201218695 六、發明說明： 【發明所屬之技術領域】 本發明係關於一種用於在無線網路中控制資料封包傳輸 之裝置、系統及方法。 【先前技術】 近來，無線網狀網路吸引愈來愈多的關注，（例如）用於 照明系統、建築物自動化、監視應用程式、感測器系統及 醫療應用之遠端控制。詳言之，室外燈具之遠端管理（所 謂之電傳管理（telemanagement))變得愈加重要。一方面， 此係由環境之關注而驅動，此係因為遠端控制系統或所謂 之電傳管理系統使得能（例如）隨時間、天氣條件及季節而 使用不同調光樣式，從而允許室外照明系統之更能源有效 之使用。另一方面，此亦藉由經濟原因驅動，此係因為增 加之能源效率亦降低操作成本。此外，系統可遠端地監^ 電力使用且侧燈故障，此情形允許判定用於修理燈具或 替換燈之最佳時間。 八〆 虽刖基於射頻（RF)之無線解決方案而使用星形網路拓樸 或網狀網路拓樸。在星形網路中，資料控制器在網路中且 有至每—節點之直接通信路徑。“’此通常需要在高層 (例，’在建築物之頂部）置放高功率/高敏感度基地台式控 制态’其使解決方案部署麻煩且昂貴。在網狀網路中 數個節點一般不直接與 且牧，、徑制态通信，而是經由所謂 躍通信與控制器通信。在多跡耀…一月之夕跳 間節點將資料封包自發详及夕個中 發送益郎點傳輸至目的地節點。節點 159105.doc 201218695 充當路由器以將資料封包自相鄰節點傳輸至距離太遠以致 以單跳躍不能到達之節點，從而產生可橫跨較大距離之網 路藉由以-連_較短跳躍打斷長距離，信號強度得以維 持i因此，藉由網狀網路中之所有節點來執行路由，從而 决疋待將資料封包發送至哪一相鄰節點。因此，網狀網路 為具有问連接性且因此具有高冗餘及可靠性之非常穩健且 穩定的網路。 在先則技術中，可將網狀網路傳輸技術劃分成兩個群 基於/乏播之網狀網路及基於路由之網狀網路。在基於 泛播之網狀網路中，藉由網路中之所有節點來轉遞所有資 料封包。因此，一節點不必進行複雜之路由決策，而是僅 廣播貝料封包藉由此等手段，技術係相當穩健的。然 而在大型網路中’歸因於轉遞之資料附加項影響總資料 速率》此外’資料封包之衝突更可能發生，從而進一步降 低總效能。因此，此解決方案之主要問題為可擴充性。可 將基於路由之網狀網路進一步劃分成主動式方案及反應式 方案。在基於路由之主動式網狀網路中，將所有需要之網 路^徑儲存於每一節點中之路由表中。路由表（例如）藉由 將定期信標訊息發送至相鄰節點以發現有效路由路徑而保 持最新。儘管資料傳輸在此種網路中係非常有效的，但可 擴充性仍為低的，此係因為在大型網路中，路由表之主動 式更新消耗大部分網路資源。此外，路由表將隨網路· 充而增長。另外，網路之設置需要時間及資源，以便建置 路由表。對比而言，反應式方案藉由按需求發現路線而避 159105.doc. 201218695 免永久附加項及大的路由表。反應式方案使用泛播來發現 網路路徑且快取作用中路線或節點。當多個路線僅用於單 一資料封包時，泛播資料封包而非執行路由發現可係更有 效的。若路線保持足夠長以避免頻繁路由，則反應式方案 退化至主動式方案。用於基於路由之反應式網狀網路的實 例用於ZigBee中。然而，此協定方案之主要問題仍係網路 之可擴充性。 在大規模多跳躍網路中，與小型網路中之跳躍距離相 比，資料封包必須行進的跳躍之數目較大。在包含數千個 節點之大龍頻電傳管㈣統中，有可能發生20至40個跳 躍。然而’個別資料封包之遞送機會隨其跳躍距離而減 小’此係因為對於每-卿，存在資料封包丢失之機會。 因此，非常有限之網路可擴充性構成常見無線網狀：路 中之大的缺陷。此係歸因於每一資料封包或訊息歸因於轉 遞而被多次傳輸的事實，藉此降低總的網路輸送量。又， 更可能發生資料封包衝突’引起資料封包丢失，從而進一 步降低總效能。因此’改良多跳躍點對點傳輸之成功及可 靠性在大規模多跳躍網路(諸如’具有大量燈具節點之街 ::明系統)中係尤其至關重要的’此係因為點對點重新 傳輸比在典型之較小型網路令更多地耗費 延遲密集的。因此’大規模無線網狀網路需要有效之路^ 協疋及點對點延遲之降低，以便達成所需之 時間及穩健性。 應 為了判定資料封包是被成功地遞送抑或*失，通常以應 159105.doc 201218695 答模式執灯資料封包傳輸。在逐跳躍應答模式下，由先前 傳輸節點之接收節點確認多跳躍傳輸之每一跳躍。然而， 此情形導致高的網路負載。因此，常使用點對點應答，其 中最終目的地節點向初始發送器節點確認資料封包之接 在此模式下，發送器節點在重新傳輸該資料封包（預 期針對該資料封包之應答）之前等待一預定時間（所謂的應 答，逾時大體而言’應答逾時係固定的且料網路之所 有節點為共同的。因為用於行進短距離之資料封包之應答 逾時接著與用於行進長距離之資料封包的應答逾時相同: 所以不必要地增加了行進短距離之資料封包之重新傳輸延 遲，從而影響網路之總的傳輸速度。此情形引起網路中之 任何兩個節點之間的通信之大的延遲以及延遲差，從而 (例如）在與照明系統之燈具節點互動時歸因於高及/或不均 勻延遲而導致不良的使用者體驗。若此缺陷藉由簡單地減 小網路之大小而解決，則將進一步降低可擴充性。因此， 應最小化成功資料封包遞送之點對點延遲。 “US 2009/0154395 A1描述-種用於叢集網狀網路之用於 稭由與相鄰節點共用效能量測結果來最小化傳輸延遲之路 由方法。藉由此等手段，選擇下一節點，該資料封包待在 多跳躍傳輸中轉遞至該下一節點 【發明内容】 馨於先前技術中之上述缺陷及問題，本發明之目標係提 七、種用於在一無線網路中傳輸資料封包之裝置、系統及 方法，其最小化點對點重新傳輸延遲且使點對點重新傳輸 159l05.doc 201218695 、句勻化同時維持或甚至增加網路可擴充性。 該目標係藉由獨立請求項之特徵來解決。 |㈣基於如下思想··基於資料封包必須涵蓋的在發 、器：點與目的地節點之間的距離而最小化用於一對給定 發j器節點及目的地節點之答覆逾時。該答覆逾時與發送 W點等待對經傳輸至目的地節點之資料封包之答覆的等 待時間有關。若該發送器節點在此時間週期屆期之後仍未 接收到答覆資料封包，則該發送器節點將重新傳輸該資料 封包。此情形消除了未應答資料封包之重新傳輸之間的不 必要的通信延遲。 在本：明之-態樣中’當一無線網路之一節點操作為一 發送器節點時，提供用於該節點之用於控制資料封包傳輸 的裝置。該裝置包含一控制單元，該控制單元可基於該發 送器節點與該目的地節點之間的距離而調整該答覆逾時。 該答覆逾時定義—時間間隔，在該時間間隔期間.，發送器 節點等待對以所發送之資料封包之答覆。#該答覆逾時 已屆期而該發送n節點尚未接收到來自該目的地節點之答 "’該發送器節點將開始重新傳輸該資料封包。藉㈣ 對每-對發送器節點及目的地節點而個別地調整該答覆逾 時’降低了偵測失敗傳輸之延遲，因此將歸因於重新傳輸 之成功傳輸之點對點延遲減小為接近於最小可能值。因 此’在大規模照明线中’此情形將進—步減小控制命令 之延遲，使得燈具節點將對（例如）調光或切換命令更快地 作出反應。 ' I59105.doc 201218695 在二較佳實施例中，該答覆逾時與應答逾時（亦即，發 送器卽點在重新傳輸資料封包 一 則專待該目的地節點之指 不成功資料封包傳輸的應答之時 间隔）有關。可能地， 來自該目的地節點之應答含有緊鄰應答部分之額外資料。 藉由此等手段，使得網路連接更有效率，此係因為不需要 發送含有額外資料之其他封包。因此，更大體而言，該應 答逾時亦可指代發送器節點等待來 的時間間隔。 手…目的地郎點之資料 較佳地，使用多跳躍傳輸以點對點應答模式來經由中間 節點將資料封包自發送器節點傳輸至目的地節點。此意謂 =該資料封包已到達其目的地節點之後對成功傳輸作出應 在-實施例中’可將該裝置添加或輕接至無線網路之現 有郎點::控制中心。因此’該裝置與—網路節點相關聯， 該網路郎點亦可為一資料收焦装銘科 ^ 针收集器即點。該資料收集器節點 可為經組態以與網路之控制中心通信的任何節點且可充當 一種閘道h舉例而言’該裝置可經調適以插入現有電路 板中或連接至節點之現有控制器。此情形尤其有用於改良 或升級現有系統(諸如’衔道照明系統)。除控制單元外， 裝置亦可進-步包含一記憶體及/或用於接收及傳輸資料 封包之收發單元。 無線網路可具有網狀扭樸，其中每一節點可充當路由 器。此網路具有增加之冗餘及可靠性。較佳地，無線網路 之郎點為固定的，此係因為該無線網路主要係針對大型室 159105.doc 201218695 外照明系統之狀況。或者或另外，至少—些節點之位置可 為網路之其他節點中之至少一些者及/或網路之控制中心 已知。舉例而t，節點中之至少—些者可儲存用於自各別 節點至最近資料收集器節點之資料封包傳輸的路由表。較 佳地，用於至最近資料收集器節點之資料封包傳輸之路由 協定係基於多對-路由料封包傳輸至較接近 該等資料收集器節點中之一者的相鄰節點。藉由此等手 段，資料封包傳輸變得較快且更有效率。此外，此情形亦 允許在大型無線網狀網路中（例如，在具有數目超過ι〇〇〇 個之燈具節點之街道照明系統中）省去多個資料收集器節 點’藉此增加冗餘且改良網路之可靠性。 在另一實施例中，藉由跳躍距離、基於GPS之距離及/或 歐幾里德（Euclidean)距離來定義兩個節點（例如，發送器節 點與目的地節點或中間節點）之間的距離。兩個節點之間 的跳躍距離之特徵可在於：跳躍計數（亦即，在兩個節點 之間傳輸一資料封包所需的跳躍之數目）或將資料封包轉 遞至最終目的地節點之中間節點的數目。基於Gps之距離 可自發送器節點及目的地節點之GPS位置導出，而歐幾里 德距離指代兩個節點之間的空間距離。可根據應用於網路 中之路由協定來挑選該距離之度量。若路由協定使用跳躍 汁數度量，則將易於判定兩個節點之間的跳躍距離。同樣 地，當筇點之網路位址係與其地理位置或Gps位置有關 時，使用基於GPS之距離或歐幾里德距離來定義發送器節 點與接收器節點之間的距離將為有利的。 I59105.doc 201218695 較佳地，可將控制單元之通信功能再分為多個不同層。 接著’較低協定層（例如，媒體存取控制（MAC)層）可經調 適以判定該資料封包傳輸是否仍保持未應答。控制單元之 較高協定層（例如，網路層、傳送層或應用層）可經調適以 考慮可用於基礎較低協定層之資訊，或控制單元之下部較 低協定層可經調適以考慮可用於較高協定層之資訊。舉例 而言，網路層可使用由MAC層判定之參數。藉助於此跨層 通信，系統變得更為可靠且靈活。在一實施例中，藉由負 責丟失資料封包之點對點重新傳輸之較高協定層來調整應 答逾時。此外，可藉由跨層通信將與發送器節點與目的地 節點之間的距離有關的距離資訊自較低協 協定層。或者或另外，可自路由表、跳躍計數器及至:;子 留時間計數器獲得距離資訊β可能地，將路由表儲存於該 裝置中。該跳躍計數器或該存留時間計數器可包括於資料 封包中。接著’目的地節點可自跳躍計數器及/或自存留 時間計數器判定距離資訊，以便調整用於待以反向(亦 I ’自目的地節點至發送器節點)傳輸之資料封包的答覆 =夺。然而’亦可針對複數個目的地節點而將跳躍計數或 ^時料㈣存於該發送器節財。亦可使用相應反向 Si始存留時間計數與最终存留時間計數之間的差來 訊。可接著儲存此資訊以用於未來資料封包傳 間計數為已知的或對於系統中 節點:相等的。或者或另外，可自接收自該目的地 A /或自用於建置無線網路之路由表之技術來 >59105.d〇c 201218695 導出距離資訊。 在另一實施例中，該答覆逾時係基於跳躍計數及跳躍時 間。舉例而言，可將答覆逾時計算為：time_〇ut= 2*h〇P__count*h〇P_time*常數。此處，該常數可大於卜跳 躍計數可為自目的地節點接收之最後資料封包的跳躍計 數、自目的地節點接收之最後《個資料封包之平均值、自 目的地節點接收之最後《個資料封包中的最大跳躍計數、 自目的地節點接收之資料封包的跳躍計數隨時間的移動平 均值’或其類似者。同樣地，跳躍時間可對應於將資料封 包自-節點轉遞至另-節點（亦即’對於一個跳躍）平均所 需之平均跳躍時間。或者，該跳躍時間可指代來自目的地 節點之最後《個資料封包的平均跳躍時間，指代最大跳躍 時間，指代跳躍時間隨時間之移動平均值，或指代成功地 執订預定百分比之跳躍的成功跳躍時間。此外’可基於待 發送之資料封包的類型而調整答覆逾時，例如，該^料封 包為時間關鍵資料封包抑或非時間關鍵資料封包，或該資 枓封包具有何種優先權等級。因此，該裝置之該控制單元 =一步能夠判定資科封包之類型。或者或另外，該答覆 疋時可基於將導致較高跳躍時間之在 載。此當前訊務負載可（例如^的田月】訊4負 ..、）稭由發送态節點Α所觀察或 接收之網路訊務量或藉由 枚 = : 中發出之^ _ 、态即點Α在最近時間週期Τ 甲赞出之封包的數目來判 nrwklb 次者，接收器節點B及/或中 曰1 P點中之任一者可將 田月〗，.周路負載之資訊（例如）作 為貝枓封包之部分報告給發送器節點。 ）作 159105.doc 201218695 在另-較佳實施例令’當節點在多跳躍資料 充當令間節點時’該裝置之該控制單元可基於資料封包已 订進之距離而調整用於轉遞所接收之資料封包之傳輸參 數。此情形增加行進長距離之資料封包在到達其目的地節 點之削經歷最終跳躍的機率。舉例而言，該 包括較低協定層處之重新傳輸之最大數目、媒體存取嘗試 之最大數目、傳輸功率位準、用於重新傳輸之延遲時間， 及用於媒體存取嘗試之後移時間。當媒體存取為空時，媒 體存取嘗試與载波感測及資料封包之後續傳輸或重新傳輪 的程序有關。因此，用於媒體存取嘗試之後移時間表示後 續媒體存取嘗試之間的時間間隔。同樣地’用於重新傳輸 之延遲時間指代後續重新傳輸之間的時間。傳輸功率位準 係與所傳輸之資料封包之信號強度有關。因此，行進長距 離之資料封包將以行進短距離之資料封包為代價而排定為 優先’從而導致遠距節點之間的資料封包傳輸之點對點延 遲降低且導致網絡中點對點延遲之固有均勻性。尤其在大 規模燈具網路中此情形將具有燈具行為同步（例如，回 應於廣播調光命令）的優勢。201218695 VI. Description of the Invention: [Technical Field] The present invention relates to an apparatus, system and method for controlling data packet transmission in a wireless network. [Prior Art] Recently, wireless mesh networks have attracted more and more attention, for example, for remote control of lighting systems, building automation, surveillance applications, sensor systems, and medical applications. In particular, remote management of outdoor luminaires (so-called telemanagement) has become increasingly important. On the one hand, this is driven by environmental concerns because remote control systems or so-called telex management systems enable, for example, different dimming styles over time, weather conditions and seasons, allowing outdoor lighting systems More energy efficient use. On the other hand, this is also driven by economic reasons, which is also due to increased energy efficiency and reduced operating costs. In addition, the system can remotely monitor power usage and sidelight failure, which allows for the optimal time to repair or replace the light. Gossip Although a radio frequency (RF) based wireless solution uses a star network topology or a mesh topology. In a star network, the data controller is in the network and has a direct communication path to each node. “'This usually requires placing a high-power/high-sensitivity base-top control state at a high level (eg, 'on top of a building'), which makes the solution deployment cumbersome and expensive. In a mesh network, several nodes are generally not Directly communicate with the animal husbandry, and the state of the system, but communicate with the controller via the so-called jump communication. In the multi-track eve... the hop node in January will transmit the information packet to the destination and send it to the destination. Node. Node 159105.doc 201218695 acts as a router to transmit data packets from neighboring nodes to nodes that are too far away to reach a single hop, resulting in a network that can span a large distance by -connect_short The jump breaks the long distance and the signal strength is maintained. Therefore, routing is performed by all the nodes in the mesh network, thereby determining which neighboring node to send the data packet to. Therefore, the mesh network has A very robust and stable network that is connected and therefore has high redundancy and reliability. In the prior art, the mesh transmission technology can be divided into two groups based on/missing mesh networks. And a routing-based mesh network. In an overcast-based mesh network, all data packets are forwarded by all nodes in the network. Therefore, a node does not have to make complex routing decisions, but only broadcasts. The material system is quite robust by means of such means. However, in the large network, the 'data attribute attached to the transfer affects the total data rate.' In addition, the 'data packet conflict is more likely to occur, thus further reducing the total Performance. Therefore, the main problem with this solution is scalability. Route-based mesh networks can be further divided into proactive and reactive solutions. In a route-based active mesh network, all needs are The network path is stored in the routing table in each node. The routing table is kept up to date, for example, by sending periodic beacon messages to neighboring nodes to discover valid routing paths, although data is transmitted over such networks. The middle system is very effective, but the scalability is still low, because in the large network, the active update of the routing table consumes most of the network resources. The table will grow with the network. In addition, the network setup requires time and resources to build the routing table. In contrast, the reactive solution avoids 159105.doc by finding routes on demand. 201218695 Free permanent attachment Item and large routing table. The reactive scheme uses flooding to discover the network path and cache the active route or node. When multiple routes are used for a single data packet, the general data packet instead of the route discovery can be More efficient. If the route remains long enough to avoid frequent routing, the reactive scheme degenerates to the active scheme. An example of a reactive mesh network for routing is used in ZigBee. However, the main problem with this protocol scheme Still the scalability of the network. In large-scale multi-hop networks, the number of hops that data packets must travel is larger than the hop distance in a small network. In the big dragon frequency with thousands of nodes In the telex tube (4), there may be 20 to 40 jumps. However, the delivery opportunity for individual data packets decreases with the distance they jump. This is because there is a chance that data packets will be lost for each. Therefore, very limited network scalability constitutes a common wireless mesh: a large defect in the road. This is due to the fact that each data packet or message is transmitted multiple times due to the transfer, thereby reducing the total network throughput. Moreover, data packet collisions are more likely to cause data packets to be lost, further reducing overall performance. Therefore, the success and reliability of the improved multi-hopping point-to-point transmission is especially critical in large-scale multi-hop networks (such as 'Street with a large number of lighting nodes:: Ming system') because of the point-to-point retransmission ratio The smaller network makes it more time consuming and intensive. Therefore, large-scale wireless mesh networks require effective paths and reduced point-to-point delays in order to achieve the required time and robustness. In order to determine whether the data packet was successfully delivered or not, it is usually transmitted in the 158105.doc 201218695 mode. In the hop-by-hop answer mode, each hop of the multi-hop transmission is acknowledged by the receiving node of the previous transmitting node. However, this situation results in a high network load. Therefore, a point-to-point response is often used, in which the final destination node acknowledges the data packet to the initial sender node. In this mode, the sender node waits for a predetermined time before retransmitting the data packet (in anticipation of the response to the data packet). (The so-called response, overtime, in general, the response timeout is fixed and all nodes of the network are common. Because the response of the data packet for traveling short distances is over time and the data used for long distance travel The response of the packet is the same as the timeout: so the retransmission delay of the data packet traveling short distance is unnecessarily increased, thereby affecting the total transmission speed of the network. This situation causes communication between any two nodes in the network. Large delays and delay differences, resulting in poor user experience, for example, due to high and/or uneven delays when interacting with the lighting nodes of the lighting system. If this defect is simply reduced by the network Resolving size will further reduce scalability. Therefore, the point-to-point delay of successful data packet delivery should be minimized. 2009/0154395 A1 describes a routing method for clustering mesh networks for sharing the effect of energy measurement with adjacent nodes to minimize transmission delay. By means of such means, selecting the next node, the data The packet is to be forwarded to the next node in the multi-hop transmission. SUMMARY OF THE INVENTION The object of the present invention is to transmit data packets in a wireless network. Apparatus, system, and method that minimizes point-to-point retransmission delay and retransmits point-to-point retransmission while maintaining or even increasing network scalability. The goal is addressed by the characteristics of the independent request item. (4) Based on the following ideas: • Based on the distance between the sender and the destination node that must be covered by the data packet, the response timeout for a given pair of sender nodes and destination nodes is minimized. The reply timeout is related to the waiting time for sending a W-point to wait for a reply to the data packet transmitted to the destination node. If the sender node does not remain after the time period of the time period Upon receipt of the reply data packet, the sender node will retransmit the data packet. This situation eliminates unnecessary communication delay between retransmission of unacknowledged data packets. In this: When a node of the network operates as a sender node, providing means for controlling data packet transmission for the node. The device includes a control unit, and the control unit is based on the sender node and the destination node The answer is adjusted by the distance. The reply is defined as the time interval during which the sender node waits for a reply to the data packet sent. # The reply timeout has expired and the answer The sending n node has not received the answer from the destination node " 'the sender node will start retransmission of the data packet. (4) individually adjust the answer timeout for each pair of sender nodes and destination nodes' The delay in detecting a failed transmission is reduced, thus reducing the point-to-point delay due to successful transmission of the retransmission to be close to the smallest possible value. Thus, in a large-scale illumination line, this situation will further reduce the delay of the control command so that the luminaire node will react more quickly, for example, to dimming or switching commands. In the second preferred embodiment, the reply timeout and the response timeout (i.e., the sender defect in retransmitting the data packet is a response to the destination node that is not successful in the data packet transmission) At the time interval) related. Possibly, the response from the destination node contains additional material in close proximity to the response portion. By doing this, the network connection is made more efficient, because there is no need to send other packets containing additional data. Therefore, in larger terms, the response timeout can also refer to the time interval that the sender node waits. Hand... Destination Point Information Preferably, the multi-hop transmission is used in a peer-to-peer mode to transmit data packets from the sender node to the destination node via the intermediate node. This means that = the data packet has arrived at its destination node and the successful transmission is made in the embodiment - the existing point can be added or tapped to the wireless network:: Control Center. Therefore, the device is associated with a network node, and the network can also be a data collection device. The data collector node can be any node configured to communicate with the control center of the network and can act as a gateway h. For example, the device can be adapted to plug into an existing circuit board or to connect to an existing control of the node. Device. This situation is especially useful for improving or upgrading existing systems (such as 'tracking lighting systems'). In addition to the control unit, the device may further comprise a memory and/or a transceiver unit for receiving and transmitting data packets. A wireless network can have a mesh twist, where each node can act as a router. This network has increased redundancy and reliability. Preferably, the wireless network is fixed in place because the wireless network is primarily directed to the condition of the large room 159105.doc 201218695 external lighting system. Alternatively or additionally, at least some of the locations of the nodes may be known to at least some of the other nodes of the network and/or to the control center of the network. For example, at least some of the nodes may store routing tables for data packet transmissions from respective nodes to the most recent data collector node. Preferably, the routing protocol for data packet transmission to the nearest data collector node is based on a plurality of pairs of routing packets being transmitted to neighboring nodes that are closer to one of the data collector nodes. By this means, data packet transmission becomes faster and more efficient. In addition, this situation also allows for the elimination of multiple data collector nodes in large wireless mesh networks (eg, in street lighting systems with more than one illuminator node), thereby increasing redundancy and Improve the reliability of the network. In another embodiment, the distance between two nodes (eg, a sender node and a destination node or an intermediate node) is defined by a jump distance, a GPS based distance, and/or an Euclidean distance. . The jump distance between two nodes may be characterized by: a hop count (i.e., the number of hops required to transmit a data packet between two nodes) or a transfer of a data packet to an intermediate node of the final destination node Number of. The distance based on Gps can be derived from the GPS position of the sender node and the destination node, and the Euclidean distance refers to the spatial distance between the two nodes. The measure of the distance can be chosen based on the routing protocol applied to the network. If the routing protocol uses the jump juice number metric, it will be easy to determine the jump distance between the two nodes. Similarly, when the network address of the defect is related to its geographic location or GPS location, it may be advantageous to use a GPS-based distance or Euclidean distance to define the distance between the transmitter node and the receiver node. I59105.doc 201218695 Preferably, the communication function of the control unit can be subdivided into a plurality of different layers. The lower protocol layer (e. g., the Medium Access Control (MAC) layer) can then be adapted to determine if the data packet transmission remains unanswered. The higher protocol layer of the control unit (eg, the network layer, transport layer, or application layer) may be adapted to take into account information available to the underlying lower agreement layer, or the lower protocol layer below the control unit may be adapted to consider available Information at the higher agreement level. For example, the network layer can use parameters determined by the MAC layer. With this cross-layer communication, the system becomes more reliable and flexible. In one embodiment, the response timeout is adjusted by relying on the higher protocol layer of the point-to-point retransmission of the lost data packet. In addition, the distance information related to the distance between the sender node and the destination node can be derived from the lower protocol layer by cross-layer communication. Alternatively or additionally, the routing table may be stored in the device from the routing table, the hop counter, and to the :; The hop counter or the persistence time counter can be included in the data packet. The destination node can then determine the distance information from the skip counter and/or the self-residence time counter to adjust the answer for the data packet to be transmitted in the reverse direction (also I' from the destination node to the sender node). However, the hop count or the ticker (4) may also be stored in the sender for a plurality of destination nodes. It is also possible to use the difference between the corresponding reverse Si start time count and the last retention time count. This information can then be stored for future data packet pass counts to be known or for nodes in the system: equal. Alternatively or additionally, the distance information may be derived from the destination A/or from the routing table used to construct the wireless network >59105.d〇 201218695. In another embodiment, the reply timeout is based on the hop count and the hop time. For example, the answer timeout can be calculated as: time_〇ut= 2*h〇P__count*h〇P_time* constant. Here, the constant may be greater than the hop count, which may be the hop count of the last data packet received from the destination node, the average of the last data packet received from the destination node, and the last data received from the destination node. The maximum hop count in the packet, the moving average of the hop count of the data packet received from the destination node over time 'or the like. Similarly, the hop time may correspond to the average hop time required to average the data packet from the - node to the other node (i.e., 'for a hop'). Alternatively, the hop time may refer to the average hop time of the last "data packet from the destination node, refers to the maximum hop time, refers to the moving average of the hop time over time, or refers to successfully ordering a predetermined percentage. Jumping success jump time. In addition, the reply timeout may be adjusted based on the type of data packet to be sent, for example, whether the packet is a time critical data packet or a non-time critical data packet, or what priority level the resource packet has. Therefore, the control unit of the device = one step can determine the type of the packet. Alternatively or additionally, the answer may be based on the load that will result in a higher jump time. The current traffic load can be (for example, ^Tianyue] 4 negative..,) The amount of network traffic observed or received by the transmitting node or by ^ _ in the suffix = : Clicking on the number of packets in the most recent time period Τ A to judge nrwklb times, any of receiver node B and/or lieutenant 1 P point can be used to report the information of Tian Yue, Zhou Lu load ( For example) report to the sender node as part of the beta packet. 159105.doc 201218695 In another preferred embodiment, 'when a node is acting as an inter-shipment node in a multi-hop data', the control unit of the device can adjust for receiving the received based on the distance the data packet has been subscribed to. The transmission parameters of the data packet. This situation increases the probability that the data packet traveling over a long distance will eventually jump when it reaches its destination node. For example, the maximum number of retransmissions at the lower protocol layer, the maximum number of media access attempts, the transmission power level, the delay time for retransmission, and the mediation time for media access attempts. When the media access is empty, the media access attempt is related to the carrier sensing and subsequent transmission or retransmission procedures of the data packet. Thus, the time interval for media access attempts represents the time interval between subsequent media access attempts. Similarly, the delay time for retransmission refers to the time between subsequent retransmissions. The transmission power level is related to the signal strength of the transmitted data packet. Therefore, the data packets traveling long distances will be prioritized at the expense of data packets traveling over short distances, resulting in a point-to-point delay in data packet transmission between distant nodes and an inherent uniformity of point-to-point delays in the network. This situation will have the advantage of synchronizing the luminaire behavior (e. g., responding to broadcast dimming commands), especially in large scale luminaire networks.
可藉由無線射頻（RF)傳輸來執行資料封包傳輸。因為RF 傳輸無需高傳輸功率且易於實施及部署’所以可降低使用 該裝置設置及操作網路之成本。此對於大型網路(例如， 用於照明系統之電傳管理網路）尤為重要。然而，資料封 包傳輸可替代地使肢外線通信、自由空間可見光通信或 電力線通信。 159105.doc 12 201218695 在一較佳實施例中，該裝置用於室外照明系統之燈具節 點中以用於燈具節點之電傳管理。因m於將燈具節 點接通/切斷，及/或可基於參數（諸如，當天時間 (daytime)、季節、天氣、環境亮度、交通事故之發生道 路施工之存在等)而控制燈具節點之調光樣式。可能地， 藉由隨燈具節點-起提供之感測器來判定此等參數中之至 少一些者且將其報告給控制中心。 在本發明之另一離檨中，样视 _ 〜银T钕供一種用於在無線網路中控 制資料封包傳輸之系統。該系統包含—控制中心及複數個 節點。該控制中心及該等節點中之至少—者包含根據上文 所描述之實施例中之-者的裝置。該控制中心經調適以控 制無線網路巾之㈣之魏絲作。舉例而言，#節點與 照明系統（例如’街道照明系統）之燈具相關聯日夺，控制中 心可基於節.點之空間分散關於其調光樣式及操作狀態而個 別地及/或成群地控制節點。較佳地，節點中之至少一者 包含記憶體及/或感測器，點包含感測器，則節點可 經調適以將感測器資料傳輸至控制中心。 在本發明之另一態樣中， Ύ 乂供一種用於在具有複數個節 點之無線網狀網路中控制眘杻4 市』貧枓封包傳輸之方法。根據此方 法’將資料封包自發送器節戟恭、玄=。u l — 亞即點發送至目的地節點，其中基 於發送器節點與目的地笳fJfc q P點之間的距離而調整用於資料封 包之答覆逾時。步驟之次庠么 人斤為任意的且可改變。因此，可 在傳輸資料封包之前執行調整應答逾時之步驟。較佳地， 此方法應詩詩照明“之電傳管理系統中β 159105.doc -13- 201218695 【實施方式】 本發明之較佳應用為室外照明系統（例如，用於街道、 停車場及么共區域）、用於一般區域照明之室内照明系統 (例如，用於購物商場、競技場、停車場、車站、隧道等） 或感測器網路。在下文中，將使用用於街道照明之室外照 明系統的實例來進-步解釋本發明。在照明控制領域中， 經由射頻網路技術的室外燈具之電傳管理正接收愈來愈多 的關注，尤其係適用於（例如）具有2〇〇個燈具以上之區段的 大規模安裝的解決方案。 在圖1中，展不具有網狀拓樸之典型網路。複數個節點 1〇(Ν)藉由無線通信路徑4㈣彼此連接i等節點中之 一些者充當資料收集器節點50(N/DC)，其經由單跳躍或多 跳躍傳輸接收來自㈣節點1()之資料封包並將料資料封 包傳輸至控制t心60，且其接收來自控制中心6()之資料封 包並將該等資料封包傳輸至節點10。因此，資料收隼器節 ㈣可以節點_控制中心60之間的閘道器之方式 ’郎點H)與資料收集器節點5G之間的無線通信路徑辦由射 頻傳輸構成’而資料收集器節點5〇與控制中心⑽之間的連 接7〇可利用網際網路、行動通信網路、無線電系統、乙太 ，冯路DSL、料或其他有線或無線資料傳輸系統。 在用於室外照明控制之電傳管理系統中，通信為非常不 對稱的。大多數訊務係由節點！〇產生，（例如）將節點ι 〇之 狀態、感測器值或功率使用報告給控制中心6〇。其他訊務 由自控制中心60至不同節點1〇之控制命令（例如，用於調 159105.doc •14· 201218695 ㈣光樣式或將燈㈣/切斷）組成。因此’大多數訊務係 由N對1戒務（單播）構成，而自控制中心6〇至節點1〇之訊務 由~1對>^訊務（以單播 '多播或廣播模式）構成。此外，燈具 即點10之數目在室外照明系統（諸如，街道照明系統）中為 極其高#。因此，尤其在與常見無線網狀網路（其通常含 有少於200個之節點）相比較時，該網路之大小係非常大 的另外’歸因於成本考慮，節點10具有有限之處理能 力’使得燈具節點10中之處理及記憶體資源將受限制。因 此，用於在單-節點10之間傳輸資料封包之通信協定應考 慮用於有效及快速資料封包傳輸的有限資源。此外，盜其 他所謂特用網狀網路相比，用於室外照明控制網路之電傳 f理系統為固定的，亦即，節點1G不移動。又，所有燈具 卽點1 0可連接至雷诉雷。 ▲ ’、 因此，網路改變將主要歸因於改 變之環境，例如，歸因於訊務。因為節點1〇為固定的，、 以節點Π)之實體位置（例如，GPS座標）在系統中可為已知 的:從而實現地理路由或基於位置之路由。此外，室外照 之訊息或資料封包需要低回二：二’存在特定類型 士 間的許多情形。舉例而 ::到父通事故時，可控制相應區 即刻切換至全功率。 疋 可於如上文所提及之室外照明系統的特定應用性質， 應用以下特徵。可藉由泛播(fl0 器節點50至各別燈旦筋赴…… g)來執仃自資料收集 路中之所有接1封包傳輪，其中藉由網 接收卽點1G轉遞所有資料封包。資料封包至少 159105.doc 201218695 含有關於發送器節點1G及-或多個目的地節點ι〇之資訊 接著藉由該至少—目的地節點10解碼資料封&。對於°自燈 具節點U)至資料收集器節點50之資料封包傳輸，基於心 之解決方案係較佳的，纟中每—節點1Q選擇較接近資料收 集器節點50中之—者的相鄰節點⑽為中間節_。較佳 地，因為定期地使用至資料收集器節點5()之路線，所以使 二主動式路由結構4主動式路由結構中路由表儲存於 母-節點H)中’指示哪—相鄰節點1()較接近於資料 節點50中之-者。因此’可以非常有效且快速之方式將資 料封包發送至最近資料收集器節點5〇。有利地，每一節點 保持關於多個下行鍵路相鄰節點1()之資訊作為替代路 線，以便增加可靠性。若一相鄰節點1〇歸因於強干擾或完 全故障而不可到達’則路由協定具有額外替代路線以將^ 料封包投送至資料收集器節點50。 在圖2A中’展示由複數個節點_繞之資料收集器節點 50說明自發送器節點A經由複數個中間節點至資 2收集器節點50(目的地節點B)之多跳躍單播資料傳輸。 P 占10具有如由半控5〇1及5〇2指示的至資料收集器節點 之不，同跳躍距離。舉例而t ’在半徑5〇1内但在半徑5〇2之 卜的節點A將需要兩個跳躍w及h2以用於將資料封包傳輸 至為目的地節點B之資料收集器節點5〇，亦即，必須經由 j *尹間崤點N1將資料封包自此節點八傳輸至資料收集 器節點50。對比而言’在半徑502内之節點】〇可以一個跳 躍將其資料封包傳輸至資科收集器節點50。當然，目的地 159105.doc • 16 · 201218695 郎點B可為任何節點1〇且未必為資料收集器節點％。因 此’可針對每—對發送器節點A及目的地節點B來定義跳 躍距離。用於特性化跳躍距離之參數為跳躍計數，亦即， 將-貝料封包自發送器節點Α傳輸至目的地節點Β所需的跳 躍之數目。 在圖2B中’說明發送器節點a與目的地節點b之間的歐 幾里德距離d。在任何兩個節點丨〇之間，將歐幾里德距離 疋義為兩個點之間的地理距離。若節點丨〇之網路位址係基 於各別節點10之GPS位置，則亦可使用基於Gps之距離。 接著將兩個節點10之間的距離定義為其GPS位置之間的距 離。詳言之，當網路之節點1〇平均地分散於網路區域上 時，兩個節點之間的歐幾里德距離或基於Gps之距離將為 在兩個節點之間傳輸資料封包時平均執行的跳躍之數目的 特性，且因此亦為傳輸時間之特性。或者，該距離可指代 由資料封包行進之實際距離。在室外照明網路中，可使用 沿街道量測之距離，而非歐幾里德距離，此係因為資料封 包將很可能沿此等路徑行進。在圖2C令說明此情形，展示 沿街道配置的街道燈具節點1〇。因此，兩個燈具節點1〇之 間的距離亦可指代街道距離，將街道距離定義為沿道路系 統之街道的空間距離或跳躍距離。 在圖3中，展示根據本發明之裝置1〇〇。裝置1〇〇可與無 線多跳躍網狀網路之節點10或資料收集器節點5〇(例如， 照明系統之燈具）相關聯。裝置1〇〇包含控制單元2〇〇。此 外，節點10或50或裝置100包含用於經由無線通信路徑 I59105.doc •17- 201218695 40(例如’經由射頻傳輸）傳輸或接收資料封包之收發單元 300 〇 裝置100之控制單元200可根據其在資料封包傳輸中之功 能而再分成多個不同層。舉例而言，當使用0SI層模型 時，控制單元200將包含用於定義裝置1〇〇與傳輸媒體之互 動的實體層；在多節點網路中提供定址及頻道存取控制機 制的MAC層；提供複數個功能及程序（例如，網路路由功 能）之網路層；使用（例如）流量控制、分段/解分段或錯誤 控制將可靠資料傳送服務提供至較高協定層之傳送層，及 用於識別通信夥伴、狀資源可用性或同步通信的應用 層。 在下文中，將使用調整應答逾時（亦即，任何發送器節 點在重新傳輸所發送之資料封包之前等待來自目的地節點 之對該資料封包的應答之時間週期）之實例來解釋本發 明。然而，本發明亦可應用於調整一般答覆逾時，在該答 覆逾時期間，發送器節點A等待來自目的地節點b之答覆 資料封包。因此，本發明係關於調整答覆逾時且不限於應 答逾時。 在圖4中，展示使用調整應答逾時之實例說明本發明之 流程圖，但本發明不限於此。當處理待發送之資料封包時 或當發送資料封包（S40)時，判定發送器節點A與目的地節 點B之間的距離（S41)。基於此距離，調整用於此資料封包 之應答逾時（S42)。因此，在調整該應答逾時之時，相應 地排程在遺漏應答之狀況下的傳送層/應用層重新傳輸。 159105.doc -18- 201218695 大體而言’應答逾時應採用儘可能小之值，以便最小化點 對點通信延遲。在先前技術中，應答逾時為固定且對於所 有節點ίο為共同的。因此，緊密相鄰之節點1〇之通信延遲 等於遠距節點10之通信延遲。然而，在針對每一個別發送 器-目的地組合來調整應答逾時（如由本發明所建議）之時， 可最小化應答逾時，因此最小化點對點通信延遲。應答逾 時之最小值等於源節點與目的地節點之間的資料封包傳輸 之預期往返時間，亦即，用於遞送資料封包之時間加上用 於遞送應答之時間》當發送器節點A在應答逾時内未接收 到對所發送之資料封包之應答時，發送器節點A將重新傳 輸該貪料封包（S43)。可改變該等步驟之次序，亦即，可 在發送資料封包（S4〇)之前執行步驟以丨及/或以之。 I將該等步射之—些者執行達奸次或㈣圈反覆地Data packet transmission can be performed by radio frequency (RF) transmission. Because RF transmissions do not require high transmission power and are easy to implement and deploy, the cost of setting up and operating the network using the device can be reduced. This is especially important for large networks (for example, telex management networks for lighting systems). However, data packet transmission can alternatively enable extra-body communication, free-space visible light communication, or power line communication. 159105.doc 12 201218695 In a preferred embodiment, the apparatus is used in a lighting fixture of an outdoor lighting system for telex management of a lighting fixture. Controlling the lighting of the luminaire node based on the switch on/off of the luminaire node and/or based on parameters such as daytime, season, weather, ambient brightness, presence of traffic accidents, etc. Light style. Possibly, at least some of these parameters are determined by the sensor provided with the luminaire node and reported to the control center. In another departure of the present invention, the sample _ _ silver T 钕 provides a system for controlling the transmission of data packets in a wireless network. The system consists of a control center and a number of nodes. The control center and at least one of the nodes comprise means according to the embodiments described above. The control center was adapted to control the Wisdom of the wireless network towel (4). For example, the # node is associated with a luminaire of a lighting system (eg, a 'street lighting system) that can be individually and/or in groups based on the spatial distribution of the points. Control node. Preferably, at least one of the nodes includes a memory and/or a sensor, and the point includes a sensor, and the node can be adapted to transmit the sensor data to the control center. In another aspect of the invention, a method for controlling the transmission of a poor packet transmission in a wireless mesh network having a plurality of nodes is provided. According to this method, the data is encapsulated from the sender's section, 戟, 玄 =. u l — The point is sent to the destination node, where the response timeout for the data packet is adjusted based on the distance between the sender node and the destination 笳fJfc q P point. The second step of the procedure is that the person is arbitrary and can be changed. Therefore, the step of adjusting the response timeout can be performed before the data packet is transmitted. Preferably, the method is applied to the poem lighting "beta" 159105.doc -13 - 201218695. [Embodiment] The preferred application of the present invention is an outdoor lighting system (for example, for street, parking lot, and Area), indoor lighting system for general area lighting (for example, for shopping malls, arenas, parking lots, stations, tunnels, etc.) or sensor networks. In the following, outdoor lighting systems for street lighting will be used. An example of the present invention is explained in a step-by-step manner. In the field of lighting control, the telex management of outdoor luminaires via radio frequency network technology is receiving more and more attention, especially for (for example) having 2 luminaires. A solution for large-scale installation of the above segments. In Figure 1, a typical network without a mesh topology is shown. A plurality of nodes 1〇(Ν) are connected to each other by a wireless communication path 4 (4). Some act as a data collector node 50 (N/DC) that receives a data packet from (4) node 1() via a single-hop or multi-hop transmission and transmits the data packet to the control t-key 60, and receives it The data packet from the control center 6() is transmitted and transmitted to the node 10. Therefore, the data collector section (4) can be used as a gateway between the node_control center 60 and the data collection. The wireless communication path between the node 5G is constituted by radio frequency transmission and the connection between the data collector node 5 and the control center (10) can utilize the Internet, the mobile communication network, the radio system, the Ethernet, the von DSL, material or other wired or wireless data transmission system. In the telex management system for outdoor lighting control, the communication is very asymmetrical. Most of the traffic is generated by the node!, for example, the node ι The status, sensor value or power usage report is sent to the control center 6. Other services are controlled by commands from the control center 60 to different nodes (for example, for 159105.doc •14·201218695 (4) light patterns Or consist of lights (4) / cut off. Therefore, 'most of the services are composed of N to 1 (unicast), and the traffic from the control center 6〇 to node 1〇 is ~1 pair> Unicast (multicast or broadcast mode) In addition, the number of luminaires, point 10, is extremely high in outdoor lighting systems (such as street lighting systems). Therefore, especially with common wireless mesh networks (which typically contain fewer than 200 nodes) In comparison, the size of the network is very large. In addition, due to cost considerations, the node 10 has limited processing power, so that the processing and memory resources in the lighting node 10 will be limited. Therefore, for the single - The communication protocol for transmitting data packets between nodes 10 should consider limited resources for efficient and fast data packet transmission. In addition, the use of other so-called special mesh networks for telemetry of outdoor lighting control networks The system is fixed, that is, node 1G does not move. In addition, all lamps and lanterns can be connected to the mine. ▲ ‘, therefore, network changes will be primarily attributed to changing environments, for example, due to traffic. Since node 1 is fixed, the physical location of the node (e.g., GPS coordinates) can be known in the system: thereby enabling geographic routing or location based routing. In addition, outdoor photo messages or data packets need to be low back to two: two. There are many situations in a particular type of relationship. For example: :: When the father passes the accident, you can control the corresponding zone to switch to full power immediately. The following features can be applied to the specific application properties of the outdoor lighting system as mentioned above. You can use the panning (fl0 node 50 to each light to go to ... g) to execute all the packets from the data collection road, where all the data packets are forwarded by the network receiving point 1G. . The data packet is at least 159105.doc 201218695 contains information about the sender node 1G and/or the plurality of destination nodes ι〇 and then decodes the datagram & by the at least-destination node 10. For the data packet transmission from the luminaire node U) to the data collector node 50, the heart-based solution is preferred, and each node 1Q selects a neighboring node that is closer to the data collector node 50. (10) is the middle section _. Preferably, because the route to the data collector node 5() is used periodically, the routing table in the active routing structure of the active routing structure 4 is stored in the parent-node H) to indicate which neighboring node 1 () is closer to the one in the data node 50. Therefore, the data packet can be sent to the most recent data collector node 5 in a very efficient and fast manner. Advantageously, each node maintains information about a plurality of downlink-keyed neighbors 1() as an alternative route to increase reliability. If an adjacent node 1〇 is not reachable due to strong interference or a complete failure, then the routing protocol has an additional alternative route to deliver the packet to the data collector node 50. The multi-hop unicast data transmission from the sender node A to the resource 2 collector node 50 (destination node B) from the sender node A is illustrated in Figure 2A by a plurality of node_wrap data collector nodes 50. P occupies 10 with the same jump distance as indicated by the semi-controls 5〇1 and 5〇2 to the data collector node. For example, node A, which is within radius 5〇1 but at radius 5〇2, will need two jumps w and h2 for transmitting the data packet to the data collector node 5 that is the destination node B. That is, the data packet must be transmitted from the node eight to the data collector node 50 via the j*instance point N1. In contrast, a node within radius 502 can transmit its data packet to the korea collector node 50 in one hop. Of course, the destination 159105.doc • 16 · 201218695 朗点 B can be any node 1〇 and not necessarily the data collector node %. Therefore, the hop distance can be defined for each of the sender node A and the destination node B. The parameter used to characterize the hop distance is the hop count, i.e., the number of hops required to transfer the packet to the destination node from the transmitter node. The Euclidean distance d between the sender node a and the destination node b is illustrated in Fig. 2B. Between any two nodes, the Euclidean distance is defined as the geographical distance between two points. If the network address of the node is based on the GPS location of the respective node 10, a distance based on Gps can also be used. The distance between the two nodes 10 is then defined as the distance between their GPS positions. In particular, when the nodes of the network are evenly distributed over the network area, the Euclidean distance or the Gps-based distance between the two nodes will be averaged when transmitting data packets between the two nodes. The nature of the number of hops performed, and therefore also the characteristics of the transmission time. Alternatively, the distance may refer to the actual distance traveled by the data packet. In an outdoor lighting network, the distance measured along the street can be used instead of the Euclidean distance because the data package will likely travel along these paths. This situation is illustrated in Figure 2C, which shows street lighting nodes 1〇 configured along the street. Therefore, the distance between two luminaire nodes 1 亦可 can also refer to the street distance, which is defined as the spatial distance or jump distance of the street along the road system. In Fig. 3, a device 1 according to the invention is shown. The device 1 can be associated with a node 10 of a wireless multi-hop mesh network or a data collector node 5 (e.g., a luminaire for a lighting system). The device 1〇〇 contains a control unit 2〇〇. In addition, the node 10 or 50 or the device 100 includes a transceiver unit 300 for transmitting or receiving data packets via a wireless communication path I59105.doc • 17-201218695 40 (eg, via RF transmission). The control unit 200 of the device 100 can be It is subdivided into multiple layers in the function of data packet transmission. For example, when using the OSI layer model, the control unit 200 will include a physical layer for defining the interaction of the device 1 with the transmission medium; a MAC layer providing addressing and channel access control mechanisms in the multi-node network; a network layer that provides a plurality of functions and procedures (eg, network routing functions); provides reliable data transfer services to the transport layer of the higher protocol layer using, for example, flow control, segmentation/de-segmentation, or error control, And an application layer for identifying communication partners, resource availability, or synchronous communication. In the following, the present invention will be explained using an example of adjusting the response timeout (i.e., the time period during which any transmitter node waits for a response from the destination node to the data packet before retransmitting the transmitted data packet). However, the present invention is also applicable to adjusting the general answer timeout during which the sender node A waits for a reply data packet from the destination node b. Accordingly, the present invention is directed to adjusting the timeout and not limited to the timeout. In Fig. 4, a flow chart showing the present invention using an example of adjusting the response timeout is shown, but the present invention is not limited thereto. When processing the data packet to be transmitted or when transmitting the data packet (S40), the distance between the sender node A and the destination node B is determined (S41). Based on this distance, the response timeout for this data packet is adjusted (S42). Therefore, when the response timeout is adjusted, the corresponding scheduling is retransmitted at the transport layer/application layer in the case of a miss response. 159105.doc -18- 201218695 In general, the response timeout should be as small as possible to minimize the point-to-point communication delay. In the prior art, the response timeout is fixed and common to all nodes ίο. Therefore, the communication delay of the closely adjacent node 1 is equal to the communication delay of the remote node 10. However, when the response timeout is adjusted for each individual transmitter-destination combination (as suggested by the present invention), the response timeout can be minimized, thus minimizing the point-to-point communication delay. The minimum value of the response timeout is equal to the expected round-trip time of the data packet transmission between the source node and the destination node, that is, the time for delivering the data packet plus the time for delivering the response" when the sender node A is responding When the response to the transmitted data packet is not received within the timeout, the sender node A will retransmit the greedy packet (S43). The order of the steps can be changed, i.e., the steps can be performed before and/or after the data packet (S4〇) is sent. I will take the steps to shoot - some of them perform rape or (four) circle repeatedly
執行直至達到最大重複數目。舉例而言，發送器節點A 可重新嘗試該資料封包之傳輸（S43)達若干次，直至達到 重新傳輸之最大數目。若允許若干傳輸嘗試且達到在M2 中所判定之逾時(針對第一次傳輸判定)，則可(例如)使用 以下公式來増加用於實際重新傳輸之逾時：tim、out=cl* 距離+c2*n—retransmissi〇ns，其中cac2為兩個常數且 n—mransmission4 ^新傳輸計數，在帛—次傳輸資料封 包時，η一retransmissi〇ns針對該第一傳輸為(^此情形防止 將逾時（重複地）設定為過低，而過低之逾時防止了成功傳 輸。 如上文所提及’相同程序適用於調整任何答覆逾時。因 159105.doc •19- 201218695 此，在發送器節點A已將資料封包傳輸至目的地節點 B(S40)之後，判定節點a與節點3兩者之間的距離（s4i)， 且相應地調整答覆逾時（S42p再次，可任意地變更步驟 S40、S41及S42之次序《若在該答覆逾時已屆期之後發送 器節點A仍未自目的地節點8接收到任何答覆，則發送器 節點A重新傳輸該資料封包（s 4 3 )。該答覆資料封包可包括 應答、資料（尤其係由發送器節點A所請求之資料），及其 類似者。又，來自目的地節點B之應答含有緊鄰應答部分 之額外資料。舉例而言，當資料收集器（發送器節點A)正 將發送回當前所使用之調光設定檔的請求發送至接收器節 點B時係如此。接收器節點B可接著發送回兩個封包亦 即，首先為已接收到請求之應答且接著為含有關於當前調 光設定播之回應之資料封包。然而，_地，將應答及關 於當前調光設定檔之資料組合成一個資料封包。藉由此等 手段，不需要任何額外傳輸，因此節省網路資源。 可使用-度量(諸如’跳躍距離、基於Gps之距離或歐幾 里德距離）來定義兩個節點之間的距離。在已知節點切之 位置或網路位址係基於各別節點丨〇之Gps位置的狀況下， 可自節點10之位置或各別節點1〇iGPS位置導出特定發送 器1的地組合之距離資訊。可連同平均傳輸速度使用基 於GPS之距離或歐幾里德距離來估計往返時間之預期值， 在成功傳輸之狀況下，將在該往返時間内接收到對所發送 之資料封包之應答。此值將接著判定應答潛時。或者，可 基於發送器節點A與目的地節點8之間的跳躍距離而導出 159105.doc -20· 201218695 該距離。該跳躍距離之特徵在於資料封包已行進以到達其 目的地節點B的跳躍之數目。在-些狀況下，（例如）當使 用具有跳躍計數度量之路由表時，跳躍距離資訊在網^層 處已可用。在此狀況下，目的地節點Β連同距離發送器節 點Α之各別距離一起儲存於發送器節點a中。 亦可使用包括於資料封包中之跳躍計數器或存留時間呀 數器（例如）藉由網路層明確地產生距離資訊。在自發送器 節點A至目的地節點B之多跳躍傳輸期間，跳躍計數器; 包括於資料封包中且隨每一跳躍而遞增。舉例而言，發送 器節點A可自包括於自目的地節點B接收之資料封包中的 跳躍計數器來導出針對特定目的地節點仏跳躍距離資訊 (亦即’跳躍計數此處’該跳躍計數可自接收自目的地 即點B之最後貧料封包來導出，或該跳躍計數可與自目的 地節點B接收之最後_資料封包之均值或平均跳躍計數有 關。或者’可將該跳躍計數挑選為自節點吨收之最後”個 資料封包之最大跳躍計數或挑選為最動個資料封包之跳 躍計數隨時間的滑動視窗平均值。替代包括於自目的地節 點B接收之資料封包中，該跳躍計數器亦可包括於來自目 的地節點B之對由發送器節點A發送之封包的應答中，或 其可包括於兩者中。或者，接收器節點B可自接收自發送 器節點A之資料封包的跳躍計數器來導出跳躍計數或該跳 躍計數之導出值（諸如，平均值或均值），且將此跳躍計數 傳達回至發送器節點A。當預期一路徑之跳躍計數不同於 反向路徑之跳躍計數（亦即，自人至3之跳躍計數不同於自 I59105.doc -21 - 201218695 二二A之跳躍什數)時’可需要此計算。可將跳躍計數資訊 錯存於網路層處，且將跳躍計數資訊用作至此目的地節點 Btf㈣&傳輸之㈣距離。調整應答 於發送至僅-個目的地節點B之 了應用 實施例應用於應答之廣播 ^ ’、可將该 廣播及多播狀況，其中將資料封包 丄至：干目的地節點B。在此等狀況下’可考慮用於多播 群組中之不同目的地節點R ^ 中之節點之最大逾時之不同逾時，或選擇用於群組 =地’可❹存留時間計數器。大體而言 =rTTL)為具有大於跳躍之最大所需數目之初始值的 =位。在每-中間節㈣處(亦即，在每一跳躍之 ^ 減存留時間計數器。將丢棄具有零之當前存留時 限轉遞。因此，可自初始存留==資科封包的無 牙間叶數（在第一跳躍之前） 跳躍=到資料封包時之最終存留時間計數之間的差導出 ° °此處’目的地節點B知曉存留時間計數器之初 或㈣存留時料數嵌人於資料封包巾。如上文所 ::述二由目的地節點B自資料封包提取最終存留時間計 ==可將最終存留時間計數嵌入於對該資料封包之 應善中或反向（例如，自目 的也卽B至發送器節點A)傳輸 的地節點B之封下包中，使得發送器節點A可計算具有同-目 的也郎點B之下_f料封包的跳躍計數。 如當:由::：由•使用用於建置路由表之其他技術(例 .。‘況心疋期地發送至潛在或重要的目的地節 159l05.doc -22- 201218695 點）而產生跳躍計數資訊。因此，可將應答逾時之調整應 用於動態路由協定，其中至特定目的地節點Β之跳躍的數 目可變化。將該距離資訊提供至處置點對點資料封包重新 傳輸之上層（例如’應用層或傳送層）。此上層可接著最小 化每一目的地節點或發送器-目的地組合之應答逾時值。 藉由此等手段’可將在資料封包傳輸失敗時的點對點延遲 減小為接近最小可能值。 當挑選跳躍計數度量時，可將應答逾時計算為： time_〇ut=2* 常數 *hop一count*hop_time，其中常數大於 i。 因數2係針對往返，此係因為跳躍計數僅與單向有關。或 者，可替代地使用針對往返之跳躍計數。因此，應答逾時 大於或等於平均往返時間。h〇p_time可與在多跳躍傳輸中 將資料封包轉遞至下-中間節點N i平均所需的平均時間週 期有關。此可為網路之特性。此外，跳躍時間可取決於資 料封包之類型，例如’資料封包是時間關鍵資料封包抑或 時間非關鍵資料封包，《資料封包Μ高優先權抑或以低 優先權標記。當然’亦可用其他特性時間(例士。，媒體跳 躍時間或成功跳躍時間，亦即’成功地進行特定百分比 (」列如’ 90%至99%)之跳躍的時間間隔）來替代ume。 當使用基於GPS之距離或歐幾里德距離時，使用用於所選 二性時間來類似地計算應答逾時。亦有可能考慮用 、。覆逾時之目的地節點B中之某-延遲，例如，回 j資料封包在目的地節點B處緩衝的緩衝時間、自目的地 即點B至發送器節點A之迴轉時間，或其類似者。 159I05.doc -23- 201218695 較佳地，應答逾時可另外基於將導致較高跳躍時間之在 網路中之當前訊務負载。此當前訊務負載可（例如）藉由發 送器節點Α所觀察或接收之網路訊務量或藉由發送器節點 A在最後時間週期T中發出之封包的數目來判定。或者， 接收器節點B(或中間節點Ni中之任一者）可將關於網路負 載之資訊（例如）作為資料封包之部分或連同應答一起報告 給發送器節點A。 ° 當應用於具有複數個燈具節點1〇之大型照明系統令時， 可因此最小化操作命令之延遲。詳言之，對於遠離發送操 作命f之資料收集器節點5〇(或控制中心6〇)且已具有長傳 輸時間之燈具節點1 〇，此最小化消除未應答資料封包之重 新傳輸之間的不必要之通信延遲。因此，可進一步增加照 明系統之可擴充性且可減小燈具節點1〇之回應時間。 引起點對點延遲增加之另—原因係去棄行進長距離之資 ::包。跨越網路已行進長距離之任何資料封包已消耗顯 :S之網路資源，且累積顯著之延遲。若丟棄此資料封 包’則必須由其發送器節點A將其重新傳輸（點對點重新傳 輸）’該重新傳輸使所累積之點對點延遲及網路資源（亦 f7。頻寬）之耗費至少加倍。在多跳躍網路中，任何節點 〇可充田發送器節點Α或充當將所接收之資料封包轉遞至 下一中間節點Ni或轉遞至最終目的地節點B之中間Ni。因 此根=本發明之另一實施例，無線網路之節點10不僅能 夠在2當發送器節點A時基於至目的地節點B之距離而最 】化資料封包之答覆逾時。而且在節點10操作為中間節點 159105.doc •24· 201218695Execute until the maximum number of repetitions is reached. For example, the sender node A can retry the transmission of the data packet (S43) several times until the maximum number of retransmissions is reached. If several transmission attempts are allowed and the timeout determined in M2 is reached (for the first transmission decision), the timeout for the actual retransmission can be added, for example, using the following formula: tim, out=cl* distance +c2*n—retransmissi〇ns, where cac2 is two constants and n-mransmission4^ new transmission count. In the case of 帛-trans transmission data packets, η-retransmissi〇ns is for the first transmission (^ this situation prevents Timeout (repetitively) is set too low, and too low a timeout prevents successful transmission. As mentioned above, 'the same procedure applies to adjusting any response timeout. 159105.doc •19- 201218695 This is sending After the node A has transmitted the data packet to the destination node B (S40), it determines the distance between the node a and the node 3 (s4i), and adjusts the response timeout accordingly (S42p again, the step can be arbitrarily changed) Sequence of S40, S41, and S42 "If the sender node A has not received any reply from the destination node 8 after the reply has expired, the sender node A retransmits the data packet (s 4 3 ). The reply data packet can be packaged Including the response, the data (especially the information requested by the sender node A), and the like. Again, the response from the destination node B contains additional information in close proximity to the response portion. For example, when the data collector (sends) This is the case when the node A) is sending a request to send back the currently used dimming profile to the receiver node B. The receiver node B can then send back two packets, ie first the response to the received request. And then a data packet containing a response to the current dimming setting. However, _, the response and the data about the current dimming profile are combined into one data packet. By this means, no additional transmission is required. This saves network resources. You can use - metrics (such as 'jump distance, Gps-based distance or Euclidean distance) to define the distance between two nodes. In the known node cut location or network address system Based on the position of the Gps position of each node, the distance information of the specific combination of the specific transmitter 1 can be derived from the position of the node 10 or the individual node 1 〇 i GPS position. The average transmission speed uses the GPS-based distance or Euclidean distance to estimate the expected value of the round-trip time. In the case of successful transmission, a response to the transmitted data packet will be received during the round-trip time. This value will be followed. Determining the response latency. Alternatively, the distance may be derived 159105.doc -20· 201218695 based on the hop distance between the sender node A and the destination node 8. The hop distance is characterized by the data packet having traveled to reach its destination. The number of hops for Node B. In some cases, for example, when using a routing table with a hop count metric, the hop distance information is available at the tier. In this case, the destination node 储存 is stored in the sender node a along with the respective distances from the transmitter node Α. The distance information can also be explicitly generated by the network layer using a hopping counter or a retention time counter included in the data packet (for example). During multi-hop transmission from Transmitter Node A to Destination Node B, the hop counter is included in the data packet and increments with each hop. For example, the sender node A may derive a jump distance information for a specific destination node from a hopping counter included in a data packet received from the destination node B (ie, 'jump count here' the hop count may be The last poor packet received from the destination, point B, is derived, or the hop count may be related to the mean or average hop count of the last _ data packet received from the destination node B. Alternatively, the hop count may be selected as The maximum hop count of the last "data packet of the node ton receives or is selected as the sliding window average of the hop count of the most dynamic data packet over time. The substitution is included in the data packet received from the destination node B, and the hop counter is also May be included in the response from the destination node B to the packet sent by the sender node A, or it may be included in both. Alternatively, the receiver node B may self-send the data packet from the sender node A. a counter to derive a hop count or a derived value of the hop count (such as an average or mean) and communicate this hop count back to the transmission Node A. When the hop count of a path is expected to be different from the hop count of the reverse path (ie, the hop count from person to 3 is different from the number of jumps from I59105.doc -21 - 201218695 22 A) This calculation may be required. The hop count information may be stored at the network layer, and the hop count information is used as the (four) distance to the destination node Btf(4)& transmission. The adjustment response is sent to only the destination node B. The application embodiment is applied to the broadcast of the response ^', and the broadcast and multicast status can be encapsulated, where the data is packetized to: the dry destination node B. In these cases, the difference can be considered for use in the multicast group. The difference in the maximum timeout of the node in the destination node R^ is overdue, or is selected for the group=ground's available timestamp counter. In general = rTTL) is the initial value having the largest required number greater than the jump = bit. At each-intermediate section (four) (ie, at each hop ^ decrease the retention time counter. The current retention time limit with zero will be discarded. Therefore, it can be from the initial retention == 资科包Number of edentulous leaves (in the first Before a jump) Jump = the difference between the final retention time counts when the data packet is deduced. ° ° Here, the destination node B knows the time of the retention time counter or (4) the number of credits is embedded in the data packet towel. The following: The second node is extracted from the data packet by the destination node B. The final retention time count == The final retention time count can be embedded in the data packet or in the reverse direction (for example, from the destination B to the sender) The node A) transmits the encapsulated packet of the ground node B, so that the sender node A can calculate the hop count of the _f packet with the same-purpose singular point B. For example: by::: by • use Other techniques for constructing routing tables (eg, 'sent to the potential or important destination section 159l05.doc -22-201218695 points) generate jump count information. Therefore, the adjustment of the response timeout can be applied to the dynamic routing protocol, where the number of hops to a particular destination node can vary. The distance information is provided to the disposal point-to-point data packet for retransmission of the upper layer (e.g., 'application layer or transport layer'). This upper layer can then minimize the response timeout value for each destination node or sender-destination combination. By this means, the point-to-point delay when the data packet transmission fails can be reduced to be close to the minimum possible value. When picking a hop count metric, the response timeout can be calculated as: time_〇ut=2* constant *hop_count*hop_time, where the constant is greater than i. Factor 2 is for round trips because the hop count is only related to unidirectional. Alternatively, a jump count for round trips may alternatively be used. Therefore, the response timeout is greater than or equal to the average round trip time. The h〇p_time may be related to the average time period required to forward the data packet to the lower-intermediate node N i in the multi-hop transmission. This can be a feature of the network. In addition, the hop time may depend on the type of data packet, such as 'data packets are time critical data packets or time non-critical data packets, and data packets are high priority or low priority. Of course, it is also possible to replace ume with other characteristic time (example, media jump time or successful jump time, that is, the time interval at which a certain percentage ("column such as '90% to 99%" jumps successfully). When using a GPS based distance or Euclidean distance, the response timeout is similarly calculated using the selected two sex time. It is also possible to consider using . a certain delay in the destination node B over time, for example, the buffer time of the back data packet buffered at the destination node B, the turnaround time from the destination point B to the sender node A, or the like . Preferably, the response timeout may additionally be based on the current traffic load in the network that would result in a higher hop time. This current traffic load can be determined, for example, by the amount of network traffic observed or received by the sender node or by the number of packets sent by the sender node A in the last time period T. Alternatively, the receiver node B (or any of the intermediate nodes Ni) may report information about the network payload (e.g., as part of the data packet or together with the response to the sender node A). ° When applied to large lighting system commands with multiple luminaire nodes, the delay in operating commands can be minimized. In particular, for a luminaire node 1 that is away from the data collector node 5 (or control center 6 〇) of the transmission operation and has a long transmission time, this minimization eliminates the retransmission between the unacknowledged data packets. Unnecessary communication delays. Therefore, the scalability of the lighting system can be further increased and the response time of the lamp node can be reduced. Another reason for the increase in point-to-point delay is to abandon the long-distance travel: package. Any data packet that has traveled long distances across the network has consumed: S network resources, and accumulated significant delays. If the data packet is discarded, it must be retransmitted by its sender node A (peer-to-peer retransmission). This retransmission doubles the accumulated point-to-point delay and network resources (also f7. bandwidth). In a multi-hop network, any node can act as a transmitter node or act as an intermediate Ni that forwards the received data packet to the next intermediate node Ni or to the final destination node B. Therefore, according to another embodiment of the present invention, the node 10 of the wireless network can not only maximize the response timeout of the data packet based on the distance to the destination node B when the sender node A is 2. And at node 10 operates as an intermediate node 159105.doc •24· 201218695
Ni時，節點10可調整用於處理待轉遞之資料封包之傳輸參 數。 在圖5中，展示說明用於待轉遞之資料封包之傳輸表數 的調整之流程圖。在步驟S50中’由中間節點沁接收來自 發送器節點A或來自另一中間節點Ni之資料封包。在已接 收到該資料封包之後，判定該接收中間節點Ni與發送器節 點A之間的距離（S51) »基於此距離’調整用於該資料封包 之傳輸參數。該等傳輸參數可指代MAC參數，例如， MAC層重新傳輸之最大數目' 頻道存取嘗試之最大數目、 傳輸功率位準、用於重新傳輸之延遲時間或用於頻道存取 之後移時間間隔，或其組合。MAC層重新傳輸之最大數目 判定允許MAC層重新嘗試資料封包之傳輸的次數。同樣 地’頻道存取嘗試之最大數目與允許MAC層執行載波感測 以便得到用於傳輸資料封包之頻道存取的最大允許次數有 關。用於重新傳輸之延遲時間表示資料封包之後續傳輸嘗 試之間的延遲，且用於頻道存取之後移時間間隔指代兩個 後續頻道存取嘗試之間的時間間隔。因此，可調整傳輸參 數’使得行進長距離之資料封包成功通過最後跳躍至其目 的地的機率增加。舉例而言，可增加MAC層重新傳輸之最 大數目或頻道存取嘗試之最大數目，或可降低用於重新傳 輸之延遲時間或用於頻道存取之後移時間間隔。當然，亦 可挑選此等調整之組合。 中間卽點Ni基於包括於資料封包中之跳躍計數器或發送 器位址而識職㈣包所行進之距離。在基於發送器位址 159105.docIn the case of Ni, node 10 can adjust the transmission parameters used to process the data packets to be forwarded. In Figure 5, a flow chart illustrating the adjustment of the number of transmission tables for a data packet to be forwarded is shown. The data packet from the sender node A or from another intermediate node Ni is received by the intermediate node 步骤 in step S50. After the data packet has been received, the distance between the receiving intermediate node Ni and the transmitter node A is determined (S51) » the transmission parameter for the data packet is adjusted based on the distance. The transmission parameters may refer to MAC parameters, such as the maximum number of MAC layer retransmissions, the maximum number of channel access attempts, the transmission power level, the delay time for retransmission, or the channel access time interval. , or a combination thereof. The maximum number of MAC layer retransmissions determines the number of times the MAC layer is allowed to retry the transmission of a data packet. Similarly, the maximum number of 'channel access attempts is related to allowing the MAC layer to perform carrier sensing in order to obtain the maximum number of allowed channel accesses for transmitting data packets. The delay time for retransmission represents the delay between subsequent transmission attempts of the data packet, and the channel access backward time interval refers to the time interval between two subsequent channel access attempts. Therefore, the transmission parameter can be adjusted so that the probability that the data packet traveling long distances successfully passes the last jump to its destination is increased. For example, the maximum number of MAC layer retransmissions or the maximum number of channel access attempts may be increased, or the delay time for retransmission or the channel access post shift time interval may be reduced. Of course, a combination of these adjustments can also be selected. The intermediate point Ni is based on the jump counter or the transmitter address included in the data packet to learn the distance traveled by the (4) packet. Based on the sender address 159105.doc
S •25· 201218695 之狀況下，發送器節點A之網路位址可為基於gps的。或 或另外τ將包含網路中之發送器節點八之Gps座標或 位置的路由表儲存於中間節酬中。行進長距離之資料封 包展現高的跳躍計數值或發送器節點A之GPS位置(例如， 在發送②位址欄位中指示）與中間節點州之GW位置（至少 為中間節點Ni本端已知)之間的大的差。 在一f例中，判定資料封包已行進之距離超過特定臨限 值°接著’可挑選用於資料封包至下-節點10之跳躍之優 先傳輸參數的集合。或者，該等傳輸參數之調整可盘由資 料封包行進之距離成比例。因此，行進長距離之資料封包 比行進較短距離之資料封包更有可能成功地跳躍至下一節 ㈣。此情形可甚至以行進短距離之資料封包為代價而發In the case of S • 25· 201218695, the network address of the sender node A can be gps-based. Or alternatively, τ stores a routing table containing the GPS coordinates or location of the transmitter node eight in the network in the intermediate pay. The long-distance data packet exhibits a high hop count value or the GPS position of the sender node A (for example, indicated in the transmission 2 address field) and the GW position of the intermediate node state (at least the intermediate node Ni is known locally The big difference between). In an example f, it is determined that the data packet has traveled a distance that exceeds a certain threshold value. Then, a set of priority transmission parameters for the data packet to the next-node 10 hop can be selected. Alternatively, the adjustment of the transmission parameters may be proportional to the distance traveled by the data packet. Therefore, data packets traveling over long distances are more likely to successfully jump to the next section (4) than data packets traveling a shorter distance. This situation can be exploited at the expense of a short-distance data packet.
在基於中間節點jsj i盘發样哭抓A 輸參數之後，使用二=離而調整傳 4傳輸參數來處理資料封包（S53)。 ㈣而言’此㈣可影響資料封包㈣之次序、優先權等 財f，使得可㈣地處理行進長距離之資料封包I 者，使用所判定之傳輸參數將資㈣ 10(S54) 〇 t王卜卽點 藉二等手：’行進長距離之資料封包之 加。此^貧料封包感知相當之延遲而不管跳躍之所需數 目，此情形對於應用層且亦對於傳送層係有益的。若在資 枓封包丟失之狀況下必須在較高層處重 則此機會增加尤其有用。此外，對於所有資料封=將 159105.doc -26- 201218695 答覆逾時及（詳言之）應答逾時設定為相等的且甚至較低。 因此，當基於資料封包之已行進距離而調整傳輸參數時’ 可另外考慮此情形以用於答覆逾時之調整。藉由此等手 段，可增加已使用大量網路頻寬之行進長距離之資料封包 的成功，且可能以行進較短距離之資料封包為代價而可減 小用於長路線之總的通信延遲。因此，進一步存在點對點 通信延遲之固有均勻性。 當應用於室外照明系統之電傳管理中時，此情形達成燈 具節點ίο之增加的同步，此係因為燈具節點1〇將以相當之 延遲接收指令。此外，遠離任何資料收集器節點5〇(或控 制中心60)之燈具節點10的均值及最大通信延遲減小，從 而導致同一網路中之燈具節點1〇之較高通信延遲均勻性。 因此，根據本發明，可針對每一對個別發送器節點A及 目的地節點B調整答覆逾時，從而降低無線網路中之點對 點傳輸延遲。此外’當節點職當轉遞所狀之資料封包 的中間節點Ni時，中間節點Ni可基於由此資料封包行進之 距離針對該資料封包而調整傳輸參數。此情形有助於進一 步減小點對點延遲且亦導致網路中之較高延遲均勻性。因 此，可節省總的網路資源。 【圖式簡單說明】 圖1展示無線網狀網路之實例； 圖2 A展示說明節點之間的跳躍距離之無線網狀網路的示 意圖； 圖2B展示說明節點之間的歐幾里德距離之無線網狀網路 159105.doc -27- 201218695 的示意圖； 圖2C說明在街道照明系統中的節點之空間分佈. 圖3展示根據本發明之裝置的示意圖； 圖4展示說明本發明之實施例的流程圖；及 圖5展示說明本發明之一另外實施例的流程圖。 【主要元件符號說明】 10 燈具節點 40 無線通信路徑 50 資料收集器節點 60 控制中心 70 連接 100 裝置 200 控制單元 300 收發單元 501 半徑 502 半徑 A 發送器節點 B 目的地節點 d 距離 hi 跳躍 h2 跳躍 N1 中間節點 159105.doc •28-After the A node input parameter is crying based on the intermediate node jsj i disk, the data packet is processed by using the second transmission parameter to adjust the transmission parameter (S53). (4) In the case of 'fourth', it may affect the order, priority, etc. of the data packet (4), so that the data packet of the long distance can be processed (4), and the transmission parameters determined by the use of the determined transmission parameters will be (4) 10 (S54) 〇t Wang Bu Yi points to the second-class hand: 'Advance of long-distance data packets. This poor packet senses a considerable delay regardless of the number of hops required, which is beneficial for the application layer and also for the transport layer. This opportunity increase is especially useful if it must be repeated at a higher level in the event that the asset packet is lost. In addition, for all data seals = 159105.doc -26- 201218695 Reply timeout and (detailed) response timeouts are set equal and even lower. Therefore, when the transmission parameters are adjusted based on the traveled distance of the data packet, this situation can be additionally considered for replying to the adjustment of the timeout. By this means, the success of data packets that have used a large amount of network bandwidth for long distances can be increased, and the total communication delay for long routes can be reduced at the expense of data packets traveling at shorter distances. . Therefore, there is further inherent uniformity of the point-to-point communication delay. When applied to telex management of an outdoor lighting system, this situation achieves an increased synchronization of the lamp node ίο, since the luminaire node 1 接收 will receive the command with considerable delay. In addition, the mean and maximum communication delay of the luminaire node 10 remote from any data collector node 5 (or control center 60) is reduced, resulting in higher communication delay uniformity for the luminaire nodes in the same network. Therefore, according to the present invention, the reply timeout can be adjusted for each pair of individual transmitter nodes A and destination nodes B, thereby reducing the point-to-point transmission delay in the wireless network. In addition, when the node serves to forward the intermediate node Ni of the data packet, the intermediate node Ni can adjust the transmission parameters for the data packet based on the distance traveled by the data packet. This situation helps to further reduce the point-to-point delay and also results in higher delay uniformity in the network. As a result, total network resources can be saved. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 shows an example of a wireless mesh network; FIG. 2A shows a schematic diagram of a wireless mesh network illustrating a jump distance between nodes; FIG. 2B shows a Euclidean distance between nodes. FIG. 2C illustrates a spatial distribution of nodes in a street lighting system. FIG. 3 shows a schematic diagram of a device in accordance with the present invention. FIG. 4 shows an embodiment of the present invention. Flowchart; and Figure 5 shows a flow chart illustrating an additional embodiment of the present invention. [Main component symbol description] 10 Lamp node 40 Wireless communication path 50 Data collector node 60 Control center 70 Connection 100 Device 200 Control unit 300 Transceiver unit 501 Radius 502 Radius A Transmitter node B Destination node d Distance hi Jump h2 Jump N1 Intermediate node 159105.doc •28-