時間無法令人滿意。因此， 1335994 需要提供一種減少尋找可見衛星所需時間的方案。 【發明内容】 本發明係提供一種衛星搜尋之方法，藉由本方法可 迅速找到定位所需之衛星。 依據本發明，選擇各衛星組成一衛星群，並於此衛 星群中進行搜尋。衛星群大小係為各階段預先決定，以 適用各種搜尋條件，衛星群之成員數（亦即衛星群大小） 及衛星群之成員係隨著持續獲得搜尋結果而更新。舉例 而言，更新依據該衛星群之掃描時間、衛星命中數、衛 星群之搜尋時間、衛星優先順序等條件。衛星群大小可 以減少或擴增。 藉由實施本發明，可以有效縮減搜尋可見衛星列表 中的所有衛星的命中時間，解決了習知技術中普遍存在 的耗時長之問題。 【實施方式】 以下將敘述本發明的詳細實施例。 第2A及第2B圖所示係美國專利申請案第 11/392,976號之衛星搜尋方法。該方法起始於一冷啓動 狀態（步驟S100)。根據步驟S110，初始化一候選衛星 的列表CandList，包括所有可能的衛星，如設定為{1， 1335994 2，......，32}。根據步驟S120，初始化一權重因素表， 記錄衛星列表之權重因素，其中將所有衛星的權重因素 均設定為0，即Wi=0，i=l，2，......，32。通常，一接 收器中包含多個實體通道，以搜尋並追蹤衛星。從步驟 S130中所述的第一通道開始，以輪替（round robin)方 式檢查及控制各實體通道。根據步驟S140，判斷當前通 道是否空閒（idle)，空閒是指當前通道並未進行衛星搜 尋和追蹤。若當前通道空閒，則根據步驟S150，判斷衛 星粗略星曆資料（almanac )以及位置/時間決定資訊是 否可用。若這些資訊可用，流程轉至步驟S160,進行熱 啓動，並且根據可用資訊選擇一顆衛星；否則，此方法 將進入步驟S200。根據步驟S200，從CandiList中選擇 並刪除權重因素最高的一顆衛星的識別碼（ID)。在初始 化狀態，第一衛星可以隨機或依照要求而選擇。輪替排 程（round-robin scheduling)係被用以搜尋衛星。換言之， 在一個輪替週期中，所有的衛星都只被搜尋一次。接著 根據步驟S210，檢查列表CandList。列表CandList係由 當前輪替週期中待搜尋之剩餘衛星構成。若列表 CandList為空，則當前輪替週期結束，下一輪替週期開 始，以搜尋未命中之衛星。因此，於步驟S220重新分配 候選衛星。於步驟S230，將選定的候選衛星分配至當前 11 1335994 空閒通道已進行搜尋。於步驟S240,選擇下一實體通道。 需注意，於步驟S140，若當前通道不是空閒，則 意味著其正在進行衛星搜尋。然後，分別於步驟S170 和S175,檢查衛星搜尋結果是命中還是去除。如果此通 道沒完成搜尋，則進入步驟S240，選擇下一通道。在一 此通道完成對某一衛星的搜尋之後，搜尋結果無論是命 中還是去除衛星的搜尋結果都會被發佈。如果某一衛星 被去除，則根據步驟S180中的條件可見度所決定的數值 來降低其他衛星的權重因素。在其他實施例中，包含被 搜尋的衛星在内的所有衛星的權重因素都要被更新。然 後，當前通道用以搜尋下一顆衛星，流程進入步驟 S200，以選擇其他的衛星進行搜尋。應該注意到，在步 驟S180中，權重因素已被更新，因而可以從候選衛星列 表中選擇出具有最高權重因素或等同為最可能被命中的 一顆衛星。 於步驟S190中，如果發現當前通道命中一顆衛星， 則程序會根據其他衛星相對於命中的衛星的條件可見度 來增加其權重因素。在一實施例中，被命中的衛星本身 的權重因素不做改變。在其他實施例中，被命中的衛星 本身的權重因素也要更新。當前通道用於追蹤被命中的 衛星，流程直接前進至步驟S240以處理下一通道。 v ^ 12 員，在上述說明中，搜尋具有最高優先順序的衛星成 貝及與此同時將此衛星成員從CandList刪除。衛星優 歹I頃序之敘述如下。對衛星的搜尋開始於從一候選衛星 幻表中搜尋一預定的或隨機選取的第一衛星，候選衛星 列表中记錄了所有或預定範圍内的可用衛星。另需事先 準備初始權重因素表，其中，初始權重因素在一顆衛 星被搜哥到之後會被更新。這些初始權重因素係用於決 疋哪顆衛星具有最高的可見可能性。舉例來説，如杲候 選：星列表中包括SVi，SV2，……，SV32，並且當前被 搜哥的衛星為SVi，那麼，更新後的衛星SV2至sv32的 權重因素被用於選擇下一顆待搜尋的衛星。權重因素可 以根據兩顆衛星之間的任一相互關係來進行更新。例 如，權重因素係根據條件可見度進行更新的。所述條件 可見度係指如果一特定衛星被看見，那麼一顆不特定衛 星被看見的條件概率。下表二示出了衛星SV1之條件可 見度的示例。其餘衛星的情況可依此推斷。The visibility of each satellite can be derived from the statistical results of observations at different points in time at the same fixed location. Figure 1 is a graphical representation of the observations of satellite visibility observed at different observation time points for a particular location. For example, at the observation point "8" in the figure, the visible satellites that can be observed are SV5, 9, 14, 15, 18, 21, 22, 26, 29, and 30. As can be seen from Table 1, the sequential search scheme takes 4.5Ts (=110.07 seconds) to hit the first satellite (hit satellite 5), and it takes 13TS+27.6=345.58 seconds to reach the first positioning (satellite 5, 9, 14 and 15 were hit, collecting accurate ephemeris data for satellite 15 takes 27.6 seconds). In order to find all the satellites in the visible satellite list, the time required is 25Ts (611.5 seconds). These hit times are unsatisfactory. Therefore, 1335994 needs to provide a solution to reduce the time required to find visible satellites. SUMMARY OF THE INVENTION The present invention provides a satellite search method by which the satellites required for positioning can be quickly found. In accordance with the present invention, each satellite is selected to form a constellation of satellites and searched for this satellite group. The satellite group size is pre-determined for each stage to suit various search conditions. The number of members of the satellite group (ie, the size of the satellite group) and the members of the satellite group are updated as the search results are continuously obtained. For example, the update is based on conditions such as the scan time of the satellite group, the number of satellite hits, the search time of the satellite group, and the satellite priority. The satellite group size can be reduced or amplified. By implementing the present invention, it is possible to effectively reduce the hit time of all satellites in the searchable visible satellite list, and solve the problem of long time consuming in the prior art. [Embodiment] Hereinafter, a detailed embodiment of the present invention will be described. Figures 2A and 2B show the satellite search method of U.S. Patent Application Serial No. 11/392,976. The method starts in a cold start state (step S100). According to step S110, a list CandList of candidate satellites is initialized, including all possible satellites, as set to {1, 1335994 2, ..., 32}. According to step S120, a weighting factor table is initialized, and the weighting factor of the satellite list is recorded, wherein the weighting factors of all the satellites are all set to 0, that is, Wi=0, i=l, 2, ..., 32. Typically, a receiver contains multiple physical channels to search for and track satellites. Starting from the first channel described in step S130, each physical channel is inspected and controlled in a round robin manner. According to step S140, it is judged whether the current channel is idle or not. Idle means that the current channel does not perform satellite search and tracking. If the current channel is idle, then according to step S150, it is determined whether the satellite rough ephemeris data (almanac) and the position/time determination information are available. If such information is available, the flow proceeds to step S160, performs a warm start, and selects a satellite based on the available information; otherwise, the method proceeds to step S200. According to step S200, the identification code (ID) of one satellite having the highest weight factor is selected and deleted from the CandiList. In the initial state, the first satellite can be selected randomly or as desired. Round-robin scheduling is used to search for satellites. In other words, in a rotation cycle, all satellites are only searched once. Next, according to step S210, the list CandList is checked. The list CandleList consists of the remaining satellites to be searched in the current rotation period. If the list CandList is empty, the current rotation period ends and the next rotation period begins to search for missed satellites. Therefore, the candidate satellites are reallocated in step S220. In step S230, the selected candidate satellites are assigned to the current 11 1335994 idle channel has been searched. In step S240, the next physical channel is selected. It should be noted that in step S140, if the current channel is not idle, it means that it is performing satellite search. Then, in steps S170 and S175, respectively, it is checked whether the satellite search result is a hit or a removal. If the search has not completed the search, the process proceeds to step S240 to select the next channel. After a search for a satellite is completed on this channel, the search result will be released whether it is a hit or a satellite removal result. If a satellite is removed, the weighting factors of other satellites are reduced according to the value determined by the conditional visibility in step S180. In other embodiments, the weighting factors for all satellites including the satellite being searched are updated. Then, the current channel is used to search for the next satellite, and the flow proceeds to step S200 to select other satellites for searching. It should be noted that in step S180, the weighting factor has been updated so that the satellite having the highest weighting factor or equivalent to being most likely to be hit can be selected from the list of candidate satellites. In step S190, if it is found that the current channel hits one satellite, the program increases its weighting factor according to the conditional visibility of other satellites relative to the hit satellite. In one embodiment, the weighting factor of the satellite itself being hit does not change. In other embodiments, the weighting factors of the satellite being hit are also updated. The current channel is used to track the satellite being hit, and the flow proceeds directly to step S240 to process the next channel. v ^ 12 members, in the above description, search for the satellites with the highest priority and at the same time delete the satellite members from the CandList. The description of the satellite excellent I sequence is as follows. The search for satellites begins by searching for a predetermined or randomly selected first satellite from a candidate satellite illusion in which all or a predetermined range of available satellites are recorded. In addition, an initial weighting factor table needs to be prepared in advance, wherein the initial weighting factor is updated after a satellite is searched. These initial weighting factors are used to determine which satellite has the highest visible probability. For example, if the candidate: the star list includes SVi, SV2, ..., SV32, and the current satellite of the search is SVi, then the weighting factors of the updated satellites SV2 to sv32 are used to select the next one. Satellite to be searched. The weighting factor can be updated based on any relationship between the two satellites. For example, weighting factors are updated based on conditional visibility. The conditional visibility is the conditional probability that an unspecified satellite is seen if a particular satellite is seen. An example of the conditional visibility of the satellite SV1 is shown in Table 2 below. The situation of the remaining satellites can be inferred from this.
Si 1 14 25 20 11 16 22 3 30 19 23 15 6 5 24 18 Vi„ 1 0.59 0.56 0.17 0.06 0.05 0.04 0.04 0 -0.04 •0,05 -0.16 -0.17 -0.19 •0.25 -0.33 sv 9 21 13 4 7 2 28 27 8 10 26 29 12 17 31 32 Vi,, -0.34 -0.36 •0.47 -0.51 -0.53 -0.64 •0.65 •0.68 -0.90 •0.92 •0.94 -1 -1 -1 -1 -1 — _ 表二衛星SV!之條件可見度 13 此權重因素表中的初始權重因素Wi(i=l，2,.....， 32)都置為“G”°當某一衛星SVi被搜尋到且判斷為可見 或不可見之後，權重因素Wj ( j = l，2，....... 1 1,1+1,......，32)將分別根據條件可見度VjM更新。Si 1 14 25 20 11 16 22 3 30 19 23 15 6 5 24 18 Vi„ 1 0.59 0.56 0.17 0.06 0.05 0.04 0.04 0 -0.04 •0,05 -0.16 -0.17 -0.19 •0.25 -0.33 sv 9 21 13 4 7 2 28 27 8 10 26 29 12 17 31 32 Vi,, -0.34 -0.36 •0.47 -0.51 -0.53 -0.64 •0.65 •0.68 -0.90 •0.92 •0.94 -1 -1 -1 -1 -1 — _ Table 2 Conditional visibility of satellite SV! 13 The initial weighting factor Wi (i=l, 2, . . . , 32) in this weighting factor table is set to “G”. When a satellite SVi is searched and judged as After being visible or invisible, the weighting factors Wj (j = l, 2, . . . 1, 1, 1, +1, ..., 32) will be updated according to the conditional visibility VjM, respectively.
最先搜尋到的第一衛星為SVi。顯然，可以將任何 顆衛星設定為第一衛星。第一衛星也可以隨機選取。 j第一衛星被搜尋之後，初始權重因素表即根據第一衛 f (本實施例中為SVi)賴尋結果進行更新。如果衛 星SVl被命中，則其他衛星SV2至SV32的權重因素都要 根據其相對於第—衛星SV1的條件可見度進行擴增，即The first satellite first searched was SVi. Obviously, any satellite can be set as the first satellite. The first satellite can also be randomly selected. After the first satellite is searched, the initial weighting factor table is updated based on the results of the first guard f (SVi in this embodiment). If the satellite SVl is hit, the weighting factors of other satellites SV2 to SV32 are amplified according to their conditional visibility relative to the first satellite SV1, ie
Wj—Wj+Vj|i，其中卜2，3，...... ’ 32。反之’如果第一Wj—Wj+Vj|i, where Bu 2, 3, ... ’ 32. Otherwise ‘if the first
衛星SVl被去除，則其他衛星的權重因素要被減少，即 Wj’-VjH，其中卜2，3，’ 32。然後，待搜尋之第 -衛星根據更新後的權4因素表選擇。例如，選擇具有 取向權重因素的衛星以搜尋。接下來搜尋的衛星依據上 述相似的方式進行選擇。 在本發明中，衛星識別碼列表（例如CandList)之 待搜尋成員係更有效地動態控制及減少，所以接收器之 相關器被使用以搜尋具有最高的可見可能性衛星群之衛 星。在微弱訊號狀n下，衛星搜尋持續相較長以擴增 獲取靈敏度。在下面的情況中，如果所有接收器之相關 1335994 器被用以搜尋具有最高的可見可能性的衛星群之衛星， 則可縮短首次定位時間，此乃因為接收器沒有用很多時 間在搜尋不可見的衛星。 優選地，候選列表之衛星群之成員動態變化且衛星 群大小收斂至一確實可見衛星之列表以縮減接收器之搜 尋範圍。然而，舉例而言，當目前衛星群之多數成員已 經被獲取及被追蹤或當目前衛星群之成員已經被搜尋很 多次，仍未獲得衛星粗略星曆資料（almanac )及衛星精 確星曆資料（ephemeris )時，衛星群也可擴展至搜尋更 多未獲取的衛星。 第3圖係根據本發明一實施例之衛星搜尋方法之流 程圖。該方法起始於步驟S300。於步驟S310，初始化一 候選衛星列表CandList，如設定為{1，2 ’ ...... ’ 32}。 舉例而言，步驟S330中，分配第一通道，以輪替方式檢 查及控制各實體通道。於步驟S340，判斷當前通道是否 空閒，空閒是指當前通道並未進行衛星搜尋和追蹤。若 當前通道空閒，則於步驟S350，判斷該衛星粗略星曆資 料（almanac)以及位置/時間決定資訊是否可用。若命 中衛星資訊可用，該流程轉至進一步的流程，此進一步 流程未揭露於本發明，在此省略不再說明。否則，流程 進入步驟S400,將在後面說明步驟S400。若當前通道係 15 1335994 被使用，即當前通道係用於搜尋，於步驟S370及S375， 判斷衛星SVi是否命中或去除。不管搜尋的結果如何， 於步驟S390更新相應的候選列表。舉例來說，該衛星群 最初包括32個衛星，被更新後只有8個衛星，將在後面 進一步說明。於步驟S400，在被稱為取得選定的候選衛 星的進一步流程中，一候選衛星從當前列表選出。所有 可能衛星（亦即候選衛星）構成候選列表CandList。在 本實施例中，各衛星依據優先順序排列。優先順序根據 前面所述的權重因素決定。最高優先順序之衛星從當前 搜尋列表中被選擇及删除，而不是從衛星群被選擇及刪 除。即當前衛星群被掃描係以輪替方式掃描一次。若被 選擇衛星已經追蹤或正在搜尋，則重複取得選定的候選 衛星的進一步流程。衛星群之被掃描次數係藉由計數值” 輪替週期”代表，其用以決定更新衛星群大小，下文再詳 細說明。於步驟S420，將從該衛星群選擇之衛星分配至 當前欲搜尋之通道。於步驟S440,選擇下一個實體通道。 第4圖係根據本發明一實施例之更新候選衛星列表 之衛星群大小之示意圖。如圖所示，列表之初始衛星群 大小（例如：衛星群之成員數）為32 (亦即G32)，即 所有可用的32個衛星識別碼都包括在列表中。舉例來 說，這個例子中在第一次搜尋後（最好是第一次就命中） 16 1335994When the satellite SV1 is removed, the weighting factors of other satellites are reduced, namely Wj'-VjH, where Bu 2, 3, '32. Then, the first satellite to be searched is selected according to the updated weight 4 factor table. For example, a satellite with an orientation weight factor is selected for searching. The satellites searched next are selected in a similar manner as described above. In the present invention, the satellite identity code list (e.g., CandList) to be searched for members is more dynamically dynamically controlled and reduced, so the correlator of the receiver is used to search for satellites having the highest visible probability satellite group. In the weak signal form n, the satellite search continues for a longer period to amplify the acquisition sensitivity. In the following case, if all receivers related to 1353994 are used to search for satellites with the highest visible probability of satellites, the first positioning time can be shortened because the receiver does not spend much time invisible in the search. Satellite. Preferably, the members of the candidate list of satellites dynamically change and the constellation size converges to a list of truly visible satellites to reduce the search range of the receiver. However, for example, when a majority of the current satellite constellation has been acquired and tracked or when members of the current constellation have been searched many times, the satellite rough ephemeris data (almanac) and satellite precise ephemeris data have not yet been obtained ( At ephemeris, the constellation of satellites can also be extended to search for more unacquired satellites. Figure 3 is a flow diagram of a satellite search method in accordance with an embodiment of the present invention. The method begins in step S300. In step S310, a candidate satellite list CandList is initialized, as set to {1, 2 s ... '32}. For example, in step S330, the first channel is allocated, and each physical channel is inspected and controlled in a rotating manner. In step S340, it is determined whether the current channel is idle, and idle means that the current channel does not perform satellite search and tracking. If the current channel is idle, then in step S350, it is determined whether the satellite's rough ephemeris data (almanac) and the location/time determination information are available. If the hit satellite information is available, the process goes to a further process, and this further process is not disclosed in the present invention and will not be described here. Otherwise, the flow advances to step S400, which will be described later. If the current channel system 15 1335994 is used, that is, the current channel is used for searching, in steps S370 and S375, it is determined whether the satellite SVi is hit or removed. Regardless of the result of the search, the corresponding candidate list is updated in step S390. For example, the constellation originally consisted of 32 satellites and was updated with only 8 satellites, as explained further below. In step S400, in a further process referred to as obtaining the selected candidate satellite, a candidate satellite is selected from the current list. All possible satellites (i.e., candidate satellites) constitute a candidate list CandList. In this embodiment, the satellites are arranged in order of priority. The order of precedence is determined by the weighting factors described above. The highest priority satellites are selected and deleted from the current search list, rather than being selected and deleted from the satellite constellation. That is, the current satellite group is scanned once in a rotating manner. If the selected satellite has been tracked or is being searched, the further process of obtaining the selected candidate satellite is repeated. The number of times the satellite group is scanned is represented by the count value "rotation period", which is used to determine the size of the updated satellite group, as described in more detail below. In step S420, the satellite selected from the satellite group is assigned to the channel currently to be searched. In step S440, the next physical channel is selected. Figure 4 is a diagram showing the size of a satellite group of a candidate satellite list updated in accordance with an embodiment of the present invention. As shown, the initial satellite group size of the list (for example, the number of members of the satellite group) is 32 (i.e., G32), that is, all available 32 satellite identification codes are included in the list. For example, in this example after the first search (preferably the first hit) 16 1335994
衛星群大小改變為8個。在衛星群大小等於8的階段（亦 即G8)，若命中數到達6 (亦即這裡已經找到6個衛星） 或當前衛星群已轉描超過1G次（”輪替週期等於 10)，衛星群大小擴張S 16。在衛星群大小等於16的階 段（亦即CH6)，若命中數到達1〇 (亦即這裡已經找到 10個衛星）或當前衛星群已經掃描超過10 :欠（亦即輪 替週期等於1G)，魅群大傾張至32。儘管此範例令 僅說明G32、G8及G16，衛星群大顿著各階段如說明 中設定。此外，衛星群大小的狀態可以從當前之大小到 更新之大小以減少或擴增而改變或進一步如說明。 甘在第4圖之實施例，衛星群大小可以動態擴張。但 是在其他實施例，衛星群大小依據情況及設計方式可以 減少或動態擴張且考慮不同的參數。舉例而言，若衛星 數已經追蹤到達預設值則衛星群大小係擴張。若衛星群 掃描次數（亦即輪替週期）到達預設數則衛星群大小係 擴張。若追蹤衛星數低於臨界值則衛星群大小係減少。 任何其它可能的情況也可以作為設計的考量因素。舉例 而&，右搜尋時間超過臨界值則衛星群大小係改變。 在一實施例中，當衛星群從初始的32個成員更新 到8個成員時，32個衛星中具有最高優先順序的8個衛 星被選擇為此8㈣貞。此優先順序可藉由前面所敘述 17 1335994 的權重因數來判斷。如同提及，在搜尋第一衛星之後係 執行從32個成員改變到8個成員。即在搜尋第一衛星之 後不^'命中或去除，所有衛星之優先順序根據搜尋的結 杲更新，並且依據衛星優先順序來選擇此8個成員。備 選或優選，最好是第一次命中之後執行從初始的32個成 員改變到8個成員。即僅在發現第一衛星時執行改變， 這可以避免訊號阻礙或是檢測錯誤的影響。 在各階段（亦即G8、G16或G32)，衛星群之成員 可以不改變直到衛星群大小改變。舉例來說，在G8的 階段’此8個成員不會改變。當階段從G8被更新到G16， 在所有可見衛星中選出16個具有最高優先順序的衛星 來作為G16的16個成員。在其他實施例的各階段，這 些成員不斷改變。舉例來説，在G8的階段，衛星群之8 個成員係不固定但依據搜尋結果更新。此8個成員依據 前面描述之衛星優先順序選擇。每當一衛星被搜尋時， 衛星優先順序就會更新。備選地，只有當命中衛星時衛 星優先順序才會更新。此外，如前所述，除了使用條件 可見度判斷優先順序之外，從衛星接收之訊號的估計載 波雜訊比（Carrier To Noise Ratio，CNR)也可以連同條 件可見度一起使用來共同判斷衛星優先順序。這可以避 免檢測不正確警報的影響。 1335994 在此情況下特定的衛星群之成員被改變，衛星群之 成員隨著衛星優先順序的更>而更新。另外，衛星群之 成員在衛星優先順序受到多次更新而不是只有一次更新 後也被更新。因此多個衛星的不正確檢測或阻礙之機率 少於只有一個衛星，藉由前面的作法衛星優先順序可以 更正確且可靠。 依據本發明，衛星群大小之改變及衛星成員之更新 係的設定十分彈性。除了前面所提的情況以外，任何可 能的情況都可以考慮。 儘管上述實施例僅以全球定位系統（GPS )及所述 衛星為範例，其他全球導航衛星系統（Global Navigation Satellite System，GNSS)像是格洛納斯（GLONASS)、 伽利略（GALILEO)及相似也是可應用的。此外，本發 明也可以應用於衛星基頻增強系統（Satellite Based Augmentation System，SBAS )，其包括廣域擴增系統 (WAAS)、同步衛星導航覆蓋服務（EGNOS)、行動服 務自動化系統（MSAS)等系統的衛星。依據本發明之方 法可以使用在任何類型的通訊系統。 雖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，任何熟悉此項技藝者，在不脫離本發明 之精神和範圍内，當可做些許更動與潤飾，因此本發明 1335994 之保護範圍當視後附之申請專利範圍所界定者為準。 【圖式簡單說明】 ‘第1圖係顯示針對-特定位置，在不同觀察時間點 觀察到的衛星可見度之觀察結果示意圖。 第2A及第2B圖係共同組成一衛星搜尋方法之流程 圖。 第3圖係根據本發明一實施例之衛星搜尋方法之流 程圖。 第4圖係根據本發明一實施例之更新候選衛星列表 之衛星群大小之示意圖。 【主要元件符號說明】 益 20The satellite group size has changed to eight. At the stage where the satellite group size is equal to 8 (ie G8), if the hit number reaches 6 (ie 6 satellites have been found here) or the current satellite group has been transcoded more than 1G times ("the rotation period is equal to 10", the satellite group Size expansion S 16. At the stage where the satellite group size is equal to 16 (ie CH6), if the hit number reaches 1〇 (that is, 10 satellites have been found here) or the current satellite group has scanned more than 10: owe (ie, rotation) The period is equal to 1G), and the charm group is tilted to 32. Although this example only describes G32, G8, and G16, the satellite group is set in various stages as specified in the description. In addition, the state of the satellite group size can be from the current size to The size of the update is changed by reduction or amplification or further as illustrated. In the embodiment of Figure 4, the satellite group size can be dynamically expanded. However, in other embodiments, the satellite group size can be reduced or dynamically expanded depending on the situation and design. And consider different parameters. For example, if the number of satellites has been tracked to reach the preset value, the satellite group size is expanded. If the number of satellite group scans (ie, the rotation period) reaches the preset number, the satellite group size Expansion. If the number of tracking satellites is below the critical value, the satellite group size will decrease. Any other possible situation can also be considered as a design consideration. For example, if the right search time exceeds the critical value, the satellite group size will change. In the embodiment, when the satellite group is updated from the initial 32 members to 8 members, the 8 satellites with the highest priority among the 32 satellites are selected as 8 (four) 贞. This priority can be as described above by 17 1335994. The weighting factor is used to judge. As mentioned, after searching for the first satellite, the change from 32 members to 8 members is performed. That is, after searching for the first satellite, it is not hit or removed, and all satellites are prioritized according to the search. The update is updated, and the 8 members are selected according to the satellite priority order. Alternatively or preferably, it is preferable to perform the change from the initial 32 members to 8 members after the first hit. That is, only when the first satellite is found. Perform changes, this can avoid signal blocking or detect the impact of errors. At each stage (ie G8, G16 or G32), members of the satellite group can not change until Wei The size of the constellation changes. For example, in the G8 phase, the 8 members will not change. When the phase is updated from G8 to G16, 16 of the highest priority satellites are selected among all visible satellites as the 16 of the G16. Members. These members are constantly changing at various stages of other embodiments. For example, in the G8 phase, the eight members of the satellite group are not fixed but are updated based on the search results. The eight members are based on the satellites described above. Sequential selection. Satellite priority is updated whenever a satellite is searched. Alternatively, satellite priority is updated only when the satellite is hit. In addition, as previously mentioned, in addition to using conditional visibility to prioritize The estimated carrier-to-noise ratio (CNR) of the signal received from the satellite can also be used together with the conditional visibility to determine the satellite priority. This avoids the effects of detecting an incorrect alarm. 1335994 In this case the members of a particular constellation are changed and the members of the constellation are updated with the satellite priority order >. In addition, members of the constellation are also updated after the satellite priority has been updated multiple times instead of only one update. Therefore, the probability of incorrect detection or obstruction of multiple satellites is less than that of only one satellite, and the satellite prioritization can be more accurate and reliable by the previous method. According to the present invention, the change in the size of the satellite group and the setting of the update system of the satellite members are very flexible. In addition to the situations mentioned above, any possible situation can be considered. Although the above embodiments are only exemplified by the Global Positioning System (GPS) and the satellites, other Global Navigation Satellite Systems (GNSS) such as GLONASS, GALILEO, and the like are also available. Applied. In addition, the present invention can also be applied to a Satellite Based Augmentation System (SBAS), which includes a Wide Area Augmentation System (WAAS), a Synchronized Satellite Navigation Coverage Service (EGNOS), a Mobile Service Automation System (MSAS), and the like. System satellite. The method according to the invention can be used in any type of communication system. While the present invention has been described in its preferred embodiments, the present invention is not intended to limit the invention, and the present invention may be modified and modified without departing from the spirit and scope of the invention. The scope of protection of 1335994 is subject to the definition of the scope of the patent application. [Simple description of the diagram] ‘The first figure shows the observation results of satellite visibility observed at different observation time points for specific locations. Figures 2A and 2B are diagrams of a satellite search method. Figure 3 is a flow diagram of a satellite search method in accordance with an embodiment of the present invention. Figure 4 is a diagram showing the size of a satellite group of a candidate satellite list updated in accordance with an embodiment of the present invention. [Main component symbol description] Benefit 20