JP3812839B2 - Wireless mesh network communication method and system - Google Patents

Description

あるいは
【数２】

但し、
【数３】

が小さいように有向化されたグラフと考えるのが妥当と考えられる。ここで、指標２はネットワーク内のハイブリッドノードの総数に相当し、指標１はハイブリッドノードにおいて、これを上位に位置付けるリンクの数と下位に位置付けるリンクの数との積の総和である。この総和が多ければ、ハイブリッドノードにより終端されるリンク数が多く、このリンク数が多ければトラヒック量の合計も多く、周波数繰り返し利用の要請も高いと想定するのが自然と考え、また、このハイブリッドノードをピュアノード化するために切り替えなければならないリンク数も多いとの観点から、ノード数に重みをつけたものと位置付けられる。
【００３６】
図６，７，８，９に示した具体例を参照して説明する。図６，７は、指標１が同じで指標２が異なる簡単な例であり、ハイブリッドノード数が、図６では「１」であるのに対して図７では「２」なので、図６が図７よりも好ましい。
【００３７】
一方、図８，９は、指標１と指標２の大小関係が異なる簡単な例である。図８では指標２が「１」であり、図９の「２」に比較して小さいが、指標１に関しては、図８の「４」に対して図９は「２」と小さい。図８，９の例では、中心のノードがハブ的な役割を果たしており、このノードにおける周波数利用効率を高めることを優先させるべきとみなせるので、図８を図９よりも好ましいとするのが妥当である。
【００３８】
これらを踏まえて、性能指標として指標１を第１優先度とし、これが等しい条件では、指標２を第２優先度として用いるのが適切と考えられる。
【００３９】
以上の検討結果から、指標１、指標２を最小とするように無向グラフを有向化するアルゴリズムを検討した。しらみつぶし的に探索した場合には、組み合わせ数が「辺の数（グラフのSize）−１」のかい乗（n！）となるので、辺の数が１０程度を超えた途端に計算時間が爆発的に増加してしまい、現実的でない。そこで、本実施形態では次のような方針に基づき、ヒューリスティックなアルゴリズムを開発した。
【００４０】
方針：奇数辺の閉路がある場合には、幾つかの辺を除去して、これを取り除き、まず得られたサブグラフを（指標は０のままで）有向化する。取り除いた辺について、指標の増加が小さい向きを選びつつ、追加していく。
【００４１】
具体的なアルゴリズムの骨子は次の通りである。
▲１▼ステップ１：基準となる点を１つ選ぶ。選んだ点から他の点への最短到達ホップ数を計算しておく。
▲２▼ステップＳ２：グラフの辺全てについて、その両端の点における上記ホップ数をチェックし、一方が偶数で、他方が奇数であれば、その辺を偶数の点から奇数の点に向かうように有向化する。もし、両方が偶数または奇数の場合には、その辺はグラフから一時取り除いておく。
▲３▼ステップ３：前記ステップＳ１，２を、基準点のループとして点の数だけ繰り返し、得られた結果の中で、取り除いた辺の数が最小のものを候補として選択する。同点の場合には、取り除いた辺を要素とするサブグラフの木の数が最大のものを選択する。（これも同点の場合には、さらなる基準を設けることなく、最初に見つかったものを候補とする）
▲４▼ステップ４：取り除いた辺からなるサブグラフに含まれる木をひとつ選び、これを有向化して、すでに有向化されたグラフに追加する。向きのつけ方は、木の辺数をStとして２^Stだけ存在するが、追加後の目的関数が最小となるものを選ぶ（全数探索）。
▲５▼ステップＳ５：前記ステップＳ３で取り除かれていた辺がすべて追加されるまで前記ステップＳ４を繰り返す。これにより、デフォルトTDDプランが生成される。
【００４２】
図２に戻り、ステップＳ３では、前記デフォルトTDDプランで規定されたTDD位置付け属性と自ノードの現在のTDD位置付け属性とが異なるときに、自ノードのTDD位置付け属性を前記デフォルトTDDプランに合わせて、上位および下位の一方から他方へ切り替える「デフォルトTDDプランの反映処理」が実行される。
【００４３】
図３は、前記「デフォルトTDDプランの反映処理」の手順を示したフローチャートであり、ステップＳ１０では、自ノードで終端される各無線リンクL1，L2，L3，L4に関する現在のTDD位置付け属性が、前記デフォルトTDDプランで規定されたTDD位置付け属性と比較され、TDD位置付け属性の切り替えが必要な無線リンクの有無が判別される。ここでは、デフォルトTDDプランで規定されたTDD位置付け属性と異なる属性の無線リンクが全て切替対象リンクと認識される。切替対象となる無線リンクが存在すればステップＳ１１へ進む。
【００４４】
ステップＳ１１では、切替対象リンクに関して、その回線品質が求められる。本実施形態では、回線品質としてCNR｛C：キャリア、N：ノイズ、R：レシオ、すなわちC/N｝を採用し、これが基準値CNRrefと比較される。CNR＞CNRrefであればステップＳ１２へ進み、TDD位置付け属性を切り替える無線リンクが選択される。なお、対象リンクが複数存在する場合には、そのうちの一つがランダムに選択される。
【００４５】
ステップＳ１３では、選択された切替対象リンクが所定の切替条件を満足しているか否かが判別される。本実施形態では、切替対象リンクと隣接し、当該切替対象リンクとTDD位置付け属性が同一の隣接リンクにおいて、当該切替対象リンクで利用されているFIDと同一のFIDが使用されていないことが切替条件として設定されている。同一FIDが使用されていれば、TDD位置付け属性の切り替え後に干渉が生じてしまうので、当該切替対象リンクに関しては、TDD位置付け属性の切り替えを中止する。これに対して、同一FIDが使用されていなければ、切替条件が満足されたものとしてステップＳ１４へ進む。
【００４６】
ステップＳ１４では、TDD位置付け属性の切替指示および切替タイミングが対向ノードへ通知される。ステップＳ１５では、今回の切り替え対象リンクのTDD位置付け属性が切り替えられ、当該無線リンクに含まれる全ての無線チャネルに関して、そのTDD位置付け属性が、上位から下位、または下位からた上位へと切り替えられる。
【００４７】
図２へ戻り、ステップＳ４では、ネットワーク制御装置３４から通知された要求チャネル数が判別される。前記ネットワーク制御装置３４は、ネットワークトポロジやトラヒック量に基づいて、一端が自ノードで終端される各無線リンクL1，L2，L3，L4に適正なチャネル数を定期的または不定期に演算し、演算結果を要求チャネル数Lreq(Lreq(1)〜Lreq(4))として無線チャネルマネージャ３２に通知する。
【００４８】
ステップＳ５では、通知された要求チャネル数Lreqと現在の無線チャネル数Lrefとが比較され、Lreq＞Lrefの無線リンクに関してはステップＳ６へ進む。ステップＳ６では、「チャネル追加の前処理」が実行され、追加する無線チャネルに割り当てるFIDが決定され、さらには、必要に応じてTDD位置付け属性の切り替えが実行される。
【００４９】
図４は、「チャネル追加の前処理」の手順を示したフローチャートであり、ステップＳ３１では、いずれの無線チャネルにも割り当てられていない未割当FIDの中に、割当条件を満足するFIDが存在するか否かが判別される。本実施形態では、割当条件として以下の３つを設定している。
【００５０】
(1)条件１：TDD位置付け属性の異なる他の無線リンクに割り当てられていないこと。
【００５１】
(2)条件２：ある一定期間連続して所定の回線品質を満足すること
【００５２】
(3)条件３：互いに隣接リンクとならないリンク同士（つまり、互いに異なるノードに終端されるリンク同士）がTDDバウンダリを一致させなければならなくなるようなFIDは選択できない。
【００５３】
すなわち、図１２に示したように、ノードN1，N2間にFIDがF1の無線リンクL1が確立され、ノードN1で終端される他の無線リンクL2のFIDがF1のとき、リンクL1，L2のTDDバウンダリは、相互の干渉を回避すべく同一に設定される。この状態で、ノードN2で終端される他の無線リンクL3にチャネルを追加する際、そのFIDをF1に設定してしまうと、ノードL3とノードL1との干渉を避けるためには両者のTDDバウンダリを一致させなければ成らず、これがノードL2にも波及するので、ノードL3とノードL2とは隣接リンクではないにもかかわらずTDDバウンダリを一致させなければならなくなる。
【００５４】
同様に、無線リンクL2，L3は隣接しないので、両者が同一のFID（例えば、F1）を利用していてもTDDバウンダリを一致させる必要はない。しかしながら、このような状態で無線リンクL1にチャネルを追加する際、そのFIDをF1に設定してしまうと、無線リンクL1に隣接する無線リンクL2，L3は、そのTDDバウンダリを無線リンクL1のそれと一致させなければならないので、結局、ノードL3とノードL2とは隣接リンクではないにもかかわらずTDDバウンダリを一致させなければならなくなる。
【００５５】
このように、複数のノードをまたいでTDDバウンダリを一致させるための制御を行おうとすれば、TDDバウンダリを決定するために必要なトラヒックの情報の伝達が煩雑となる。本実施形態では、このような事態を回避すべく、互いに隣接リンクとならないリンク同士（図１２では、リンクL2，L3）がTDDバウンダリを一致させなければならなくなるようなFIDは選択できないことが第３の条件となる。そして、この割当条件を満足するFIDが存在すればステップＳ３２へ進み、その中の一つが選択される。
【００５６】
これに対して、割当条件を満足するFIDが存在しなければステップ３３へ進み、未割当FIDのいずれかが任意に選択される。ステップＳ３４では、TDD位置付け属性を切り替えても割当条件を満足できるFIDであるか否かが判定され、満足されるのであればステップＳ３５へ進む。ステップＳ３５では、無線チャネルを追加しようとしている無線リンクのTDD位置付け属性を切り替えても、既設の無線チャネルで利用中のFIDが全て割当条件を満足できるか否かが判定される。満足できるのであればステップＳ３６へ進み、無線チャネルが追加される無線リンクのTDD位置付け属性が切り替えられ、ここに無線チャネルが追加される。
【００５７】
一方、前記ステップＳ３４において、TDD位置付け属性を切り替えても割当条件が満足されないと判定されるか、あるいはステップＳ３５において、TDD位置付け属性を切り替えてしまうと既設の無線チャネルで利用中のFIDが割当条件を満足できなくなると判定されるとステップＳ３７へ進む。ステップＳ３７では、他の未割当FIDが存在するか否かが判定され、存在すればステップＳ３３へ戻って上記各処理が他のFIDに関して繰り返される。他の未割当FIDが存在しなければそのまま終了する。
【００５８】
図２へ戻り、以上のようにして追加するチャネルに割り当てるFIDが決定されると、ステップＳ７では、対向ノードとTDDバウンダリやフレーム同期に関するネゴシエーションが行われ、その後、所定のタイミングで無線チャネルが追加される。
【００５９】
図５は、前記TDD位置付け属性の切り替えによってFIDの利用効率が向上する様子を模式的に表現した図であり、ここでは、システムで利用可能なFIDの集合を「Fn」で表現し、ハイブリッドノードを上位に位置付ける無線リンクのみで利用可能なFIDの集合を「Fa」で表現し、ハイブリッドノードを下位に位置付けるリンクのみで利用可能なFIDの集合を「Fn-a」で表現している。したがって、「Fn」、「Fa」、「Fn-a」は、「Fn＝Fa＋Fn-a」の関係を有する。
【００６０】
同図左側に示したように、ノードN1，N4が上位ノード（●）、ノードN2，N5が下位ノード（○）、ノードN3がハイブリッドノードの状態では、ノードN1−N3間の無線リンクL1で利用可能なFIDの集合はFn-aであり、ノードN2−N3間の無線リンクL2およびN3−N5間の無線リンクL3で利用可能なFIDの集合はFaである。したがって、これらの無線リンクで利用可能なFIDには制約が多く、無線チャネルの追加が難しい。
【００６１】
この状態でリンクL1，L3に対してチャネル追加が要求されると、ノードN3では、リンクL2，L3に関してTDD位置付け属性の切り替えが実行され、その結果、同図右側に示したように、ノードN3がハイブリッドノードから下位ノードに切り替わり、これに応答して、ノードN5が下位ノードから上位ノードに切り替わり、ノードN2が下位ノードからハイブリッドノードに切り替わる。
【００６２】
この結果、トラヒック需要の増加した無線リンクL1，L3で利用可能なFIDの集合が、いずれもFnとなるので、トラヒック需要の増大に応じて無線チャネルを追加できるようになる。
【００６３】
【発明の効果】
本発明によれば、以下のような効果が達成される。
(1)各ノードは各無線リンクに対する自ノードの位置付け（TDD位置付け属性）を上位および下位の一方から他方へ切り替えることができるので、ハイブリッドノードからピュアノードへの移行およびその逆が可能になる。
(2)各ノードは自ノードがハイブリッドノードの時にトラヒック需要が増大し、FID不足により無線チャネルを追加できない場合に、自ノードのTDD位置付け属性を切り替えることでピュアノードとなれる。そして、ピュアノードではハイブリッドノードよりも割当可能なFIDが増えるので、トラヒック需要の増大に応じて無線チャネルを追加できるようになる。
(3)各ノードは自ノードのTDD位置付け属性をデフォルトTDDプランで規定されたTDD位置付け属性に戻すことができるので、TDD位置付け属性の切り替えが頻発しない環境を簡単に再現できる。
【図面の簡単な説明】
【図１】 各無線ノードに搭載される無線通信システムの主要部の構成を示したブロック図である。
【図２】 各無線ノードの動作を示したフローチャートである。
【図３】 デフォルトTDDプラン反映処理のフローチャートである。
【図４】 チャネル追加前処理のフローチャートである。
【図５】 DD位置付け属性の切り替えによってFIDの利用効率が向上する様子を模式的に表現した図である。
【図６】 デフォルトTDDプランの作成方法を説明するための図（その１）である。
【図７】 デフォルトTDDプランの作成方法を説明するための図（その２）である。
【図８】 デフォルトTDDプランの作成方法を説明するための図（その３）である。
【図９】 デフォルトTDDプランの作成方法を説明するための図（その４）である。
【図１０】 奇数多角形の閉路を含むネットワークの構成を模式的に示した図である。
【図１１】 偶数多角形の閉路のみを含むネットワークの構成を模式的に示した図である。
【図１２】 割当条件を説明するための図である。
【符号の説明】
２…屋外無線装置，３…屋内無線装置，３１…スイッチ，３２…無線チャネルマネージャ，３３…無線リンク制御装置，３４…ネットワーク制御装置[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a communication method and system in a mesh network, and in particular, a pair of radio nodes that terminate each radio link is exclusively positioned in either the upper or lower level, and is positioned in one of the upper and lower levels. The present invention relates to a communication method and system in which a wireless node autonomously establishes a TDD wireless channel on the wireless link.
[0002]
[Prior art]
In wireless networks where the link topology is mesh, adoption of the TDD (Time Division Duplex) method is being considered as a communication method that can effectively allocate communication resources even if the traffic volume between opposing nodes is asymmetric. Yes. The TDD method is an access method that realizes mutual communication by time-sharing a time axis with a constant period (time slot) in one frequency band (frequency slot) and assigning it to the uplink and downlink. Generally used in multiplexing systems such as (Time Division Multiple Access) and CDMA (Code Division Multiple Access).
[0003]
In such a TDD scheme, a frequency slot (FID) to be allocated to each radio channel is determined, and a time slot allocation boundary (TDD Boundary: TDD boundary) is set in accordance with uplink and downlink traffic volume. Settings related to various TDD attributes are required.
[0004]
Here, one control node on the mesh network grasps the traffic information and network topology of all nodes in the mesh network, determines the TDD attribute and FID for each link, and distributes them to all nodes. This is unrealistic in terms of increasing the load on the control node, the ability to follow instantaneously changing traffic volume, or the scalability of the network.
[0005]
Therefore, it is considered effective to use an autonomous distributed method in which each node determines its own TDD attribute and FID while obtaining neighboring information instead of central control by a specific control node. However, a radio communication system that employs the TDD scheme and uses a plurality of FIDs has the following technical problems.
[0006]
(1) In a mesh network, one node may have a plurality of radio links, and a plurality of antennas and radio stations are installed in each node. At this time, if the same FID is used for a plurality of links and the TDD boundary is different for each link, interference due to wraparound from an adjacent antenna of the same node may occur. Therefore, it is necessary to share a TDD boundary between adjacent links using the same FID, and an optimum TDD boundary cannot be set for each link according to the traffic volume.
[0007]
(2) When updating the FID assigned to a radio link, if a pair of radio nodes sharing this radio link do not have a common knowledge about their TDD boundary or FID, each node has a different link to the same link. A FID will be assigned, and a trap will be generated, making communication impossible.
[0008]
In order to solve such a technical problem, the inventors of the present invention, as an allocation attribute related to a TDD time slot, set a young time slot ID (from the start of the TDD frame to the TDD boundary) for each pair of opposing radio nodes. ) Is the transmission period, and the node with the old number (from the TDD boundary to the end of the TDD frame) as the reception period is positioned “upper”, and conversely, the old number of the time slot ID is the transmission period and the young number is the reception period. Is assigned to the subordinate node, and for each radio link, one of the radio nodes located higher or lower than the radio link is assigned a frequency slot, so that a radio channel is assigned to each radio link. We have invented a system that can be assigned without difficulty and filed a patent application (Japanese Patent Application No. 2001-149535).
[0009]
Furthermore, by allowing each node to terminate not only a wireless link that positions its own node at the higher level but also a wireless link that positions its own node as a lower level, the network is an odd number having a plurality of nodes as vertices. A system has been invented and applied for a patent (Japanese Patent Application No. 2002-70259) in which a radio channel can be assigned to each radio link without fail even when a polygonal circuit is included.
[0010]
FIG. 10 is a diagram showing a configuration of a wireless network that adopts the above-described prior art (Japanese Patent Application No. 2002-70259) and allocates wireless channels (frequency slots in the present embodiment) based on the TDD method in an autonomous and distributed manner. It is.
[0011]
A pair of nodes N that terminate each radio link L1 to Ln, one of which is positioned as "upper" and the other as "lower", regardless of whether it is higher or lower, only that one node, A frequency slot may be assigned to the radio link. For example, in the case of the wireless link L1, if one node N1 that terminates the wireless link L1 is positioned at the upper level (●), the other node N4 is positioned at the lower level (◯). Similarly, in the case of the radio link L2, one node N1 that terminates the radio link L2 is positioned higher, and the other node N2 is positioned lower.
[0012]
Here, nodes that are positioned higher than all of the wireless links that terminate at the own node, such as wireless nodes N1 and N7, are defined as “upper nodes”, and all of the wireless links that terminate at nodes N2 and N4, for example. Nodes that are positioned in the lower order are defined as “subordinate nodes”.
[0013]
By the way, if all of the closed loops constituted by a plurality of wireless nodes are even polygons, all the nodes can be defined as either upper nodes or lower nodes as shown in FIG. However, as shown in FIG. 10, when an odd-polygonal closed circuit (for example, a triangle having nodes N1, N3, and N4 as vertices) is included, some nodes are forced to be positioned at both higher and lower levels. The
[0014]
For example, the node N3 in FIG. 10 is positioned higher than the radio link L4, but is positioned lower than the other radio links L3, L5, and L6. Similarly, the node N5 is positioned higher than the radio link L8, but is positioned lower than the other radio links L7 and L9. Like the nodes N3 and N5, nodes positioned both higher and lower than the radio link terminated at the own node are the “upper node” and “lower node” (collectively, “pure anode”). In some cases, it is defined as “hybrid node”.
[0015]
[Problems to be solved by the invention]
In the case of a pure anode, even if the same FID is used in adjacent links, problems such as interference do not occur as long as the TDD boundary matches. On the other hand, in the hybrid node, the same FID as that used in the radio link that positions the own node “upper” cannot be used in the adjacent link that positions the own node “lower”. Therefore, in the hybrid node, as compared with the pure anode, there are more restrictions on the frequency slots that can be allocated, and the frequency utilization efficiency is inevitably lowered. The occurrence of such hybrid nodes should be suppressed as much as possible. Even when the occurrence is unavoidable, it should be placed in a place with a small amount of traffic.
[0016]
However, in the above-described prior art, since the hybrid node is fixedly arranged, even if the traffic amount of the hybrid node increases for some reason, it cannot cope with this.
[0017]
An object of the present invention is to provide a communication method and system for a wireless mesh network that can solve the above-described problems of the prior art and can appropriately establish a wireless channel according to traffic fluctuations.
[0018]
[Means for Solving the Problems]
In order to achieve the above-described object, the present invention exclusively sets a pair of wireless nodes that terminate each wireless link to either the upper or lower order on a mesh network in which a plurality of wireless nodes are distributed and arranged. In a wireless communication system that positions and autonomously establishes a TDD wireless channel on the wireless link, each wireless node is characterized by the following measures.
[0019]
(1) characterized in that it includes means for detecting a request for switching between the positioning of the own node and the positioning of the opposite node, and means for switching between the positioning of the own node and the positioning of the opposite node in response to the switching request. .
[0020]
(2) As the switching request, a means for detecting a request to add a radio channel to the radio link, a means for selecting a frequency slot to be used in a radio channel to be added in response to the addition request, and a selected frequency Means for determining whether or not a slot satisfies a predetermined allocation condition; means for adding a radio channel using a frequency slot satisfying the allocation condition; and when the frequency slot cannot satisfy the allocation condition Means for determining whether the allocation condition is satisfied if the positioning of the own node and the positioning of the opposite node are switched, and when the allocation condition is satisfied if the positioning is switched, the own node Is switched from one of the upper and lower positions to the other, and a radio channel is added at the position after switching.
[0021]
(3) means for recognizing the topology configuration of the network and generating topology information, means for generating a default TDD plan that defines the standard positioning of each node for each radio link based on the topology information, and the default Means for comparing the positioning specified in the TDD plan with the positioning of the own node and detecting that the two are different as the switching request, and responding to the switching request, the positioning of the own node to the default TDD plan. In addition, switching from one of the upper and lower levels to the other is characterized.
[0022]
According to the feature (1) described above, each node can switch its own node positioning (TDD positioning attribute) with respect to each radio link from one of the upper and lower ones to the other. The reverse is possible.
[0023]
According to the feature (2) described above, when each node is a hybrid node, traffic demand increases, and when a radio channel cannot be added due to a lack of FID, each node can switch its own node's TDD positioning attribute to I can be. In the pure node, more FIDs can be allocated than in the hybrid node, so that a radio channel can be added in response to an increase in traffic demand.
[0024]
According to the above feature (3), each node can return its own TDD positioning attribute to the TDD positioning attribute specified in the default TDD plan, so that an environment where switching of TDD positioning attributes does not occur frequently can be easily reproduced. it can.
[0025]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a main part of a radio node according to the present invention, and all of “upper node”, “lower node”, and “hybrid node” have the same configuration.
[0026]
The wireless node N includes an outdoor wireless device 2 and an indoor wireless device 3. The outdoor wireless device 2 includes a plurality of outdoor wireless units (# 1 to #n) 21. The indoor wireless device 3 includes a plurality of wireless channel units (# 1 to #n) 33, a switch 31, a wireless channel manager (CHM) 32, and a network control device 34.
[0027]
The network control device 34 includes a topology information generator 34a and a default TDD plan generator 34b. The topology information generator 34a recognizes the topology configuration of the network and generates topology information. As will be described in detail later, the default TDD plan generation unit 34b obtains the standard positioning (upper or lower) of each node for each radio link based on the topology information, and uses this as the default TDD plan for the radio channel manager 32. To notify.
[0028]
The radio channel manager 32 dynamically assigns the radio channel units 33 (# 1 to #n) to the radio links L1 to L4 according to the notification from the network control device 34. Each of the radio channel units 33 establishes a TDD radio channel with the radio channel unit of the opposite node, and controls the TDD frame and TDD boundary. The switch 31 connects each radio channel unit 33 (# 1 to #n) to each radio link L1 to L4 based on the assignment result.
[0029]
Next, the operation of each wireless node will be described in detail with reference to the flowchart. FIG. 2 is a flowchart showing the main operation of the wireless node. In step S1, the topology generation unit 34a of the network control device 34 recognizes the network topology. In step S2, the default TDD plan described above is generated based on the recognition result. The procedure for generating a default TDD plan will be described below.
[0030]
If the network includes anything other than even polygonal cycles, a hybrid node is forced to occur. When traffic demand increases in a hybrid node, it is desirable to make it pure anodized to increase the efficiency of FID usage. For this purpose, each node dynamically switches its own TDD positioning attribute for each radio link. Need to be able to.
[0031]
However, if TDD positioning attribute switching occurs frequently, transmission efficiency decreases, so even if TDD positioning attribute switching is allowed, it is necessary to minimize the occurrence, and for this purpose, positioning of each node is required. It is desirable to always set appropriately. Therefore, in this embodiment, a “default TDD plan” is defined that is determined only from network topology information without depending on traffic demand.
[0032]
The “Default TDD Plan” gives a standard setting for the TDD positioning attribute to all wireless nodes on the network. It does not fix the instantaneous instantaneous allocation, but dynamically switches the TDD positioning attribute. It also serves as a guide so that it can be carried out efficiently without being generated unnecessarily and even when switching occurs. Therefore, in this embodiment, not only immediately after the start of operation, but also after switching some TDD positioning attributes according to traffic fluctuations, if the traffic volume decreases, guidance to return to the “default TDD plan” The principle is working.
[0033]
The present inventors have devised an algorithm for creating a “default TDD plan” that can set the TDD positioning attribute in all links so as not to generate hybrid nodes as much as possible for an arbitrary network topology. Introduced in the “Research Results Report” for the Telecommunications Broadcasting Organization created in the process of FWA research and development. The contents will be briefly described below.
[0034]
Although the task of “Default TDD Plan” itself belongs to the task of directing undirected graphs in graph theory, as a result of investigation, there is no conventional example of the definition of the objective function itself. In a simple directed graph, the number of sides incident to and outgoing from one focused point Vi (that is, the incoming order and outgoing order) are represented by DEGin (Vi) and DEGout (Vi), respectively. Here, if the direction from the lower node to the upper node is incident and the opposite is defined as the emission, there is only the incident side or the emission side except for the hybrid node.
DEGin (Vi) = 0
Or
DEGout (Vi) = 0
Is established. Therefore, as a hybrid node definition,
DEGin (Vi) DEGout (Vi)> 0
It can be expressed as point Vi that satisfies
[0035]
So, the TDD positioning attribute is optimized
[Expression 1]

Or
[Expression 2]

However,
[Equation 3]

It is considered appropriate to think of the graph as directed so that is small. Here, the index 2 corresponds to the total number of hybrid nodes in the network, and the index 1 is the total sum of the products of the number of links positioned higher and the number of links positioned lower in the hybrid node. If this sum is large, it is natural to assume that the number of links terminated by the hybrid node is large. If the number of links is large, it is natural to assume that the total traffic volume is high and the demand for frequency repetition is high. From the viewpoint that the number of links that must be switched in order to make a node pure anodized, the number of nodes is weighted.
[0036]
This will be described with reference to specific examples shown in FIGS. FIGS. 6 and 7 are simple examples in which the index 1 is the same and the index 2 is different. The number of hybrid nodes is “1” in FIG. 6 and “2” in FIG. 7 is preferable.
[0037]
On the other hand, FIGS. 8 and 9 are simple examples in which the magnitude relationship between the index 1 and the index 2 is different. In FIG. 8, index 2 is “1”, which is smaller than “2” in FIG. 9, but regarding index 1, FIG. 9 is smaller than “4” in FIG. 8. In the examples of FIGS. 8 and 9, the central node plays a hub role, and it can be considered that priority should be given to increasing the frequency utilization efficiency at this node. Therefore, it is appropriate to make FIG. 8 preferable to FIG. It is.
[0038]
Based on these, it is considered appropriate to use index 1 as the first priority as a performance index and use index 2 as the second priority under the same conditions.
[0039]
From the above examination results, an algorithm for directing an undirected graph so as to minimize the indices 1 and 2 was examined. In the case of exhaustive search, the number of combinations is a power (n!) Of “the number of sides (size of graph) −1”, so that the calculation time immediately after the number of sides exceeds about 10 Is explosive and unrealistic. Therefore, in this embodiment, a heuristic algorithm is developed based on the following policy.
[0040]
Policy: If there is an odd-numbered circuit, remove some edges, remove them, and first direct the resulting subgraph (with the index at 0). The removed edges are added while selecting the direction in which the increase in index is small.
[0041]
The outline of the specific algorithm is as follows.
(1) Step 1: Select one reference point. The number of the shortest arrival hops from the selected point to another point is calculated in advance.
(2) Step S2: Check the number of hops at both ends of the graph for all the edges of the graph. If one is an even number and the other is an odd number, the edge is directed from an even number to an odd number. Directed. If both are even or odd, the edge is temporarily removed from the graph.
{Circle around (3)} Step 3: Steps S1 and S2 are repeated as many times as the number of points as a loop of reference points, and the obtained result having the smallest number of removed sides is selected as a candidate. In the case of a tie, the one with the largest number of subgraph trees with the removed edge as the element is selected. (If this is also a tie, the first one found as a candidate without setting further criteria)
{Circle around (4)} Step 4: Select one tree included in the subgraph consisting of the removed edges, direct it, and add it to the already directed graph. The direction of orientation is 2 ^ St, where the number of sides of the tree is St, but the one with the smallest objective function after addition is selected (exhaustive search).
(5) Step S5: Step S4 is repeated until all the sides removed in step S3 are added. As a result, a default TDD plan is generated.
[0042]
Returning to FIG. 2, in step S3, when the TDD positioning attribute specified in the default TDD plan is different from the current TDD positioning attribute of the own node, the TDD positioning attribute of the own node is adjusted to the default TDD plan, A “default TDD plan reflection process” for switching from one of the upper and lower levels to the other is executed.
[0043]
FIG. 3 is a flowchart showing the procedure of the “default TDD plan reflection process”. In step S10, the current TDD positioning attributes for the radio links L1, L2, L3, and L4 terminated at the own node are as follows. It is compared with the TDD positioning attribute defined in the default TDD plan, and the presence / absence of a radio link that requires switching of the TDD positioning attribute is determined. Here, all radio links having an attribute different from the TDD positioning attribute defined in the default TDD plan are recognized as the switching target links. If there is a wireless link to be switched, the process proceeds to step S11.
[0044]
In step S11, the line quality of the switching target link is obtained. In this embodiment, CNR {C: carrier, N: noise, R: ratio, that is, C / N} is adopted as the channel quality, and this is compared with the reference value CNRref. If CNR> CNRref, the process proceeds to step S12, and the radio link for switching the TDD positioning attribute is selected. If there are a plurality of target links, one of them is selected at random.
[0045]
In step S13, it is determined whether or not the selected switching target link satisfies a predetermined switching condition. In this embodiment, the switching condition is that an adjacent link that is adjacent to the switching target link and has the same TDD positioning attribute as the switching target link does not use the same FID as that used in the switching target link. Is set as If the same FID is used, interference occurs after the switching of the TDD positioning attribute, so the switching of the TDD positioning attribute is stopped for the switching target link. On the other hand, if the same FID is not used, it is determined that the switching condition is satisfied, and the process proceeds to step S14.
[0046]
In step S14, the switching instruction and switching timing of the TDD positioning attribute are notified to the opposite node. In step S15, the TDD positioning attribute of the current switching target link is switched, and the TDD positioning attribute is switched from the upper level to the lower level or from the lower level to the upper level for all radio channels included in the radio link.
[0047]
Returning to FIG. 2, in step S <b> 4, the number of requested channels notified from the network control device 34 is determined. The network controller 34 periodically or irregularly calculates the appropriate number of channels for each of the radio links L1, L2, L3, and L4 whose one end is terminated at its own node based on the network topology and traffic volume. The result is notified to the radio channel manager 32 as the requested channel number Lreq (Lreq (1) to Lreq (4)).
[0048]
In step S5, the notified request channel number Lreq is compared with the current radio channel number Lref, and for a radio link of Lreq> Lref, the process proceeds to step S6. In step S6, “pre-processing for channel addition” is executed, the FID assigned to the radio channel to be added is determined, and further, switching of the TDD positioning attribute is executed as necessary.
[0049]
FIG. 4 is a flowchart showing the procedure of “channel addition preprocessing”. In step S31, among the unassigned FIDs that are not assigned to any radio channel, there is an FID that satisfies the assignment condition. Is determined. In the present embodiment, the following three are set as allocation conditions.
[0050]
(1) Condition 1: It is not assigned to another radio link having a different TDD positioning attribute.
[0051]
(2) Condition 2: Satisfy predetermined line quality continuously for a certain period
[0052]
(3) Condition 3: The FID cannot be selected such that links that are not adjacent to each other (that is, links terminated at different nodes) must have the same TDD boundary.
[0053]
That is, as shown in FIG. 12, when the radio link L1 with F1 of F1 is established between the nodes N1 and N2, and the FID of the other radio link L2 terminated at the node N1 is F1, the links L1 and L2 The TDD boundary is set to be the same to avoid mutual interference. In this state, when adding a channel to another radio link L3 terminated at the node N2, if the FID is set to F1, in order to avoid interference between the node L3 and the node L1, the TDD boundary between the two is set. Must be matched, and this also affects the node L2, so that the TDD boundary must be matched even though the node L3 and the node L2 are not adjacent links.
[0054]
Similarly, since the radio links L2 and L3 are not adjacent to each other, it is not necessary to match the TDD boundary even if both use the same FID (eg, F1). However, when a channel is added to the radio link L1 in such a state, if the FID is set to F1, the radio links L2 and L3 adjacent to the radio link L1 set the TDD boundary to that of the radio link L1. As a result, the node L3 and the node L2 must match the TDD boundary even though they are not adjacent links.
[0055]
As described above, if control for matching TDD boundaries is performed across a plurality of nodes, transmission of traffic information necessary to determine TDD boundaries becomes complicated. In the present embodiment, in order to avoid such a situation, it is not possible to select an FID in which links that are not adjacent to each other (links L2 and L3 in FIG. 12) must match the TDD boundary. Condition 3 is satisfied. If there is an FID that satisfies this allocation condition, the process proceeds to step S32, and one of them is selected.
[0056]
On the other hand, if there is no FID satisfying the allocation condition, the process proceeds to step 33, and any one of the unallocated FIDs is arbitrarily selected. In step S34, it is determined whether or not the FID can satisfy the allocation condition even if the TDD positioning attribute is switched. If satisfied, the process proceeds to step S35. In step S35, even if the TDD positioning attribute of the radio link to which the radio channel is to be added is switched, it is determined whether or not all the FIDs currently used in the existing radio channel can satisfy the allocation condition. If satisfied, the process proceeds to step S36, where the TDD positioning attribute of the radio link to which the radio channel is added is switched, and the radio channel is added here.
[0057]
On the other hand, if it is determined in step S34 that the allocation condition is not satisfied even if the TDD positioning attribute is switched, or if the TDD positioning attribute is switched in step S35, the FID being used in the existing radio channel is allocated. If it is determined that the condition cannot be satisfied, the process proceeds to step S37. In step S37, it is determined whether or not there is another unassigned FID. If there is, the process returns to step S33 and the above processes are repeated for other FIDs. If there is no other unallocated FID, the process ends.
[0058]
Returning to FIG. 2, when the FID to be assigned to the channel to be added is determined as described above, in step S7, negotiation with the opposite node and TDD boundary and frame synchronization is performed, and then the radio channel is added at a predetermined timing. Is done.
[0059]
FIG. 5 is a diagram schematically showing how FID utilization efficiency is improved by switching the TDD positioning attribute. Here, a set of FIDs that can be used in the system is represented by “Fn”, and a hybrid node A set of FIDs that can be used only by radio links that are positioned on the upper side is expressed by “Fa”, and a set of FIDs that can be used only by links that position the hybrid node on the lower side is expressed by “Fn-a”. Therefore, “Fn”, “Fa”, and “Fn−a” have a relationship of “Fn = Fa + Fn−a”.
[0060]
As shown on the left side of the figure, when the nodes N1 and N4 are upper nodes (●), the nodes N2 and N5 are lower nodes (◯), and the node N3 is a hybrid node, the wireless link L1 between the nodes N1 and N3 The set of available FIDs is Fn-a, and the set of FIDs available on the radio link L2 between the nodes N2 and N3 and the radio link L3 between N3 and N5 is Fa. Therefore, there are many restrictions on the FID that can be used in these radio links, and it is difficult to add radio channels.
[0061]
In this state, when a channel addition is requested for the links L1 and L3, the node N3 performs switching of the TDD positioning attribute with respect to the links L2 and L3. As a result, as shown on the right side of FIG. Is switched from the hybrid node to the lower node, and in response, the node N5 is switched from the lower node to the upper node, and the node N2 is switched from the lower node to the hybrid node.
[0062]
As a result, the set of FIDs that can be used on the radio links L1 and L3 that have increased traffic demand becomes Fn, so that radio channels can be added in response to an increase in traffic demand.
[0063]
【The invention's effect】
According to the present invention, the following effects are achieved.
(1) Since each node can switch its own node positioning (TDD positioning attribute) with respect to each radio link from one of the upper and lower levels to the other, it is possible to shift from the hybrid node to the pure node and vice versa.
(2) Each node can become a pure node by switching its own TDD positioning attribute when traffic demand increases when the node is a hybrid node and a radio channel cannot be added due to lack of FID. In the pure node, more FIDs can be allocated than in the hybrid node, so that a radio channel can be added in response to an increase in traffic demand.
(3) Since each node can return its own TDD positioning attribute to the TDD positioning attribute specified in the default TDD plan, it is possible to easily reproduce an environment in which switching of TDD positioning attributes does not occur frequently.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a main part of a radio communication system mounted on each radio node.
FIG. 2 is a flowchart showing the operation of each wireless node.
FIG. 3 is a flowchart of a default TDD plan reflection process.
FIG. 4 is a flowchart of pre-channel addition processing.
FIG. 5 is a diagram schematically illustrating how FID utilization efficiency is improved by switching DD positioning attributes.
FIG. 6 is a diagram (No. 1) for describing a creation method of a default TDD plan;
FIG. 7 is a diagram (No. 2) for describing a creation method of a default TDD plan;
FIG. 8 is a diagram (No. 3) for explaining a creation method of a default TDD plan;
FIG. 9 is a diagram (No. 4) for describing a creation method of a default TDD plan;
FIG. 10 is a diagram schematically showing a configuration of a network including an odd polygonal closed circuit.
FIG. 11 is a diagram schematically showing the configuration of a network including only even polygonal cycles.
FIG. 12 is a diagram for explaining allocation conditions.
[Explanation of symbols]
2 ... Outdoor wireless device, 3 ... Indoor wireless device, 31 ... Switch, 32 ... Wireless channel manager, 33 ... Wireless link control device, 34 ... Network control device

Claims (4)

On a mesh network in which multiple wireless nodes are distributed, a pair of wireless nodes that terminate each wireless link is based on whether the transmission period is before or after the TDD boundary set in each TDD frame. In a wireless mesh network that is positioned exclusively in either the upper or lower level and autonomously launches a TDD wireless channel on the wireless link based on this positioning,
Each wireless node is
Means for detecting a request to switch between the positioning of the own node and the positioning of the opposite node;
A wireless mesh network communication system, comprising: means for switching between the positioning of the own node and the positioning of the opposite node in response to the switching request. Means for detecting a request to add a radio channel to the radio link as the switching request;
Means for selecting a frequency slot to be used in a radio channel to be added in response to the addition request;
Means for determining whether or not the selected frequency slot satisfies a predetermined allocation condition;
Means for adding a radio channel using a frequency slot satisfying the allocation condition;
Means for determining whether the allocation condition is satisfied by switching between the positioning of the own node and the positioning of the opposite node when the frequency slot cannot satisfy the allocation condition;
2. When the allocation is satisfied if the positioning is switched, the positioning of the own node is switched from one of the upper level and the lower level to the other, and a radio channel is added at the position after switching. A communication system of the wireless mesh network described. Means for recognizing the topology configuration of the network and generating topology information;
Means for generating a default TDD plan that defines the standard positioning of each node for each radio link based on the topology information;
Comparing the positioning specified in the default TDD plan with the positioning of the own node, and detecting that the two are different as the switching request,
3. The wireless mesh network communication system according to claim 1, wherein, in response to the switching request, the position of the own node is switched from one of the upper level and the lower level to the other in accordance with the default TDD plan. On a mesh network in which multiple wireless nodes are distributed, a pair of wireless nodes that terminate each wireless link is based on whether the transmission period is before or after the TDD boundary set in each TDD frame. In a radio communication system that is positioned exclusively in either the upper or lower level and autonomously launches a TDD radio channel on the radio link based on this positioning,
Each wireless node
Detecting a request to add a radio channel to the radio link;
Selecting a frequency slot to be used in the additional radio channel;
Determining whether the selected frequency slot satisfies a predetermined allocation condition;
A procedure for determining whether or not the allocation condition is satisfied by switching between the positioning of the own node and the positioning of the opposite node when the frequency slot cannot satisfy the allocation condition;
When the positioning is switched and the allocation condition is satisfied, the positioning of the own node is switched from one of the upper and lower levels to the other, and a radio channel is added in the position after switching,
Generate a default TDD plan that defines the standard positioning of each node for each radio link based on the topology configuration of the network;
A procedure for comparing the position defined in the default TDD plan with the current position of the own node;
A step of switching the positioning of the own node from one of the upper level and the lower level to the other in accordance with the default TDD plan when the positioning specified in the default TDD plan is different from the positioning of the own node. Wireless mesh network communication method.

Images