201206116 六、發明說明： 【發明所屬之技術領域】 本發明有關於一種網路設備之網路頻寬上限統計及調 整方法，不需修改先前頻寬偵測技術，即可對於錯估的頻 寬上限值進行修正及調整。 【先前技術】 # 現今的網路設備，例如：寬頻路由器或無線網路寬頻 分享器，對於設備本身的網路流量、頻寬分配及限制功能 採用網路頻寬管理（Quallty 〇f Service ; Q〇s)及服務類型 (Type of Service’· T0S)機制，並利用優先權技術對於各 種不同協定的流量或是特定端點的流量進行頻寬的分配與 限制的設定。此外’這些網路設備之頻寬分配與限制功能 必須事先得知對外頻寬的上限值才能針對各種流量做比例 分配頻寬的動作。然而，現今許多頻寬债測的技術因為其 • 演算法或是網路佈建狀況的不同，多少都會對於實際頻寬 的偵測產生誤差的狀況，例如：高估或低估。 以往網路設備常使用一封包來回時間技術 TripTime; RTT)估算網路頻寬，對於特定端點（例如：DNS、 GATEWAY或NTP server)傳送封包，並藉由接收回應封包的 時間差進行數學運算以估算出頻寬。然而以上的技術與做 法，會因為網路設備與特定端點中的網路狀態以及路徑長 短的不同而導致誤差範圍大幅變動，例如：當網路擁塞時， 可能估算出較低的頻寬（2Mbps頻寬只估算出丨仙的頻 201206116 寬）’則QoS機制採用該較低的頻寬值（iMbps)進行頻寬的 分配與限制’將會造成使用者只能使用到lMbps頻寬的狀 況’進而造成頻寬使用上的浪費。反之，若估算出較高的 頻寬值（2Mbps頻寬卻估算出3Mbps頻寬）’ Q〇S進行頻寬分 配與限制時，假設給予高優先權分配2Mbps，而低優先權 給予分配1Mbps，則高優先權的流量頻寬2Mbps中有一 1Mbps將被低優先權所佔據，將使得高優先權在滿載傳輸 時最高只能有1Mbps流量頻寬。如此，將會造成實際頻寬 並;又有按照原先設定比例分配的狀況。 面對上述頻寬偵測所產生的誤差，先前的作法只能修 改頻寬偵測演算法的演算方式或是設計不同的偵測演算法 f量使偵測的誤差幅度變小來改善，但是無論再精準的演 算法或偵測方式仍會出無法容許的誤差範圍或是處在盔法 精確的偵測狀況與環境條件，而令頻寬偵測的結果缺= 確性。 【發明内容】 本發明之主要目的，在於提供一種網路設備之網路頻 寬上限統計及調整方法，不需修改先前已提出的頻寬侦測 技術，以將先前頻寬偵測技術所偵測出的網路頻寬值導入 於本發明統計及調整方法中’即可對於估算錯誤 限值進行修正及調整。 本發明之次要目的，在於提供一種網路設備之網路頻 寬上限統計及調整方法，其可應用於對内網路、對外網路、 4 201206116 有線網路或無線網路之上/下載的頻寬上限值進行調整 此以符合各種網路系統的需求。 曰 ▲為此，本發明提供一種網路設備之網路頻寬上限統計 及調整方法，其步驟包括：取得網路設備中現行的一頻寬 上限值；設定-取樣量及一接受比例值，以決定一接 檻值；提供至少-統計表，各統計表分別用以記錄有:取 樣次數及複數個索引值的統計次數，各索引值分別對應一 •頻寬階層，且各頻寬階層間差距一級距單位；進行一頻寬 上限統計流程，步驟包括：取得目前在網路設備中運行之 頻寬偵測技術所摘測出的一網路頻寬值；換算網路頻寬值 為索引值’以根據網路頻寬值所換算出的索引值對於統計 表:同等數值的索引值其統計次數往上加卜以及對於取 ，數在上加1，及比較取樣次數是否達到取樣量，若取 人數未達到取樣量，繼續頻寬上限統計流程，否則，進 行一頻寬^限調整流程；及進行頻寬上限調整流程，步驟 I =括換算頻寬上限值為索引值，以根據頻寬上限值所換 舁出的索引值從統計表中取出同等數值的索引值其記錄的 =計次數；及比較取出的統計次數是否大於接受門檻值， 若取出的統計次數大於接受門播值，再度進行下一次頻寬 ^限統計流程’否則’在統計表中判斷出最多統計次數的 '引ΐ，之後，比對最多統計次數的索引值及頻寬上限值 f換算出的索i值間之數值大小，以對於頻寬上限值所換 _ 的索引值往下減）、往上加i或維持現值並乘上級距 早位以计算出更新的頻寬上限值而取代目前的頻寬上限 5 201206116 次頻寬上限 值，並在更新頻寬上限值之後，再度進行下一 統計流程。 【實施方式】 請參閱第1圖及第2圖，為本發明網路設備之網路頻 寬上限統计及調整方法一較佳實施例之流程圖及本發 索引值之統計:欠數-實_之統計表。本發明網路設 網路頻寬上限統計及調整方法係可應用於—網路設備上， 並可利用先前已提出的頻寬偵測技術所偵測出的二頻寬 而進行統計，以調整出較精準的頻寬上限值， 行頻寬分配。 β』幻進 如第1圖所示，本發明所述之網路頻寬上限統 整方法，其主要步驟包括如下： °° 首先，步驟S111，在網路設備開始運作時，網路設 備中-頻寬統計調整常駐程式將會進行啟動，並取得ς 路設備令現行的一頻寬上限值（Upper Limit201206116 VI. Description of the Invention: [Technical Field] The present invention relates to a network bandwidth upper limit statistics and adjustment method for a network device, which can be used for a mis-estimated bandwidth without modifying the previous bandwidth detection technique. The upper limit is corrected and adjusted. [Prior Art] # Today's network devices, such as broadband routers or wireless network broadband sharing devices, use network bandwidth management for network traffic, bandwidth allocation and restriction functions of the device itself (Quallty 〇f Service; Q 〇s) and the Type of Service's (T0S) mechanism, and use the priority technology to set the bandwidth allocation and restriction for traffic of various protocols or traffic of a specific endpoint. In addition, the bandwidth allocation and limitation functions of these network devices must know the upper limit of the external bandwidth in advance to allocate bandwidth for various traffic. However, many of today's bandwidth-denominated technologies suffer from errors in the detection of actual bandwidth, such as overestimation or underestimation, because of their different algorithms or network deployment conditions. In the past, network devices often used a packet back and forth time technology TripTime; RTT) to estimate the network bandwidth, transmit packets for a specific endpoint (for example, DNS, GATEWAY or NTP server), and perform mathematical operations by receiving the time difference of the response packets. Estimate the bandwidth. However, the above techniques and practices may cause large variations in the error range due to the network state and the length of the path in the network device. For example, when the network is congested, a lower bandwidth may be estimated ( 2Mbps bandwidth is only estimated to be the frequency of 201206116 wide) 'The QoS mechanism uses the lower bandwidth value (iMbps) for bandwidth allocation and limitation' will cause the user to use only 1Mbps bandwidth. 'There is a waste of bandwidth usage. On the other hand, if a higher bandwidth value is estimated (2 Mbps bandwidth is estimated to be 3 Mbps bandwidth) 'Q 〇 S bandwidth allocation and limitation, it is assumed that a high priority allocation of 2 Mbps is given, and a low priority is given to allocate 1 Mbps. A high priority traffic bandwidth of 1 Mbps will be occupied by low priority, which will make the high priority only 1 Mbps traffic bandwidth at full load. In this way, the actual bandwidth will be caused; and the situation will be allocated according to the original set ratio. In the face of the above-mentioned error caused by bandwidth detection, the previous method can only modify the calculation method of the bandwidth detection algorithm or design different detection algorithms to reduce the error amplitude of the detection to improve, but No matter the precise algorithm or detection method, there is still an unacceptable error range or the accurate detection condition and environmental conditions of the helmet method, and the result of the bandwidth detection is lacking. SUMMARY OF THE INVENTION The main purpose of the present invention is to provide a network bandwidth upper limit statistics and adjustment method for a network device, without modifying the previously proposed bandwidth detection technology to detect the previous bandwidth detection technology. The measured network bandwidth value is introduced into the statistical and adjustment method of the present invention', and the estimated error limit can be corrected and adjusted. A secondary object of the present invention is to provide a network bandwidth upper limit statistics and adjustment method for a network device, which can be applied to an internal network, an external network, or on a 201206116 wired network or a wireless network. The upper bandwidth limit is adjusted to meet the needs of various network systems. For this reason, the present invention provides a network bandwidth upper limit statistics and adjustment method for a network device, the steps comprising: obtaining a current upper bandwidth upper limit value in the network device; setting-sampling amount and an acceptance ratio value To determine at least one value; provide at least - a statistical table, each of which is used to record: the number of times of sampling and the number of statistics of the plurality of index values, each index value corresponding to a bandwidth class, and each bandwidth class The first-order gap unit is used to perform a bandwidth upper-limit statistical process, and the steps include: obtaining a network bandwidth value measured by the bandwidth detection technology currently running in the network device; and converting the network bandwidth value The index value 'is an index value converted according to the network bandwidth value. For the statistical table: the index value of the same value is counted up by the number of statistics, and for the number of times, the number of samples is increased by 1 and the number of samples is compared. If the number of people does not reach the sampling amount, continue the bandwidth upper limit statistical process, otherwise, perform a bandwidth adjustment process; and perform a bandwidth upper limit adjustment process, step I = including the upper limit of the conversion bandwidth Value, the index value of the equivalent value is taken from the statistical table according to the index value exchanged by the upper limit value of the bandwidth, and the number of counts of the record is counted; and whether the counted number of times of the comparison is greater than the acceptance threshold value, if the number of statistics is taken out If the value is greater than the acceptance of the gatecast value, the next time the bandwidth is limited. The flow is 'other'. The maximum number of statistics is counted in the statistics table. After that, the index value and the upper bandwidth limit of the maximum number of statistics are compared. The value of the value between the converted values is reduced by the index value of _ for the upper limit of the bandwidth, plus i or the current value is maintained and multiplied by the previous level to calculate the updated bandwidth. The upper limit value replaces the current upper limit of bandwidth 5 201206116 times of the upper bandwidth, and after updating the upper limit of the bandwidth, the next statistical process is performed again. [Embodiment] Please refer to FIG. 1 and FIG. 2, which are flowcharts of a preferred embodiment of a network bandwidth upper limit statistics and adjustment method for a network device according to the present invention, and statistics of the present index value: Real _ statistics table. The network bandwidth upper limit statistics and adjustment method of the network of the invention can be applied to the network device, and can be adjusted by using the second bandwidth detected by the previously proposed bandwidth detection technology to adjust A more accurate bandwidth upper limit value and a line bandwidth allocation. As shown in FIG. 1 , the network bandwidth upper limit integration method of the present invention includes the following main steps: °° First, step S111, when the network device starts to operate, in the network device. - The bandwidth statistics adjustment resident program will start and obtain the current upper bandwidth limit of the circuit device (Upper Limit)
Ba^ndth ; ULBW) ’此頻寬上限值（此別）可由使用者手動 制定或者透過-先前已提出的頻寬❹m術自動制定。 步驟S112’設定一取樣的總數量s及一可接受的比例 值R (0 1間的浮點數），以決定一接受門檻值A=rxs。 接著，步驟S113，提供至少一統計表，各統計表分別 用=記錄-取樣讀之變數c，以及複數個索引值L的統 计-人數T〔 IN〕，各索引值In分別對應一頻寬階層，各頻寬 户白層間差距一級距單位，本實施例係以128 Kbps作為頻 201206116 寬階層的級距單位，則㈣值Ιβ對應於QKbps，索引值 Ιι對應於128Kbps ’索引值l對應於256Kfc>ps，索引值 13對應於職bps·..⑽層方式依此類推）。當然，本發明 另：實施财係可以不同單位數值來作為頻t階層的級 距，位，例如：32KbpS、64Kbps、256Kbps、512Kbps 或任 何單位數值。之後，進入一頻寬上限統計流程12〇。 本發明所述之頻寬上限統計流程12〇，其流程如下 所述.步驟S121，取樣得到一網路頻寬值（Netw〇rk Bandwidth ; NBW)，取得該網路頻寬值（NBW)的方法係可使 用目前在網路設備中所運行的一般頻寬偵測技術，例如·· 封包來回時間RTT技術（封包容量大小/時間 差）、流量統計技術（收集封包流量/收集時間）或 者其他先前已提出的頻寬偵測技術。 步驟S122’網路頻寬值（NBw)根據一第一頻寬級距方 程式換算為一對應的索引值h，其方程式如下所示：Ba^ndth ; ULBW) 'This bandwidth upper limit (this) can be set manually by the user or automatically by the previously proposed bandwidth ❹m technique. Step S112' sets a total number of samples s and an acceptable ratio value R (a floating point number between 0 1) to determine an acceptance threshold A = rxs. Next, in step S113, at least one statistical table is provided. Each statistical table uses a variable of the record-sampling read c, and a statistical number-number of the index values L (IN), and each index value In corresponds to a bandwidth. The hierarchy is the first-order distance unit between the white-bands of each bandwidth. In this embodiment, 128 Kbps is used as the span unit of the frequency 201206116, and the (four) value Ιβ corresponds to QKbps, and the index value Ιι corresponds to 128Kbps. The index value l corresponds to 256Kfc> ps, index value 13 corresponds to the job bps ·.. (10) layer mode and so on). Of course, the present invention is another embodiment: the financial system can use different unit values as the level of the frequency t level, such as: 32 KbpS, 64 Kbps, 256 Kbps, 512 Kbps or any unit value. After that, enter a bandwidth upper limit statistical flow 12〇. The bandwidth upper limit statistical process 12〇 of the present invention has the following flow. In step S121, a network bandwidth value (NetW〇rk Bandwidth; NBW) is obtained, and the network bandwidth value (NBW) is obtained. The method can use the general bandwidth detection technology currently running in the network device, such as · packet round-trip time RTT technology (packet capacity size / time difference), traffic statistics technology (collect packet flow / collection time) or other previous A bandwidth detection technique has been proposed. Step S122' The network bandwidth value (NBw) is converted into a corresponding index value h according to a first bandwidth step formula, and the equation is as follows:
Ia=(NBW + mod A ⑴ 其中，Ια為換算網路頻寬值（NBW)所求得的索引值，NBW為 網路頻寬值，mod為除法取商數運算函數，a為級距單位（如 128Kbps)。藉由網路頻寬值（NBW)所換算出的索引值ιΑ以 對於統計表中同等數值的索引值IN其統計次數T〔 ιΝ〕往 上加1，例如：偵測出網路頻寬值（NBW)為963 Kbps，則可 透過方程式（1)計算出索引值IA=(963 + 64) mod 128 = 8, 對於第2圖中之統計表IN=8的統計次數τ〔 I〇往上加1， 201206116 再者，同時對於網路頻寬值（NBW)的取樣次數c也會往上 加 1(C=C+1)。 接續，步驟S123，若目前網路頻寬值（NBW)的取樣次 數C未達到取樣量S，則繼續頻寬上限統計流程12〇，直 到取樣次數C已相同於取樣量S，後續才進行一頻寬上限 調整流程130。 當頻寬上限統計流程120對於網路頻寬值（NBW)收 集到足夠的取樣筆數後’進入本發明所述之頻寬上限調 整流程130 ’其流程如下所述：步驟S131，頻寬上限值 (ULBW)根據一第二頻寬級距方程式換算為一對應的索引值 Ιβ ’其方程式（2)如下所示： (2)Ia=(NBW + mod A (1) where Ια is the index value obtained by converting the network bandwidth value (NBW), NBW is the network bandwidth value, mod is the division quotient operation function, and a is the step size unit. (eg 128Kbps). The index value ι converted by the network bandwidth value (NBW) is incremented by one for the index value IN of the equivalent value in the statistical table, for example, the detection is performed. If the network bandwidth value (NBW) is 963 Kbps, the index value IA=(963 + 64) mod 128 = 8 can be calculated by equation (1). For the statistical number of the statistical table IN=8 in Fig. 2, τ [ I〇1, 201206116 Furthermore, the number of samples c for the network bandwidth value (NBW) will be increased by 1 (C=C+1). Continue, step S123, if the current network frequency If the sampling number C of the wide value (NBW) does not reach the sampling amount S, the bandwidth upper limit statistical flow is continued until the sampling number C is the same as the sampling amount S, and then a bandwidth upper limit adjustment process 130 is performed. The upper limit statistics process 120 enters the bandwidth upper limit adjustment process 130 of the present invention after collecting enough number of samples for the network bandwidth value (NBW). The process is as follows: Step S131, the bandwidth upper limit value (ULBW) from the equation in terms of an index value corresponding to a second bandwidth in accordance Ιβ stage 'which equation (2) shown below: (2)
,=〇JLBW)mod A 其中，Ib為換算頻寬上限值（ULBW)所求得的索引值，讥训 為頻寬上限值’ m〇d為除法取商數運算函數，A為級距單位 (如128KbPS)。藉由頻寬上限值（ULBW)所換算出的索引值 h從統計表中取出同等數值的索引值h其記錄的統計次數 τ〔 In〕’例如：當頻寬上限值（ULBW)制定為896肋卵時， 換异出的索引值Ιβ=7’則從第2圖之統計表中取出In=7 的統計次數了〔17〕=7;當頻寬上限值（咖）敎為刪 bPS時’換算出的*引值Ιβ=8，則從第2圖之統計表中 制統計次數T〔 M = 81 ;當頻寬上限值(_) 之統；#由時’換算出的索引值Ιβ=9，則從第2圖 先什表中取出“的統計次數T〔l9〕=4。 201206116 步驟132，統計次數τ〔 In〕取出之後，比對統計次數 T〔 In〕是否大於接受門檻值a。若統計次數τ〔 In〕大於 接文門檻值A’再度進行下一次頻寬上限統計流程ι2〇，以 進行下一週期之頻寬統計。反之，比對統計次數T〔 IN〕小 於接受門檻值A，則頻寬上限值（ULBW)有可能制定或估計 =誤，而令頻寬上限值（ULBW)的索引值h在統計表中非 系態分佈，後續，將對於現行的頻寬上限值（ULBW)進行調 秦整的相關步驟之後，才會再度進行頻寬上限統計流程120。 係以一範例進行解說，若取樣的總數量S設定為1〇〇 次取樣次數及接受的比例值R為〇 8，則接收門檻值A將 可決定為80次’當頻寬上限值（ULBW)換算為索引值h=8, 且在第2圖之統計表中對應取出索引值h=8的統計次數 T〔 Is〕-81將可超過接受門檻值A，則目前所估算或制定 的頻寬上限值（ULBW)係為一較準確的上限值，不需調整， 即了重返頻寬上限統計流程12〇 ;相反的，當頻寬上限值 φ (ULBW)換算出的索引值b不等於8而為其他值時，則在 第2圖之統计表中所對應取出的索引值丨N的統計次數T 〔U〕必定小於接受門檻值A，則目前所估算或制定的頻寬 上限值（ULBW)亦可能為高估、低估或錯估的上限值，必須 進一步調整頻寬上限值（ULBW)，並在調整完之後，才重返 頻寬上限統計流程120。 承上，步驟133’當取出的索引值In的統計次數T〔 h〕 小於接受門播值A的情況下’將從第2圖之統計表中判斷 出最多統計次數的索引值IN(如本實施例最多取樣次數的 9 201206116 索引值為I pj=8 ’其統計次數τ〔 18〕=81)。 步驟134，比對最多統計次數τ〔 iN〕的索引值iN及該 頻寬上限值（ULBW)所換算出的索引值Ib間之數值大小。若 最多統計次數T〔L·〕的索引值In小於索引值Ιβ，則18往 下減ι(ιΒ=ιΒ-ι)，並乘上級距單位（如128Kbps)，以取代 現行的頻寬上限值（ULBW=(IB-i)xl28Kbps)，換言之，現 行的頻寬上限值（ULBW)有高估的情況，必須調降一級頻寬 階層。若最多統計次數T〔 IN〕的索引值In大於索引值Ιβ， 則Ib往上加1(IB=IB+1)，並乘上級距單位（如128Kbps)， 以取代現行的頻寬上限值（ULBW=(IB+l)xl 28Kbps)，換言 之，現行的頻寬上限值（ULBW)有低估的情況，必須調升一 級頻寬階層。當然，在步驟134中，頻寬品質的不穩亦可 能造成頻寬上限值（ULBW)錯估的情況發生，而令最多統計 次數T〔 h〕的索引值In雖等於索引值Ib，但，最多統計 次數Τ〔 Μ仍小於所設定的接受門檻值a，此時，暫且不 而對於頻寬上限值（ULBW)進行調整，以避免對於頻寬不穩 的網路設備進行頻寬上限值（ULBW)的調整。 在進行完步驟134之後，再度進行下一週期頻寬上限 統計流程120及頻寬上限流程13〇,繼續對於網路設備所 偵測的頻寬值作統計，且週期性地對於網路設備之頻寬上 限值（ULBW)進行調整。 又’本發明實施例所述之取樣總數量S係以一百次取 樣次數為例’然，為了令取樣的網路頻寬值（NBW)集中度越 冋，亦可進一步設定為一千次 '一千次、一萬次或其他更 201206116 南數罝的取樣次數，而比例值R隨著取樣的精確度要求提 南比例比例值R设的越高並接近1，其取樣的精確度要 求也會相對提高。,=〇JLBW)mod A where Ib is the index value obtained by the upper limit of the conversion bandwidth (ULBW), and the upper limit of the bandwidth is 'm〇d' is the division function of the division function, A is the level Distance unit (such as 128KbPS). The index value h of the equivalent value is taken from the statistical table by the index value h converted from the upper bandwidth limit value (ULBW), and the statistical number of records τ[In]' is recorded, for example, when the upper bandwidth limit (ULBW) is established. For the 896 rib egg, the index value of the change Ιβ=7' is taken from the statistical table of Fig. 2, and the number of statistics of In=7 is [17]=7; when the upper limit of the bandwidth (coffee) is When bPS is deleted, the converted *index Ιβ=8, the statistical number T [M = 81; from the upper limit of the bandwidth (_); when converted from time to time] is calculated from the statistical table in Fig. 2 If the index value Ιβ=9, the "statistical number T[l9]=4 is taken from the table in the second figure. 201206116 Step 132, after the statistical number τ[In] is taken out, compare the statistical number T[In] It is greater than the acceptance threshold value a. If the statistical number τ[In] is greater than the access threshold A', the next upper bandwidth upper limit statistical process ι2〇 is performed to perform the bandwidth statistics of the next cycle. Conversely, the comparison statistics T[ If IN] is less than the acceptance threshold A, the upper bandwidth limit (ULBW) may be formulated or estimated = error, and the index value h of the upper bandwidth limit (ULBW) is in the system. The non-systematic distribution in the meter will be followed by the relevant step of adjusting the current bandwidth upper limit value (ULBW), and then the bandwidth upper limit statistical process 120 will be performed again. When the total number of samples S is set to 1 number of sampling times and the accepted ratio value R is 〇8, the receiving threshold value A can be determined as 80 times. When the upper bandwidth limit value (ULBW) is converted into an index value h= 8, and in the statistical table of Fig. 2, the statistical number T[Is]-81 corresponding to the index value h=8 will exceed the acceptance threshold A, and the currently estimated or established bandwidth upper limit value (ULBW) The system is a more accurate upper limit value, and no adjustment is needed, that is, the re-return bandwidth upper limit statistical flow is 12〇; conversely, when the bandwidth upper limit value φ (ULBW) is converted, the index value b is not equal to 8 For other values, the statistical number T [U] of the index value 丨N taken in the statistical table of Fig. 2 must be less than the acceptance threshold A, and the current upper limit value (ULBW) is estimated or established. ) may also be the upper limit of overestimation, underestimation or miscalculation, and the upper bandwidth limit (ULBW) must be further adjusted, and After the adjustment is completed, the bandwidth upper limit statistical process 120 is returned. In the step 133', when the statistical value T[h] of the extracted index value In is smaller than the acceptance of the gated value A, the statistics from the second figure will be The index value IN of the maximum number of statistics is judged in the table (as in the embodiment, the maximum number of samples is 9 201206116, and the index value is I pj=8 ', and the number of statistics is τ [18]=81). Step 134, compare the maximum number of statistics τ The numerical value between the index value iN of [iN] and the index value Ib converted by the bandwidth upper limit value (ULBW). If the index value In of the maximum number of statistics T[L·] is smaller than the index value Ιβ, then 18 is reduced by ι(ιΒ=ιΒ-ι) and multiplied by the step size unit (such as 128Kbps) to replace the current bandwidth limit. The value (ULBW = (IB-i) xl28Kbps), in other words, the current upper bandwidth limit (ULBW) is overestimated, and the first-order bandwidth level must be lowered. If the index value In of the most statistical number T[IN] is greater than the index value Ιβ, then Ib is incremented by 1 (IB=IB+1) and multiplied by the step size unit (such as 128Kbps) to replace the current bandwidth upper limit value. (ULBW = (IB + l) xl 28 Kbps), in other words, the current upper bandwidth limit (ULBW) is underestimated, and the first-order bandwidth level must be raised. Of course, in step 134, the instability of the bandwidth quality may also cause a miscalculation of the upper bandwidth limit (ULBW), and the index value In of the most statistical number T[h] is equal to the index value Ib, but The maximum number of statistics Τ[ Μ is still less than the set acceptance threshold a. At this time, the upper bandwidth limit (ULBW) is not adjusted to avoid bandwidth fluctuation for network devices with unstable bandwidth. Adjustment of the limit value (ULBW). After the step 134 is performed, the next cycle bandwidth upper limit statistics process 120 and the bandwidth upper limit process 13 are performed again, and the bandwidth value detected by the network device is continuously counted, and periodically for the network device. The upper bandwidth limit (ULBW) is adjusted. In addition, the total number S of samplings in the embodiment of the present invention is taken as an example of the number of sampling times of one hundred. However, in order to make the concentration of the sampled network bandwidth (NBW) more ambiguous, it may be further set to one thousand times. '1000 times, 10,000 times or other more 201206116 South number 取样 sampling times, and the proportional value R with the accuracy of the sampling requirements, the higher the ratio of the ratio R is set to be close to 1, the precision of the sampling requirements It will also increase relatively.
再者，本發明頻寬上限統計及調整方法可對於對外、 =内網路線路之網路設備所包括的各類有線網路 面（例如.乙太網路界面）或是無線網路介面（例 如802. lla、8〇2. lib/g)之上/下傳時的網路頻寬值(NBW) 進订偵測及統計，再者，頻寬上限值（ULBw)係可為一頻寬 傳 9 上限值（Upiink Upper Limit Bandwidth; UULBW) 或一頻寬下載的上限值（D〇wnlink Upper LimitFurthermore, the bandwidth upper limit statistics and adjustment method of the present invention can be used for various wired network planes (for example, Ethernet interfaces) or wireless network interfaces included in network devices of external and internal network lines (for example, 802.lla, 8〇2. lib/g) Network bandwidth value (NBW) subscription detection and statistics, in addition, the bandwidth upper limit (ULBw) can be a frequency Upiink Upper Limit Bandwidth (UULBW) or the upper limit of a bandwidth download (D〇wnlink Upper Limit)
Bandwidth ; DULBW)。 ❹Ϊΐίΐ’本發明利用統計原理對於先前各種頻寬债 錄，:ί法每次偵測估計的網路頻寬進行相對的記 Γ借由^由每、▲ 4週期的循環分析比較，調整更新網路 ㈣寬值’改善各類頻寬债測技術的誤差， 值“佈上較準確的網路設備實體線路頻寬上限 以上所述者 第 【圖式簡單說明】 1圖：為本發明網路設備之頻寬上 限統計及調整方法一 201206116 較佳實施例之流程圖。 第2圖：為本發明各索引值之統計次數一實施例之統計表。 【主要元件符號說明】 120 頻寬上限統計流程 130 頻寬上限調整流程Bandwidth; DULBW). ❹Ϊΐίΐ' The invention utilizes the statistical principle for the previous various bandwidth debts: ί method to detect the estimated network bandwidth each time relative to the record by the ^ _ 4 cycle analysis and comparison, adjust the update network Road (four) wide value 'improves the error of various bandwidth debt measurement technology, the value is "the more accurate network device physical line bandwidth upper limit above." [Simplified description] 1 picture: the network of the present invention The upper limit of the bandwidth of the device and the adjustment method of the device is as follows: Figure 2 is a flow chart of the preferred embodiment of the present invention. Fig. 2 is a statistical table of an embodiment of the statistical value of each index value of the present invention. [Description of main component symbols] 120 upper limit of bandwidth Statistical Process 130 Bandwidth Upper Limit Adjustment Process
1212