JP3742571B2

JP3742571B2 - Route maintenance method, route maintenance system, and wireless node device in wireless network

Info

Publication number: JP3742571B2
Application number: JP2001258319A
Authority: JP
Inventors: 穣片山; 晃平水野; 栄亮工藤; 正芳中山; 博人須田
Original assignee: Nippon Telegraph and Telephone Corp; NTT Inc USA
Current assignee: NTT Inc; NTT Inc USA
Priority date: 2001-08-28
Filing date: 2001-08-28
Publication date: 2006-02-08
Anticipated expiration: 2021-08-28
Also published as: JP2003069625A

Description

【０００１】
【発明の属する技術分野】
本発明は、各無線ノードが特定の無線ノードに頻繁にアクセスし、他の無線ノード間のアクセスが少ない無線ネットワークにおいて、特定の無線ノードとの経路維持を行う経路維持方法、経路維持システムおよび無線ノード装置に関する。
【０００２】
【従来の技術】
無線ネットワークにおける無線ノード（以下「ノード」という）間でのパケット転送に関する従来の経路管理は、経路を常時構築しておくテーブル駆動方式と、通信の必要が生じた際に転送経路を構築するオンデマンド方式と、テーブル駆動方式とオンデマンド方式を組み合わせたハイブリッド方式とがある。
【０００３】
テーブル駆動方式は、各ノードが定期的あるいはトポロジーの変化を検出したときに、経路情報を各ノードに通知し、各ノードがその情報に基づいて経路を再構築して経路を維持するので、通信が生じたときにすぐにパケットの送信を開始することができる。しかし、特定のノードが多くのノードから頻繁にアクセスされる通信形態では、不要な経路構築のための処理や通信が生じる。
【０００４】
一方、オンデマンド方式は、例えば文献（Charles E. Perkins, "Ad Hoc On-Demand Distance Vector Routing", in Proceedings of the 2nd IEEE Workshop on Mobile Computing Systems and Applications, pp.90-100, 1999 ）で提案されているＡＯＤＶ（Ad Hoc On-Demand Distance Vector Routing) では、通信が生じたときにネットワーク全体に経路探索パケットをブロードキャストし、宛先ノードなどがその返信を返すことにより経路構築が可能になる。したがって、特定の経路に対して経路維持のための処理は不要となるが、通信が生じてから最初のデータパケットが送信されるまでの遅延が避けられない。また、無線リンクで直接通信可能な範囲にある隣接ノードを認識できなければ通信ができないので、経路維持パケットを直接通信可能な範囲にある隣接ノードにブロードキャストする。この経路維持パケットを受信したノードは、経路維持パケットを受信することにより隣接ノード（送信ノード）を認識するが、一定時間この経路維持パケットが受信されなければ、障害等によるトポロジーの変化が起きたことを認識する。このため、トポロジーの変化が起きたことを認識するまでに、この一定時間以上の時間が必要になる。
【０００５】
また、ハイブリッド方式は、例えば文献（Zygmunt J. Hass,"A New Routing Protocol for the Reconfigurable Wireless Networks", in IEEE ICUPC '97,pp.652-566, 1997)で提案されているＺＲＰ（Zone Routing Protocol)では、ゾーン半径という概念を導入し、各ノードはゾーン半径内のノードへの経路はテーブル駆動方式で管理し、ゾーン半径外のノードへの経路はオンデマンド方式で構築することとしている。このため、特定のノードがゾーン半径内になるようにゾーン半径を設定すると、テーブル駆動方式と同様に、特定のノードが多くのノードから頻繁にアクセスされる通信形態では、不要な経路構築のための処理や通信が生じる。一方、特定のノードがゾーン半径外になるようにゾーン半径を設定すると、オンデマンド方式と同様に、特定のノードが多くのノードから頻繁にアクセスされる通信形態では常に遅延が生じることになる。
【０００６】
【発明が解決しようとする課題】
ところで、外部のネットワークとのゲートウェイとなる特定のノードは、多くのノードから頻繁にアクセスされることになるので、このような通信形態では上記のような問題が起こる。
【０００７】
すなわち、従来のテーブル駆動方式では、すべてのノードに対する経路情報をもつ必要があり、障害等によりトポロジーの変化を検出すると、検出したノードは経路の再構築のために全ノードに通知し、各ノードはこの通知を元に経路を再構築する必要がある。このように、すべてのノードの経路を維持することは、「特定のノード」へのアクセス頻度が高いシステムでは、「特定のノード」以外の経路構築処理、経路維持のための処理、これらの処理に伴う無線帯域が無駄になる。
【０００８】
また、従来のオンデマンド方式は、経路維持を行う必要がない代わりに、特定のノードが多くのノードから頻繁にアクセスされる通信形態では常に遅延が生じることになる。特に、各ノードは、経路維持パケットを隣接ノードにのみ送信するだけで、隣接ノードから経路維持パケットの返信を受けることはなく、隣接ノードが経路維持パケットを受信しているか否かを確認できなかった。したがって、隣接ノードから受信できるが隣接ノードへ送信できない場合には対処できず、別の手段でトポロジーの変化を認識する必要があったので、トポロジーの変化を認識して経路を再構築するまでに時間がかかり、遅延が大きくなっていた。
【０００９】
また、従来のハイブリッド方式は、ノードからの距離によってテーブル駆動方式とオンデマンド方式とを切り替えるので、従来のテーブル駆動方式およびオンデマンド方式の各問題点をそのまま有している。
【００１０】
本発明は、各ノードが特定のノードに頻繁にアクセスし、他のノード間のアクセスが少ない無線ネットワークにおいて、特定のノードとの経路維持に対しては遅延を小さくし、無駄な経路構築処理、経路維持のための処理、これらの処理に伴う無線帯域の無駄を省くことができる経路維持方法、経路維持システムおよび無線ノード装置を提供することを目的とする。
【００１１】
【課題を解決するための手段】
本発明の無線ネットワークにおける経路維持方法、経路維持システムおよび無線ノード装置は、特定無線ノードが各無線ノードに対して、または各無線ノードが特定無線ノードに対して経路維持パケットを送信し、経路維持パケットを受信したノードが経路維持返信パケットを送信し、経路維持パケットの送信ノードが経路維持返信パケットの受信確認により、特定無線ノードと各無線ノードとの間の経路を維持するものである。
【００１２】
請求項１，７，１３，１４に記載の発明では、特定無線ノードは、「宛先」、「隣接」、「ホップ数」、「経路チェック」を記した経路管理表を保持し、「隣接」を同一とする「宛先」すべてを記載した経路維持パケットを生成し、この経路維持パケットを「隣接」に該当する無線ノードへユニキャストし、経路維持パケットを受信した無線ノードが送信した経路維持返信パケットを受信して各経路を維持していることを確認する。
【００１３】
無線ノードは、「宛先」、「隣接」、「ホップ数」を記した経路管理表を保持し、経路維持パケットを受信したときに、この経路維持パケットに記述された全ての「宛先」を取り出し、その「宛先」に対して経路管理表の「隣接」を同一とする全ての「宛先」を記載した経路維持パケットを生成し、この経路維持パケットを「隣接」に該当する無線ノードへユニキャスト中継し、経路維持パケットを中継する必要がない場合または経路維持パケットを中継して所定時間経過した場合に、特定無線ノードへの経路維持返信パケットを生成してユニキャスト送信し、経路維持返信パケットを受信したときはその経路維持返信パケットに自ノードで経路維持パケットを受信したことを示す情報を付加してユニキャスト中継する。
【００１４】
請求項２，８，１３，１４に記載の発明では、特定無線ノードは、「宛先」、「隣接」、「ホップ数」、「経路チェック」を記した経路管理表を保持し、経路維持パケットを生成し、この経路維持パケットを無線リンクで直接通信可能な範囲にある隣接する無線ノードへブロードキャストし、経路維持パケットを受信した無線ノードから送信された経路維持返信パケットを受信して各経路を維持していることを確認する。
【００１５】
無線ノードは、「宛先」、「隣接」、「ホップ数」を記した経路管理表を保持し、経路維持パケットを受信したときに、その発信元を確認して特定無線ノードからの初めての経路維持パケットである場合には、特定無線ノードへの経路維持返信パケットを生成してユニキャスト送信する。
【００１６】
請求項３，９，１３，１４に記載の発明では、無線ノードは、「宛先」、「隣接」、「ホップ数」、「経路チェック」を記した経路管理表を保持し、所定の時刻に経路維持パケットを生成して特定無線ノードへユニキャストし、経路維持パケットを受信したときに自ノードを一意に識別できる情報を経路維持パケットに付加して特定無線ノードへ中継し、特定無線ノードが送信した経路維持返信パケットを受信すると特定無線ノードとの経路を経路管理表に記録し、この経路維持返信パケットが自ノード宛ての場合はそれを廃棄し、他の無線ノード宛ての場合はそれを中継する。
【００１７】
特定無線ノードは、「宛先」、「隣接」、「ホップ数」を記した経路管理表を保持し、経路維持パケットを受信したときに、この経路維持パケットから通過した無線ノードの情報を取り出し、その経路情報を経路管理表に記録し、経路維持パケットから送信元を取り出してその送信元へ経路維持返信パケットをユニキャストする。
【００１８】
請求項４，１０，１３，１４に記載の発明では、無線ノードは、「宛先」、「隣接」、「ホップ数」、「経路チェック」を記した経路管理表を保持し、所定の時刻に経路維持パケットを生成して隣接する無線ノードへブロードキャストし、経路維持パケットを受信したときにその送信元へ経路維持返信パケットをユニキャストし、隣接する無線ノードから経路維持返信パケットを受信できないときは、経路管理表からその隣接する無線ノードに対応する情報を削除して経路の再構築を行う。
【００１９】
請求項５，１１，１３に記載の発明では、特定無線ノードは、経路管理表に記された「経路チェック」に用いるパケットの受信に要する所定時間について、無線ノードからその所定時間内にパケットが送られてこない場合には、無線ノードとの間の経路を再構築する。
【００２０】
請求項６，１２，１４に記載の発明では、無線ノードは、他の無線ノードに経路管理表に基づいた経路でパケットを送信してから所定時間経過しても返信パケットがない場合に、経路管理表に記された「隣接」のうち、返信パケットのない経路の「隣接」と同一の経路を削除し、障害通知パケットを無線リンクで直接通信可能な範囲にある隣接する無線ノードへブロードキャストし、無線ノードとの間の経路を再構築する。
【００２１】
【発明の実施の形態】
図１は、無線ネットワークの構成例を示す。図において、外部のネットワークとのゲートウェイとなる特定のノードをＡＰ（アクセスポイント）とし、特定ノードＡＰにノードＷＲ１，ＷＲ２が無線リンクを介して接続され、さらにノードＷＲ１とノードＷＲ４が接続され、ノードＷＲ４とノードＷＲ３、ＷＲ５、ＷＲ６およびＷＲ７が接続され、ノードＷＲ２とノードＷＲ５が接続され、さらにノードＷＲ５とＷＲ７が接続され、ノードＷＲ３とノードＷＲ６が接続される。
【００２２】
各ノードの経路管理表は、「宛先（宛先ノード）」、「隣接（宛先ノードにパケットを届けるために転送する隣接ノード）」、「ホップ数（パケットが宛先ノードに届くまでに必要な経路のメトリック）」から構成される。例えば、特定ノードＡＰにおいては、宛先ＷＲ１に対して隣接ＷＲ１、ホップ数１となり、宛先ＷＲ３に対して隣接ＷＲ１、ホップ数２となる。また、各ノードＷＲ１〜７の経路管理表には、特定ノードＡＰを宛先とした場合の隣接およびホップ数が設定されている。
【００２３】
本発明は、特定ノードＡＰが各ノードＷＲ１〜ＷＲ７に対して、または各ノードＷＲ１〜ＷＲ７が特定ノードＡＰに対して経路維持パケットを送信し、経路維持パケットを受信したノードが経路維持返信パケットを送信し、経路維持パケットの送信ノードが経路維持返信パケットの受信確認により、特定ノードＡＰと各ノードＷＲ１〜ＷＲ７との間の経路を維持するものである。以下、特定ノードＡＰから各ノードＷＲ１〜ＷＲ７に対して経路維持パケットを送信するＡＰ主導の実施形態と、各ノードＷＲ１〜ＷＲ７が特定ノードＡＰに対して経路維持パケットを送信するＷＲ主導の実施形態について説明する。
【００２４】
＜ＡＰ主導の実施形態：請求項１，２，５，７，８，１１＞
図２は、ＡＰ主導の場合の特定ノードＡＰの経路管理表を示す。なお、無線ネットワークの構成およびノードＷＲ１〜ＷＲ７の経路管理表は、図１に示すものと同じである。
【００２５】
特定ノードＡＰの経路管理表は、「宛先」、「隣接」、「ホップ数」の他に「経路チェック」の欄が設けられる。「経路チェック」は、経路維持返信パケットの受信確認によって、各ノードが所定時間内に経路維持パケットを受信したか否かを表すものである。
【００２６】
（ＡＰ主導の第１の制御シーケンス：請求項１，７）
ＡＰ主導の第１の制御シーケンスを図３〜図７を参照して説明する。図３は、特定ノードＡＰにおける経路維持パケット送信時の制御シーケンスを示す。図４は、ノードＷＲ１〜ＷＲ７における制御シーケンスを示す。図５は、特定ノードＡＰにおける経路維持返信パケット受信時の制御シーケンスを示す。図６は、経路維持パケットの構成例を示す。図７は、経路維持返信パケットの構成例を示す。
【００２７】
（特定ノードＡＰにおける経路維持パケット送信時の制御シーケンス）
図２，３において、特定ノードＡＰは、経路管理表に「宛先」、「隣接」、「ホップ数」からなる未処理の組があるか否かを調べ、未処理の組があれば、経路維持パケットを生成する。経路維持パケットは、図６に示すように、パケット種別、宛先、隣接、ホップ数、送信元、発信元、実宛先から構成される。なお、実宛先とは、経路維持パケットを受信すべきノードを意味する。
【００２８】
ここで、特定ノードＡＰの経路管理表の「宛先」、「隣接」、「ホップ数」の組を［宛先、隣接、ホップ数］と記述する。まず、図２に示すように［ＷＲ１，ＷＲ１，１］の組があり、隣接ノードがＷＲ１であることが分かる（s1）。ここでは隣接ＷＲ１への経路維持パケットが存在しない（初めての経路維持パケットである）ので（s2）、図６(1) に示すような経路維持パケット（パケット種別：経路維持、宛先：ＷＲ１、隣接：ＷＲ１、ホップ数：１、送信元：ＡＰ、発信元：ＡＰ、実宛先：ＷＲ１）を生成する（s3）。次に、［ＷＲ２，ＷＲ２，１］については、同様に隣接ＷＲ２への経路維持パケット（パケット種別：経路維持、宛先：ＷＲ２、隣接：ＷＲ２、ホップ数：１、送信元：ＡＰ、発信元：ＡＰ、実宛先：ＷＲ２）を生成する（s3）。
【００２９】
次に、［ＷＲ３，ＷＲ１，３］については、すでに隣接ＷＲ１への経路維持パケットが存在するので（s2）、この経路維持パケットの実宛先にＷＲ３を追加する（s4、図６(2) ）。［ＷＲ４，ＷＲ１，２］については、すでに隣接ＷＲ１への経路維持パケットが存在するので（s2）、この経路維持パケットの実宛先にＷＲ４を追加する（s4、図６(3) ）。以下同様に、［ＷＲ５，ＷＲ２，２］、［ＷＲ６，ＷＲ１，３］、［ＷＲ７，ＷＲ２，３］に対しても、それぞれ実宛先の追加による経路維持パケットを生成する。
【００３０】
以上により、経路維持パケットはＷＲ１方向とＷＲ２方向の２種類が生成される。ＷＲ１方向の経路維持パケットは、パケット種別：経路維持、宛先：ＷＲ１、隣接：ＷＲ１、ホップ数：１、送信元：ＡＰ、発信元：ＡＰ、実宛先：ＷＲ１，ＷＲ３，ＷＲ４，ＷＲ６である（図６(4) ）。一方、ＷＲ２方向の経路維持パケットは、パケット種別：経路維持、宛先：ＷＲ２、隣接：ＷＲ２、ホップ数：１、送信元：ＡＰ、発信元：ＡＰ、実宛先：ＷＲ２，ＷＲ５，ＷＲ７である（図６(5) ）。そして、各経路の経路チェックを経路調査中にし、２種類の経路維持パケットをノードＷＲ１，ＷＲ２に送信する。
【００３１】
（ノードＷＲ１〜ＷＲ７における制御シーケンス）
図２，４において、ノードＷＲ１が経路維持パケットを受信すると（s11)、実宛先に自ノード以外のノードがあるか否かを調べる（s12)。ここでは、実宛先に自ノード以外のノードＷＲ３，ＷＲ４，ＷＲ６があるので、ノードＷＲ１の経路管理表に従って経路維持パケットを生成する(s13〜s17)。実宛先ＷＲ４に対しては、経路管理表の［ＷＲ４，ＷＲ４，１］から隣接がＷＲ４であることが分かり、隣接ＷＲ４への経路維持パケットが存在しないので、隣接ＷＲ４への経路維持パケット（パケット種別：経路維持、宛先：ＷＲ４、隣接：ＷＲ４、ホップ数：１、実宛先：ＷＲ４）を生成する。［ＷＲ６，ＷＲ４，２］については、すでに隣接ＷＲ４への経路維持パケットが存在するので、この経路維持パケットの実宛先にＷＲ６を追加する。［ＷＲ３，ＷＲ４，２］についても同様である。これにより、ノードＷＲ１は、パケット種別：経路維持、宛先：ＷＲ４、隣接：ＷＲ４、ホップ数：１、実宛先：ＷＲ３，ＷＲ４，ＷＲ６となる経路維持パケットをノードＷＲ４に送信し（s18)、経路維持返信パケットの受信待ちとなる（s19)。
【００３２】
ノードＷＲ４は、ノードＷＲ１からの経路維持パケットを受信すると、ノードＷＲ１と同様に処理する。すなわち、ノードＷＲ３に対して、パケット種別：経路維持、宛先：ＷＲ３、隣接：ＷＲ３、ホップ数：１、実宛先：ＷＲ３となる経路維持パケットを送信する（s18)。さらに、ノードＷＲ６に対して、パケット種別：経路維持、宛先：ＷＲ６、隣接：ＷＲ６、ホップ数：１、実宛先：ＷＲ６となる経路維持パケットを送信する（s18)。そして、経路維持返信パケットの受信待ちとなる（s19)。
【００３３】
ノードＷＲ３は、ノードＷＲ４からの経路維持パケットを受信すると、実宛先に自ノード以外のノードが存在しないので（s12)、送信元である特定ノードＡＰに対する経路維持返信パケットを生成し、ノードＷＲ４に返信する（s23,s24)。経路維持返信パケットは、図７に示すように、パケット種別、宛先、隣接、ホップ数、送信元、発信元、正常経路から構成される。この経路維持返信パケットは、パケット種別：経路維持返信、宛先：ＡＰ、隣接：ＷＲ４、ホップ数：３、送信元：ＷＲ３、発信元：ＷＲ３、正常経路：ＷＲ３に設定される（図７(1))。ノードＷＲ６もノードＷＲ３と同様の処理を行う。
【００３４】
ノードＷＲ４は、ｓ19において経路維持返信パケットの受信待ちとなっており、ノードＷＲ３とＷＲ６から経路維持返信パケットを受信すると（s20,s21)、それらの正常経路に自ノードを追加した新たに経路維持返信パケットを生成し、受信した経路維持返信パケットを破棄する。新たな経路維持返信パケットは、パケット種別：経路維持返信、宛先：ＡＰ、隣接：ＷＲ１、ホップ数：２、送信元：ＷＲ４、発信元：ＷＲ４、正常経路：ＷＲ３，ＷＲ４，ＷＲ６に設定され（図７(2))、ノードＷＲ１に送信される（s22)。
【００３５】
ノードＷＲ１は、ノードＷＲ４と同様の処理を行う。経路維持返信パケットは、パケット種別：経路維持返信、宛先：ＡＰ、隣接：ＡＰ、ホップ数：１、送信元：ＷＲ１、発信元：ＷＲ１、正常経路：ＷＲ１，ＷＲ３，ＷＲ４，ＷＲ６と設定され（図７(3))、特定ノードＡＰに送信される。
【００３６】
ノードＷＲ２，ＷＲ５，ＷＲ７のルートについても同様に、ノードＷＲ２からパケット種別：経路維持返信、宛先：ＡＰ、隣接：ＡＰ、ホップ数：１、送信元：ＷＲ２、発信元：ＷＲ２、正常経路：ＷＲ２，ＷＲ５，ＷＲ７と設定された経路維持返信パケットが特定ノードＡＰに送信される。
【００３７】
（特定ノードＡＰにおける経路維持返信パケット受信時の制御シーケンス）
図２，５において、特定ノードＡＰは２種類の経路維持パケットをノードＷＲ１，ＷＲ２に送信した後に、経路維持返信パケットの受信待ちになっており（s31)、所定時間内にノードＷＲ１からの経路維持返信パケットを受信すると（s32)、経路維持返信パケットから正常経路ＷＲ１，ＷＲ３，ＷＲ４，ＷＲ６を取り出し、経路管理表の宛先ＷＲ１，ＷＲ３，ＷＲ４，ＷＲ６に対応する各経路チェックを経路調査中からＯＫに変更する（s33)。ノードＷＲ２からの経路維持返信パケットについても同様に処理する（s34)。
【００３８】
（ＷＲ５とＷＲ７との間の障害発生時の処理：請求項５，１１）
図２において、ノードＷＲ５は、ノードＷＲ２から経路維持パケットを受信し、上述した手順に従ってノードＷＲ７に経路維持パケットを転送し、経路維持返信パケットの受信待ち状態に入る（図４、s19)。ここで、所定時間経過してもノードＷＲ７から経路維持返信パケットが受信されない場合には、ノードＷＲ５は経路維持返信パケットを生成する。この経路維持返信パケットは、パケット種別：経路維持返信、宛先：ＡＰ、隣接：ＷＲ２、ホップ数：２、送信元：ＷＲ５、発信元：ＷＲ５、正常経路：ＷＲ５を設定してノードＷＲ２に送信される。
【００３９】
ノードＷＲ２は、受信した経路維持返信パケットのパケット種別：経路維持返信、宛先：ＡＰ、隣接：ＡＰ、ホップ数：１、送信元：ＷＲ２、発信元：ＷＲ２、正常経路：ＷＲ２，ＷＲ５を設定して特定ノードＡＰに送信される。
【００４０】
特定ノードＡＰは、受信した経路維持返信パケットから正常経路ＷＲ２，ＷＲ５を取り出し、経路管理表の宛先ＷＲ２，ＷＲ５に対応する各経路チェックを経路調査中からＯＫに変更する。特定ノードＡＰは、経路維持パケットを送信後から所定時間経過すると、経路管理表の各経路チェックを調べ、経路調査中の宛先ＷＲ７の経路に対しては経路の再構築を行う（図５、s35)。
【００４１】
（ＡＰ主導の第２の制御シーケンス：請求項２，８）
ＡＰ主導の第２の制御シーケンスを図８〜図１１を参照して説明する。図８は、特定ノードＡＰの制御シーケンスを示す。図９は、ノードＷＲ１〜ＷＲ７における制御シーケンスを示す。図１０は、経路維持パケットの構成例を示す。図１１は、経路維持返信パケットの構成例を示す。
【００４２】
経路維持パケットは、図１０に示すように、パケット種別、宛先、隣接、ホップ数、送信元、発信元から構成される。ここでは、パケット種別：経路維持、宛先：ブロードキャスト、隣接：ブロードキャスト、ホップ数：経路管理表のホップ数の最大値（３）、送信元：ＡＰ、発信元：ＡＰに設定される。
【００４３】
図２，８において、特定ノードＡＰは、経路管理表の各経路に対する経路チェックを経路調査中に設定し（s41)、送信元をＡＰにした図１０に示す経路維持パケットをブロードキャストし（s42)、経路維持返信パケットの受信待ちに入る（s43)。
【００４４】
図９において、ノードＷＲ１は、経路維持パケットの受信待ちの状態にあり（s51)、所定時間内に特定ノードＡＰからの経路維持パケットを受信すると（s52)、経路維持パケットの発信元を確認する（s53)。そして、ノードＷＲ１の経路管理表の宛先ＡＰに対する隣接と経路維持パケットの発信元が一致する（ここではともにＡＰ）である場合には (s54)、発信元である特定ノードＡＰに対して経路維持返信パケットを送信する（s55)。経路維持返信パケットは、図１１に示すように、パケット種別、宛先、隣接、ホップ数、送信元、発信元から構成される。ここでは、パケット種別：経路維持返信、宛先：ＡＰ、隣接：ＡＰ、ホップ数：１、送信元：ＷＲ１、発信元：ＷＲ１に設定される。
【００４５】
また、ノードＷＲ１は、発信元と自ノードに設定した経路維持パケットをブロードキャストする（s56)。この経路維持パケットは、パケット種別：経路維持、宛先：ブロードキャスト、隣接：ブロードキャスト、ホップ数：２、送信元：ＡＰ、発信元：ＷＲ１に設定される。
【００４６】
この経路維持パケットを受信した他のノードも同様に処理し、それぞれ経路維持返信パケットを発信元に送信しながら経路維持パケットをブロードキャストする。ただし、ノードＷＲ３，ＷＲ６，ＷＲ７は、ノードＷＲ４からの経路維持パケットのホップ数が１であるので経路維持パケットをブロードキャストしない。経路維持返信パケットを受信した各ノードでは、経路管理表の宛先ＡＰに対する隣接のノード宛に順次経路維持返信パケットを転送し、最終的に特定ノードＡＰに到達する。
【００４７】
図８において、特定ノードＡＰは、所定時間内に受信した経路維持返信パケットの送信元ＷＲ１を取り出し、経路管理表の宛先ＷＲ１に対応する各経路チェックを経路調査中からＯＫに変更する（s44,s45)。他の送信元からの経路維持返信パケットについても同様である。特定ノードＡＰは、経路維持パケットをブロードキャスト後から所定時間経過すると、経路管理表の各経路チェックを調べ、経路調査中の宛先の経路に対しては経路の再構築を行う（s46)。
【００４８】
（ＷＲ５とＷＲ７との間の障害発生時の処理：請求項５，１１）
ＷＲ５とＷＲ７との間に障害が発生した場合には、ノードＷＲ１〜ＷＲ６については経路維持返信パケットが特定ノードＡＰに返信されてくるので、特定ノードＡＰの経路管理表の各経路チェックはＯＫになっている。しかし、ノードＷＲ７からの経路維持返信パケットが特定ノードＡＰに到達しないので、対応する経路チェックは経路調査中のままである。したがって、特定ノードＡＰは、経路維持パケットをブロードキャスト後から所定時間経過すると、経路管理表の各経路チェックを調べ、経路調査中の宛先ＷＲ７の経路に対しては経路の再構築を行う。
【００４９】
＜ＷＲ主導の実施形態：請求項３，４，６，９，１０，１２＞
図１２は、ＷＲ主導の場合のノードＷＲ１〜ＷＲ７の経路管理表を示す。なお、無線ネットワークの構成および特定ノードＡＰの経路管理表は、図１に示すものと同じである。
【００５０】
ノードＷＲ１〜ＷＲ７の経路管理表は、「宛先」、「隣接」、「ホップ数」の他に「経路チェック」の欄が設けられる。「経路チェック」は、経路維持返信パケットの受信確認によって、各ノードが所定時間内に経路維持パケットを受信したか否かを表すものである。
【００５１】
（ＷＲ主導の第１の制御シーケンス：請求項３，９）
ＷＲ主導の第１の制御シーケンスを図１３を参照して説明する。図１３は、ノードＷＲ１〜ＷＲ７における制御シーケンスを示す。ここでは、ノードＷＲ６の場合について説明するが、他のノードＷＲ３，ＷＲ７についても同様である。
【００５２】
図１２，１３において、ノードＷＲ６は、所定時間経過しても特定ノードＡＰ宛ての経路維持パケットを受信しないので（s61)、特定ノードＡＰ宛ての経路維持パケットを生成し、経路管理表に基づいて隣接のノードＷＲ４宛てに送信する（s62)。この経路維持パケット、パケット種別：経路維持、宛先：ＡＰ、隣接：ＷＲ４、ホップ数：３、送信元：ＷＲ６、発信元：ＷＲ６に設定される。そして経路維持返信パケットの受信待ちに入る（s63)。
【００５３】
ノードＷＲ４は、所定時間内にノードＷＲ６から特定ノードＡＰ宛ての経路維持パケットを受信すると（s61)、経路管理表から宛先ＡＰに対する隣接ノードを検索し、ここではノードＷＲ１宛てに経路維持パケットを転送する（s64)。この経路維持パケットは、パケット種別：経路維持、宛先：ＡＰ、隣接：ＷＲ１、ホップ数：２、送信元：ＷＲ６、発信元：ＷＲ４に設定が変更される。そして経路維持返信パケットの受信待ちに入る（s65)。
【００５４】
ノードＷＲ１は、所定時間内にノードＷＲ４から特定ノードＡＰ宛ての経路維持パケットを受信すると、ノードＷＲ４と同様に処理して特定ノードＡＰに経路維持パケットを転送する。この経路維持パケットは、パケット種別：経路維持、宛先：ＡＰ、隣接：ＡＰ、ホップ数：１、送信元：ＷＲ６、発信元：ＷＲ１に設定が変更される。
【００５５】
特定ノードＡＰは、ノードＷＲ１から経路維持パケットを受信すると、経路維持パケットから送信元ＷＲ６を取り出し、ノードＷＲ６を宛先として経路管理表から得た隣接のノードＷＲ１に経路維持返信パケットを送信する。この経路維持返信パケットは、パケット種別：経路維持返信、宛先：ＷＲ６、隣接：ＷＲ１、ホップ数：３、送信元：ＡＰ、発信元：ＡＰに設定される。
【００５６】
ノードＷＲ１は、経路維持返信パケットの受信待ちに入っており（s65)、所定時間内に特定ノードＡＰからノードＷＲ６宛ての経路維持返信パケットを受信すると（s66)、特定ノードＡＰへの経路が正常であることを認識し、経路管理表の宛先ＡＰの経路チェックをＯＫとする。そして、ノードＷＲ１はノードＷＲ６を宛先として、経路管理表から得た隣接のノードＷＲ４に経路維持返信パケットを転送する（s67)。この経路維持返信パケットは、パケット種別：経路維持返信、宛先：ＷＲ６、隣接：ＷＲ４、ホップ数：２、送信元：ＡＰ、発信元：ＷＲ１に設定が変更される。
【００５７】
ノードＷＲ４は、所定時間内にノードＷＲ１からノードＷＲ６宛ての経路維持返信パケットを受信すると、特定ノードＡＰへの経路が正常であることを認識し、経路管理表の宛先ＡＰ，ＷＲ１の経路チェックをＯＫとする。そして、ノードＷＲ１と同様に処理してノードＷＲ６に経路維持返信パケットを転送する。この経路維持返信パケットは、パケット種別：経路維持、宛先：ＷＲ６、隣接：ＷＲ６、ホップ数：１、送信元：ＡＰ、発信元：ＷＲ４に設定が変更される。
【００５８】
ノードＷＲ６は、経路維持返信パケットの受信待ちに入っており（s63)、所定時間内にノードＷＲ４から経路維持返信パケットを受信すると（s68)、特定ノードＡＰへの経路が正常であることを認識し、経路管理表の宛先ＡＰ，ＷＲ４の経路チェックをＯＫとする。このとき、宛先が自ノードであるので、経路維持返信パケットの転送を行わずに処理を終了する。
【００５９】
（ＷＲ２とＷＲ５との間の障害発生時の処理：請求項６，１２）
図１２，１３において、ノードＷＲ５は、ノードＷＲ７から経路維持パケットを受信し、上述した手順に従ってノードＷＲ２に経路維持パケットを転送し、経路維持返信パケットの受信待ち状態に入る（s65)。ここで、所定時間経過してもノードＷＲ２から経路維持返信パケットが返信されず (s66)、後述する障害通知パケットも受信せずに所定時間経過した場合 (s69,s70)や、受信した障害通知パケットが特定ノードＡＰの方向（宛先がＡＰの隣接）から送信されたものであれば（s71)、障害通知パケットを１ホップ分ブロードキャストする (s72)。この障害通知パケットは、パケット種別：障害情報、宛先：ブロードキャスト、隣接：ブロードキャスト、ホップ数：１、送信元：ＷＲ５、発信元：ＷＲ５に設定される。そして、経路管理表の隣接がＷＲ２の経路の設定を削除する（s73)。
【００６０】
ノードＷＲ７は、経路維持返信パケットではなく障害通知パケットを受信し、経路管理表の宛先ＡＰに対する隣接と障害通知パケットの発信元がともにＷＲ５で一致した場合（s75)や、両パケットを受信しないまま所定時間経過すると（s76)、障害通知パケットを１ホップ分ブロードキャストし（s72)、特定ノードＡＰへの経路の設定を削除する（s73)。この障害通知パケットは、パケット種別：障害情報、宛先：ブロードキャスト、隣接：ブロードキャスト、ホップ数：１、送信元：ＷＲ５、発信元：ＷＲ７に設定される。
【００６１】
また、ノードＷＲ４は障害通知パケットを受信すると、経路管理表の宛先ＡＰに対する隣接がＷＲ１であり、障害通知パケットの発信元がＷＲ５で一致しないので、この障害通知パケットを無視する。
【００６２】
以上により、ノードＷＲ５、ＷＲ７は、特定ノードＡＰに対する経路の再構築を行う（s74)が、その方法は上述の本発明による方法や従来のＡＯＤＶ等の経路構築方法を用いてもよい。
【００６３】
（ＷＲ主導の第２の制御シーケンス：請求項４，１０）
ＷＲ主導の第２の制御シーケンスを図１４を参照して説明する。図１４は、ノードＷＲ１〜ＷＲ７における制御シーケンスを示す。ここでは、ノードＷＲ５の場合について説明するが、他のノードについても同様である。
【００６４】
図１２，１４(1) において、ノードＷＲ５は、ある時刻になると経路管理表の経路チェックをすべて経路調査中にする（s81)。そして、経路維持パケットを生成して１ホップ分ブロードキャストする (s82)。この経路維持パケットは、パケット種別：経路維持、宛先：ブロードキャスト、隣接：ブロードキャスト、ホップ数：１、送信元：ＷＲ５、発信元：ＷＲ５に設定される。そして経路維持返信パケットの受信待ちに入る（s83)。
【００６５】
図１４(2) において、ノードＷＲ２が経路維持パケットを受信すると（s91)、経路維持返信パケットを生成し、経路維持パケットを送信したノードＷＲ５宛てに送信する（s92)。この経路維持返信パケットは、パケット種別：経路維持返信、宛先：ＷＲ５、隣接：ＷＲ５、ホップ数：１、送信元：ＷＲ２、発信元：ＷＲ２に設定される。経路維持パケットを受信するノードＷＲ４，ＷＲ７も同様である。
【００６６】
図１４(1) において、経路維持返信パケットの受信待ち中のノードＷＲ５が、所定時間内にノードＷＲ２，ＷＲ４，ＷＲ７からの経路維持返信パケットを受信すると（s84)、経路管理表の隣接ＷＲ２，ＷＲ４，ＷＲ７の経路チェックのＯＫに設定する（s85)。
【００６７】
（ＷＲ２とＷＲ５との間の障害発生時の処理：請求項６，１２）
図１２，図１４(1) において、ノードＷＲ５は、所定時間内にノードＷＲ２からの経路維持返信パケットが受信されず（s84)、ノードＷＲ２との間に障害が発生したことを経路管理表の経路チェックが経路調査中のままであることから認識すると、経路管理表の隣接がＷＲ２の経路を経路構築中に変更する（s86)。経路構築中の経路は特定ノードＡＰの方向であるので、障害通知パケットを１ホップ分ブロードキャストする (s87)。この障害通知パケットは、パケット種別：障害情報、宛先：ブロードキャスト、隣接：ブロードキャスト、ホップ数：１、送信元：ＷＲ５、発信元：ＷＲ５に設定される。これにより、ノードＷＲ５は、特定ノードＡＰに対する経路の再構築を行う（s88)が、その方法は上述の本発明による方法や従来のＡＯＤＶ等の経路構築方法を用いてもよい。
【００６８】
図１４(3) において、ノードＷＲ７は、障害通知パケットを受信し（s93)、経路管理表の宛先ＡＰに対する隣接と障害通知パケットの発信元がともにＷＲ５で一致した場合に（s94)、経路管理表の隣接がＷＲ２の経路を経路構築中に変更し（s95)、障害通知パケットを１ホップ分ブロードキャストする（s96)。この障害通知パケットは、パケット種別：障害情報、宛先：ブロードキャスト、隣接：ブロードキャスト、ホップ数：１、送信元：ＷＲ５、発信元：ＷＲ７に設定される。これにより、ノードＷＲ７は、特定ノードＡＰに対する経路の再構築を行う（s97)が、その方法は上述の本発明による方法や従来のＡＯＤＶ等の経路構築方法を用いてもよい。
【００６９】
また、ノードＷＲ４は障害通知パケットを受信すると、経路管理表の宛先ＡＰに対する隣接がＷＲ１であり、障害通知パケットの発信元がＷＲ５で一致しないので、この障害通知パケットを無視する。
【００７０】
【発明の効果】
以上説明したように、本発明は、各無線ノードが特定の無線ノードに頻繁にアクセスし、他の無線ノード間のアクセスが少ない無線ネットワークにおいて、特定無線ノードが各無線ノードに対して、または各無線ノードが特定無線ノードに対してユニキャストで経路維持パケットを送信し、経路維持パケットを受信したノードが経路維持返信パケットをユニキャスト送信し、経路維持パケットの送信ノードが経路維持返信パケットの受信確認により、特定無線ノードと各無線ノードとの間の経路を維持することができる。したがって、特定の無線ノードとの経路維持に対しては、経路維持に伴う遅延を小さくし、無駄な経路構築処理、経路維持のための処理、これらの処理に伴う無線帯域の無駄を省くことができる。
【図面の簡単な説明】
【図１】無線ネットワークの構成例を示す図。
【図２】ＡＰ主導の場合の特定ノードＡＰの経路管理表を示す図。
【図３】特定ノードＡＰにおける経路維持パケット送信時の制御シーケンスを示す図。
【図４】ノードＷＲ１〜ＷＲ７における制御シーケンスを示す図。
【図５】特定ノードＡＰにおける経路維持返信パケット受信時の制御シーケンスを示す図。
【図６】経路維持パケットの構成例を示す図。
【図７】経路維持返信パケットの構成例を示す図。
【図８】特定ノードＡＰの制御シーケンスを示す図。
【図９】ノードＷＲ１〜ＷＲ７における制御シーケンスを示す図。
【図１０】経路維持パケットの構成例を示す図。
【図１１】経路維持返信パケットの構成例を示す図。
【図１２】ＷＲ主導の場合のノードＷＲ１〜ＷＲ７の経路管理表を示す図。
【図１３】ノードＷＲ１〜ＷＲ７における制御シーケンスを示す図。
【図１４】ノードＷＲ１〜ＷＲ７における制御シーケンスを示す図。
【符号の説明】
ＡＰ特定無線ノード（特定ノード）
ＷＲ無線ノード（ノード）[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a path maintenance method, a path maintenance system, and a radio for maintaining a path with a specific radio node in a radio network in which each radio node frequently accesses a specific radio node and there is little access between other radio nodes. It relates to node equipment.
[0002]
[Prior art]
Conventional route management related to packet transfer between wireless nodes (hereinafter referred to as “nodes”) in a wireless network is a table-driven method that always constructs a route, and on that constructs a forwarding route when communication is required. There are a demand method and a hybrid method combining a table driving method and an on-demand method.
[0003]
The table-driven method notifies each node of route information periodically or when a topology change is detected, and each node reconstructs the route based on the information and maintains the route. The transmission of the packet can be started immediately when the error occurs. However, in a communication mode in which a specific node is frequently accessed from many nodes, unnecessary route construction processing and communication occur.
[0004]
On-demand methods, on the other hand, are proposed in literature (Charles E. Perkins, "Ad Hoc On-Demand Distance Vector Routing", in Proceedings of the 2nd IEEE Workshop on Mobile Computing Systems and Applications, pp. 90-100, 1999). In the AODV (Ad Hoc On-Demand Distance Vector Routing), when a communication occurs, a route search packet is broadcast to the entire network, and a destination node or the like returns its reply, thereby making it possible to construct a route. Therefore, processing for maintaining a route is not required for a specific route, but a delay from when communication occurs until the first data packet is transmitted is unavoidable. Further, since communication is not possible unless an adjacent node within a range in which direct communication can be performed through a wireless link is recognized, a route maintenance packet is broadcast to the adjacent node in a range where direct communication is possible. The node that has received this route maintenance packet recognizes an adjacent node (transmission node) by receiving the route maintenance packet, but if this route maintenance packet is not received for a certain period of time, a topology change due to a failure or the like has occurred. Recognize that. For this reason, it takes a certain time or more to recognize that a change in topology has occurred.
[0005]
Moreover, the hybrid method is, for example, ZRP (Zone Routing Protocol) proposed in literature (Zygmunt J. Hass, “A New Routing Protocol for the Reconfigurable Wireless Networks”, in IEEE ICUPC '97, pp. 652-566, 1997). ) Introduces the concept of zone radius, where each node manages routes to nodes within the zone radius in a table-driven manner, and routes to nodes outside the zone radius are constructed on demand. For this reason, when the zone radius is set so that the specific node is within the zone radius, in the communication mode in which the specific node is frequently accessed from many nodes, as in the table driving method, an unnecessary route is constructed. Processing and communication occur. On the other hand, if the zone radius is set so that the specific node is outside the zone radius, a delay always occurs in a communication mode in which the specific node is frequently accessed from many nodes, as in the on-demand method.
[0006]
[Problems to be solved by the invention]
By the way, since a specific node serving as a gateway to an external network is frequently accessed from many nodes, the above problem occurs in such a communication form.
[0007]
That is, in the conventional table driving method, it is necessary to have route information for all nodes. When a topology change is detected due to a failure or the like, the detected node notifies all nodes to reconstruct the route. Needs to reconstruct the route based on this notification. In this way, maintaining the routes of all nodes means that in a system with a high access frequency to “specific nodes”, route construction processing other than “specific nodes”, processing for maintaining routes, these processing The wireless band associated with is wasted.
[0008]
In addition, the conventional on-demand method does not need to maintain a route, but a delay always occurs in a communication mode in which a specific node is frequently accessed from many nodes. In particular, each node only sends a route maintenance packet to an adjacent node, does not receive a route maintenance packet from the adjacent node, and cannot check whether the adjacent node has received the route maintenance packet. It was. Therefore, if it is possible to receive from the adjacent node but cannot transmit to the adjacent node, it is not possible to cope with it, and it is necessary to recognize the change in topology by another means, so it is necessary to recognize the change in topology and rebuild the route. It took time and the delay was large.
[0009]
Further, the conventional hybrid method switches between the table driving method and the on-demand method depending on the distance from the node, and thus has the respective problems of the conventional table driving method and the on-demand method.
[0010]
In a wireless network in which each node frequently accesses a specific node and there is little access between other nodes, a delay is reduced with respect to maintaining a route with the specific node, and a wasteful route construction process, It is an object of the present invention to provide a route maintenance method, a route maintenance system, and a wireless node device that can eliminate processing for maintaining a route, waste of a radio band associated with these processes.
[0011]
[Means for Solving the Problems]
According to the route maintenance method, route maintenance system, and wireless node device of the present invention, a specific wireless node transmits a route maintenance packet to each wireless node or each wireless node transmits a route maintenance packet to a specific wireless node. The node that receives the packet transmits a route maintenance reply packet, and the transmission node of the route maintenance packet maintains the route between the specific wireless node and each wireless node by confirming reception of the route maintenance reply packet.
[0012]
In the inventions described in claims 1, 7, 13, and 14, the specific wireless node holds a route management table in which “destination”, “adjacent”, “hop count”, and “route check” are stored, and “adjacent” A route maintenance packet that describes all the “destination” with the same as the destination is generated, this route maintenance packet is unicast to the wireless node corresponding to “adjacent”, and the route maintenance reply sent by the wireless node that received the route maintenance packet Confirm that the packet is received and each route is maintained.
[0013]
The wireless node maintains a route management table in which “destination”, “adjacent”, and “hop count” are stored. When a route maintenance packet is received, all the “destination” described in the route maintenance packet are extracted. , Generate a route maintenance packet describing all “destinations” that have the same “adjacent” in the route management table for the “destination”, and unicast the route maintenance packet to the wireless node corresponding to “adjacent” When a relay does not need to relay a route maintenance packet or when a predetermined time has elapsed after relaying the route maintenance packet, a route maintenance reply packet to a specific wireless node is generated and unicasted, and the route maintenance reply packet Is received, information indicating that the node has received the route maintenance packet is added to the route maintenance reply packet, and unicast relay is performed.
[0014]
In the inventions according to claims 2, 8, 13, and 14, the specific wireless node holds a route management table in which “destination”, “adjacent”, “hop count”, and “route check” are stored, and a route maintenance packet This route maintenance packet is broadcast to adjacent wireless nodes in the range where direct communication can be performed via the wireless link, and the route maintenance reply packet transmitted from the wireless node that has received the route maintenance packet is received, and each route is Make sure it is maintained.
[0015]
The wireless node maintains a route management table that describes “destination”, “adjacent”, and “hop count”. When a route maintenance packet is received, the wireless node confirms its source and receives the first route from a specific wireless node. If it is a maintenance packet, a route maintenance reply packet to the specific wireless node is generated and unicasted.
[0016]
In the third, ninth, thirteen, and fourteenth aspects of the present invention, the wireless node holds a route management table in which “destination”, “adjacent”, “hop count”, and “route check” are stored at a predetermined time. A route maintenance packet is generated and unicasted to a specific wireless node. When the route maintenance packet is received, information that can uniquely identify the own node is added to the route maintenance packet and relayed to the specific wireless node. When the transmitted route maintenance reply packet is received, the route to the specific wireless node is recorded in the route management table. If this route maintenance reply packet is addressed to its own node, it is discarded, and if it is addressed to another wireless node, it is discarded. Relay.
[0017]
The specific wireless node holds a route management table in which “destination”, “adjacent”, and “hop count” are recorded, and when a route maintenance packet is received, takes out information of the wireless node that has passed from this route maintenance packet, The route information is recorded in the route management table, the transmission source is extracted from the route maintenance packet, and the route maintenance reply packet is unicast to the transmission source.
[0018]
In the inventions according to the fourth, tenth, thirteenth and fourteenth aspects, the wireless node holds a route management table in which “destination”, “adjacent”, “hop count”, and “route check” are stored, at a predetermined time. When a route maintenance packet is generated and broadcast to an adjacent wireless node, when a route maintenance packet is received, a route maintenance reply packet is unicast to the transmission source, and when a route maintenance reply packet cannot be received from an adjacent wireless node The information corresponding to the adjacent wireless node is deleted from the route management table, and the route is reconstructed.
[0019]
In the fifth, eleventh and thirteenth aspects of the invention, the specific wireless node receives a packet from the wireless node within the predetermined time for a predetermined time required for reception of the packet used for the “route check” described in the route management table. If it is not sent, the route to the wireless node is reconstructed.
[0020]
In the inventions of claims 6, 12, and 14, when a wireless node transmits a packet to another wireless node through a route based on the route management table and there is no reply packet even after a predetermined time has elapsed, Of the “adjacent” listed in the management table, delete the route that is the same as the “adjacent” of the route that does not have a reply packet, and broadcast the failure notification packet to the adjacent wireless node in the range that can be directly communicated with the wireless link. , Rebuild the route to and from the wireless node.
[0021]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 shows a configuration example of a wireless network. In the figure, a specific node serving as a gateway to an external network is defined as an AP (access point), nodes WR1 and WR2 are connected to the specific node AP via a wireless link, and a node WR1 and a node WR4 are further connected to each other. WR4 and nodes WR3, WR5, WR6, and WR7 are connected, node WR2 and node WR5 are connected, nodes WR5 and WR7 are connected, and node WR3 and node WR6 are connected.
[0022]
The route management table of each node includes “destination (destination node)”, “adjacent (adjacent node to transfer the packet to the destination node)”, “hop count (route necessary for the packet to reach the destination node) Metric) ”. For example, in the specific node AP, the adjacent WR1 and the number of hops are 1 for the destination WR1, and the adjacent WR1 and the number of hops are 2 for the destination WR3. In the route management table of each of the nodes WR1 to WR, the adjacency and the number of hops when the specific node AP is the destination are set.
[0023]
In the present invention, the specific node AP transmits a route maintenance packet to each of the nodes WR1 to WR7, or each of the nodes WR1 to WR7 transmits the route maintenance packet to the specific node AP. The transmission node of the route maintenance packet maintains the route between the specific node AP and each of the nodes WR1 to WR7 based on the receipt confirmation of the route maintenance reply packet. Hereinafter, an AP-driven embodiment in which a specific node AP transmits a path maintenance packet to each of the nodes WR1 to WR7, and a WR-driven embodiment in which each of the nodes WR1 to WR7 transmits a path maintenance packet to the specific node AP Will be described.
[0024]
<AP-led embodiment: claims 1, 2, 5, 7, 8, 11>
FIG. 2 shows a route management table of a specific node AP in the case of AP initiative. The configuration of the wireless network and the route management table of the nodes WR1 to WR7 are the same as those shown in FIG.
[0025]
The route management table of the specific node AP includes a “route check” column in addition to “destination”, “adjacent”, and “hop count”. “Route check” indicates whether or not each node has received a route maintenance packet within a predetermined time based on the receipt confirmation of the route maintenance reply packet.
[0026]
(AP-first control sequence: claims 1 and 7)
The first control sequence led by the AP will be described with reference to FIGS. FIG. 3 shows a control sequence when a route maintenance packet is transmitted in the specific node AP. FIG. 4 shows a control sequence in the nodes WR1 to WR7. FIG. 5 shows a control sequence when a route maintenance reply packet is received in the specific node AP. FIG. 6 shows a configuration example of the route maintenance packet. FIG. 7 shows a configuration example of the route maintenance reply packet.
[0027]
(Control sequence for route maintenance packet transmission in specific node AP)
2 and 3, the specific node AP checks whether or not there is an unprocessed set including “destination”, “adjacent”, and “hop count” in the route management table. Generate a maintenance packet. As shown in FIG. 6, the route maintenance packet includes a packet type, a destination, an adjacency, the number of hops, a transmission source, a transmission source, and an actual destination. The actual destination means a node that should receive the route maintenance packet.
[0028]
Here, a set of “destination”, “adjacent”, and “hop count” in the route management table of the specific node AP is described as [destination, adjacency, hop count]. First, as shown in FIG. 2, there is a set of [WR1, WR1, 1], and it can be seen that the adjacent node is WR1 (s1). Here, since there is no route maintenance packet to the adjacent WR1 (this is the first route maintenance packet) (s2), a route maintenance packet (packet type: route maintenance, destination: WR1, neighbor) as shown in FIG. 6 (1) : WR1, hop count: 1, source: AP, source: AP, actual destination: WR1) is generated (s3). Next, for [WR2, WR2, 1], similarly, a route maintenance packet to adjacent WR2 (packet type: route maintenance, destination: WR2, neighbor: WR2, number of hops: 1, source: AP, source: AP (actual destination: WR2) is generated (s3).
[0029]
Next, for [WR3, WR1, 3], since there is already a route maintenance packet to the adjacent WR1 (s2), WR3 is added to the actual destination of this route maintenance packet (s4, FIG. 6 (2)). . For [WR4, WR1, 2], there is already a route maintenance packet to the adjacent WR1 (s2), so WR4 is added to the actual destination of this route maintenance packet (s4, FIG. 6 (3)). Similarly, a route maintenance packet is generated for each of [WR5, WR2, 2], [WR6, WR1, 3], and [WR7, WR2, 3] by adding actual destinations.
[0030]
As described above, two types of route maintenance packets are generated in the WR1 direction and the WR2 direction. The route maintenance packet in the WR1 direction is packet type: route maintenance, destination: WR1, neighbor: WR1, hop count: 1, source: AP, source: AP, actual destination: WR1, WR3, WR4, WR6 ( Fig. 6 (4)). On the other hand, the route maintenance packet in the WR2 direction is packet type: route maintenance, destination: WR2, neighbor: WR2, hop count: 1, source: AP, source: AP, actual destination: WR2, WR5, WR7 ( Fig. 6 (5)). Then, the route check of each route is performed during route investigation, and two types of route maintenance packets are transmitted to the nodes WR1 and WR2.
[0031]
(Control sequence in nodes WR1 to WR7)
2 and 4, when the node WR1 receives the route maintenance packet (s11), it is checked whether or not there is a node other than the own node at the actual destination (s12). Here, since there are nodes WR3, WR4, and WR6 other than the own node as actual destinations, a route maintenance packet is generated according to the route management table of node WR1 (s13 to s17). For the real destination WR4, it can be seen from [WR4, WR4, 1] of the route management table that the neighbor is WR4, and there is no route maintenance packet for the adjacent WR4. Type: Route Maintenance, Destination: WR4, Adjacent: WR4, Number of Hops: 1, Real Destination: WR4). As for [WR6, WR4, 2], since there is already a route maintenance packet to the adjacent WR4, WR6 is added to the actual destination of this route maintenance packet. The same applies to [WR3, WR4, 2]. As a result, the node WR1 transmits a route maintenance packet having the packet type: route maintenance, destination: WR4, neighbor: WR4, number of hops: 1, real destination: WR3, WR4, WR6 to the node WR4 (s18). It waits for reception of a maintenance reply packet (s19).
[0032]
When the node WR4 receives the route maintenance packet from the node WR1, the node WR4 processes the same as the node WR1. That is, a route maintenance packet with packet type: route maintenance, destination: WR3, neighbor: WR3, number of hops: 1, actual destination: WR3 is transmitted to the node WR3 (s18). Further, a route maintenance packet with packet type: route maintenance, destination: WR6, neighbor: WR6, hop count: 1, real destination: WR6 is transmitted to the node WR6 (s18). Then, it waits for reception of a route maintenance reply packet (s19).
[0033]
Upon receiving the route maintenance packet from the node WR4, the node WR3 generates a route maintenance reply packet for the specific node AP that is the transmission source because there is no node other than its own node at the actual destination (s12), and sends it to the node WR4. Reply (s23, s24). As shown in FIG. 7, the route maintenance reply packet includes a packet type, a destination, an adjacency, the number of hops, a transmission source, a transmission source, and a normal route. This route maintenance reply packet is set to packet type: route maintenance reply, destination: AP, neighbor: WR4, number of hops: 3, transmission source: WR3, transmission source: WR3, normal route: WR3 (FIG. 7 (1) )). The node WR6 performs the same process as the node WR3.
[0034]
Node WR4 is waiting to receive a route maintenance reply packet in s19. When node WR4 receives a route maintenance reply packet from nodes WR3 and WR6 (s20, s21), a new route maintenance is performed by adding its own node to these normal routes. A reply packet is generated, and the received route maintenance reply packet is discarded. The new route maintenance reply packet is set to packet type: route maintenance reply, destination: AP, neighbor: WR1, hop count: 2, source: WR4, source: WR4, normal route: WR3, WR4, WR6 ( In FIG. 7 (2)), it is transmitted to the node WR1 (s22).
[0035]
Node WR1 performs the same processing as node WR4. The route maintenance reply packet is set as packet type: route maintenance reply, destination: AP, neighbor: AP, hop count: 1, source: WR1, source: WR1, normal route: WR1, WR3, WR4, WR6 ( FIG. 7 (3)) is transmitted to the specific node AP.
[0036]
Similarly, for the routes of the nodes WR2, WR5 and WR7, the packet type: route maintenance reply from the node WR2, destination: AP, neighbor: AP, hop count: 1, source: WR2, source: WR2, normal route: WR2 , WR5, WR7, the route maintenance reply packets set are transmitted to the specific node AP.
[0037]
(Control sequence when receiving route maintenance reply packet in specific node AP)
2 and 5, the specific node AP is waiting to receive a path maintenance reply packet after transmitting two types of path maintenance packets to the nodes WR1 and WR2 (s31), and the path from the node WR1 within a predetermined time. When the maintenance reply packet is received (s32), the normal routes WR1, WR3, WR4, WR6 are extracted from the route maintenance reply packet, and each route check corresponding to the destinations WR1, WR3, WR4, WR6 in the route management table is being investigated. Change to OK (s33). The route maintenance reply packet from the node WR2 is similarly processed (s34).
[0038]
(Processing when a failure occurs between WR5 and WR7: Claims 5 and 11)
In FIG. 2, the node WR5 receives the route maintenance packet from the node WR2, transfers the route maintenance packet to the node WR7 in accordance with the above-described procedure, and enters the state of waiting for reception of the route maintenance reply packet (FIG. 4, s19). Here, if the route maintenance reply packet is not received from the node WR7 even after the predetermined time has elapsed, the node WR5 generates a route maintenance reply packet. This route maintenance reply packet is set to packet type: route maintenance reply, destination: AP, neighbor: WR2, number of hops: 2, transmission source: WR5, transmission source: WR5, normal route: WR5 and transmitted to node WR2. The
[0039]
The node WR2 sets the packet type of the received route maintenance reply packet: route maintenance reply, destination: AP, neighbor: AP, hop count: 1, source: WR2, source: WR2, normal route: WR2, WR5. To the specific node AP.
[0040]
The specific node AP extracts the normal routes WR2 and WR5 from the received route maintenance reply packet, and changes each route check corresponding to the destinations WR2 and WR5 in the route management table to “OK” during the route investigation. When a predetermined time elapses after transmitting the route maintenance packet, the specific node AP checks each route check in the route management table, and reconstructs the route for the route of the destination WR7 under route investigation (FIG. 5, s35). ).
[0041]
(AP-led second control sequence: claims 2 and 8)
The second control sequence led by the AP will be described with reference to FIGS. FIG. 8 shows a control sequence of the specific node AP. FIG. 9 shows a control sequence in the nodes WR1 to WR7. FIG. 10 shows a configuration example of the route maintenance packet. FIG. 11 shows a configuration example of a route maintenance reply packet.
[0042]
As shown in FIG. 10, the route maintenance packet includes a packet type, a destination, an adjacency, the number of hops, a transmission source, and a transmission source. Here, packet type: route maintenance, destination: broadcast, neighbor: broadcast, number of hops: maximum number of hops in route management table (3), transmission source: AP, transmission source: AP are set.
[0043]
2 and 8, the specific node AP sets the route check for each route in the route management table during the route investigation (s41), and broadcasts the route maintenance packet shown in FIG. 10 with the transmission source as the AP (s42). Then, it waits for reception of a route maintenance reply packet (s43).
[0044]
In FIG. 9, the node WR1 is in a state of waiting for a route maintenance packet (s51), and receives a route maintenance packet from a specific node AP within a predetermined time (s52), and confirms the source of the route maintenance packet. (S53). If the source of the route maintenance packet in the route management table of the node WR1 matches the source of the route maintenance packet (both are APs here) (s54), the route maintenance is performed for the specific node AP that is the source. A reply packet is transmitted (s55). As shown in FIG. 11, the route maintenance reply packet includes a packet type, a destination, an adjacency, the number of hops, a transmission source, and a transmission source. Here, packet type: route maintenance reply, destination: AP, neighbor: AP, hop count: 1, source: WR1, and source: WR1 are set.
[0045]
Further, the node WR1 broadcasts the route maintenance packet set in the transmission source and the own node (s56). This route maintenance packet is set to packet type: route maintenance, destination: broadcast, neighbor: broadcast, hop count: 2, source: AP, source: WR1.
[0046]
Other nodes that have received this path maintenance packet process the same, and broadcast the path maintenance packet while transmitting a path maintenance reply packet to the source. However, the nodes WR3, WR6, and WR7 do not broadcast the route maintenance packet because the number of hops of the route maintenance packet from the node WR4 is 1. Each node that has received the route maintenance reply packet sequentially forwards the route maintenance reply packet to the adjacent node with respect to the destination AP in the route management table, and finally reaches the specific node AP.
[0047]
In FIG. 8, the specific node AP extracts the transmission source WR1 of the path maintenance reply packet received within a predetermined time, and changes each path check corresponding to the destination WR1 in the path management table from being investigated to OK (s44, s45). The same applies to route maintenance reply packets from other transmission sources. The specific node AP checks each route check in the route management table when a predetermined time has elapsed after broadcasting the route maintenance packet, and reconstructs the route for the destination route under route investigation (s46).
[0048]
(Processing when a failure occurs between WR5 and WR7: Claims 5 and 11)
If a failure occurs between WR5 and WR7, the route maintenance reply packet is returned to the specific node AP for the nodes WR1 to WR6, so that each route check in the route management table of the specific node AP is OK. It has become. However, since the route maintenance reply packet from the node WR7 does not reach the specific node AP, the corresponding route check remains under route investigation. Therefore, when a predetermined time elapses after the route maintenance packet is broadcast, the specific node AP checks each route check in the route management table, and reconstructs the route for the route of the destination WR7 under route investigation.
[0049]
<WR-led embodiment: claims 3, 4, 6, 9, 10, 12>
FIG. 12 shows a route management table of the nodes WR1 to WR7 in the case of WR initiative. The configuration of the wireless network and the route management table of the specific node AP are the same as those shown in FIG.
[0050]
The route management table of the nodes WR1 to WR7 has a “route check” column in addition to “destination”, “adjacent”, and “hop count”. “Route check” indicates whether or not each node has received a route maintenance packet within a predetermined time based on the receipt confirmation of the route maintenance reply packet.
[0051]
(WR-led first control sequence: claims 3 and 9)
A first control sequence led by WR will be described with reference to FIG. FIG. 13 shows a control sequence in the nodes WR1 to WR7. Although the case of the node WR6 will be described here, the same applies to the other nodes WR3 and WR7.
[0052]
12 and 13, since the node WR6 does not receive the route maintenance packet addressed to the specific node AP even after a predetermined time has elapsed (s61), the node WR6 generates a route maintenance packet addressed to the specific node AP, based on the route management table. Transmit to the adjacent node WR4 (s62). This route maintenance packet, packet type: route maintenance, destination: AP, neighbor: WR4, hop count: 3, transmission source: WR6, transmission source: WR6 are set. Then, it waits for reception of a route maintenance reply packet (s63).
[0053]
When the node WR4 receives a route maintenance packet addressed to the specific node AP from the node WR6 within a predetermined time (s61), the node WR4 searches for a neighboring node for the destination AP from the route management table, and transfers the route maintenance packet to the node WR1 here. (S64). The setting of this route maintenance packet is changed to packet type: route maintenance, destination: AP, neighbor: WR1, hop count: 2, transmission source: WR6, transmission source: WR4. Then, it waits for reception of a route maintenance reply packet (s65).
[0054]
When the node WR1 receives a route maintenance packet addressed to the specific node AP from the node WR4 within a predetermined time, the node WR1 processes the same as the node WR4 and transfers the route maintenance packet to the specific node AP. The setting of this route maintenance packet is changed to packet type: route maintenance, destination: AP, neighbor: AP, hop count: 1, transmission source: WR6, transmission source: WR1.
[0055]
When the specific node AP receives the path maintenance packet from the node WR1, the specific node AP extracts the transmission source WR6 from the path maintenance packet, and transmits the path maintenance reply packet to the adjacent node WR1 obtained from the path management table with the node WR6 as the destination. This route maintenance reply packet is set to packet type: route maintenance reply, destination: WR6, neighbor: WR1, hop count: 3, transmission source: AP, transmission source: AP.
[0056]
The node WR1 is waiting to receive a route maintenance reply packet (s65), and if a route maintenance reply packet addressed to the node WR6 is received from the specific node AP within a predetermined time (s66), the route to the specific node AP is normal. And the route check of the destination AP in the route management table is OK. The node WR1 then forwards the route maintenance reply packet to the adjacent node WR4 obtained from the route management table with the node WR6 as the destination (s67). The setting of this route maintenance reply packet is changed to packet type: route maintenance reply, destination: WR6, neighbor: WR4, hop count: 2, transmission source: AP, transmission source: WR1.
[0057]
When the node WR4 receives the route maintenance reply packet addressed to the node WR6 from the node WR1 within a predetermined time, the node WR4 recognizes that the route to the specific node AP is normal and checks the route of the destination AP and WR1 in the route management table. OK. Then, processing is performed in the same manner as the node WR1, and the path maintenance reply packet is transferred to the node WR6. The setting of this route maintenance reply packet is changed to packet type: route maintenance, destination: WR6, neighbor: WR6, number of hops: 1, source: AP, source: WR4.
[0058]
The node WR6 is waiting to receive the route maintenance reply packet (s63), and if it receives the route maintenance reply packet from the node WR4 within a predetermined time (s68), it recognizes that the route to the specific node AP is normal. Then, the route check of the destination AP and WR4 in the route management table is OK. At this time, since the destination is the own node, the processing is terminated without transferring the route maintenance reply packet.
[0059]
(Processing when a failure occurs between WR2 and WR5: Claims 6 and 12)
12 and 13, the node WR5 receives the path maintenance packet from the node WR7, transfers the path maintenance packet to the node WR2 in accordance with the above-described procedure, and enters the path maintenance waiting state for the path maintenance reply packet (s65). Here, the route maintenance reply packet is not replied from the node WR2 even after the predetermined time has passed (s66), and when the predetermined time has passed without receiving the failure notification packet described later (s69, s70), or the received failure notification If the packet is transmitted from the direction of the specific node AP (destination is adjacent to the AP) (s71), the failure notification packet is broadcast for one hop (s72). This failure notification packet is set to packet type: failure information, destination: broadcast, neighbor: broadcast, hop count: 1, transmission source: WR5, transmission source: WR5. Then, the setting of the route with the adjacent WR2 in the route management table is deleted (s73).
[0060]
The node WR7 receives the failure notification packet instead of the route maintenance reply packet, and when both the neighbor to the destination AP in the route management table and the transmission source of the failure notification packet match in WR5 (s75), both packets are not received. When a predetermined time has elapsed (s76), the failure notification packet is broadcast for one hop (s72), and the setting of the route to the specific node AP is deleted (s73). This failure notification packet is set to packet type: failure information, destination: broadcast, neighbor: broadcast, hop count: 1, transmission source: WR5, transmission source: WR7.
[0061]
Further, when receiving the failure notification packet, the node WR4 ignores this failure notification packet because the adjacent to the destination AP in the path management table is WR1 and the source of the failure notification packet does not match WR5.
[0062]
As described above, the nodes WR5 and WR7 reconstruct the route to the specific node AP (s74). As the method, the above-described method according to the present invention or a conventional route construction method such as AODV may be used.
[0063]
(WR-led second control sequence: claims 4 and 10)
A second control sequence led by WR will be described with reference to FIG. FIG. 14 shows a control sequence in the nodes WR1 to WR7. Here, the case of the node WR5 will be described, but the same applies to other nodes.
[0064]
In FIGS. 12 and 14 (1), the node WR5 makes all route checks in the route management table under route investigation at a certain time (s81). Then, a route maintenance packet is generated and broadcast for one hop (s82). This route maintenance packet is set to packet type: route maintenance, destination: broadcast, neighbor: broadcast, hop count: 1, source: WR5, source: WR5. Then, it waits for reception of a route maintenance reply packet (s83).
[0065]
In FIG. 14 (2), when the node WR2 receives the path maintenance packet (s91), it generates a path maintenance reply packet and transmits it to the node WR5 that transmitted the path maintenance packet (s92). This route maintenance reply packet is set to packet type: route maintenance reply, destination: WR5, neighbor: WR5, hop count: 1, transmission source: WR2, transmission source: WR2. The same applies to the nodes WR4 and WR7 that receive the route maintenance packet.
[0066]
In FIG. 14 (1), when the node WR5 waiting to receive the path maintenance reply packet receives the path maintenance reply packet from the nodes WR2, WR4 and WR7 within a predetermined time (s84), the adjacent WR2 in the path management table is received. The route check is set to OK for WR4 and WR7 (s85).
[0067]
(Processing when a failure occurs between WR2 and WR5: Claims 6 and 12)
In FIG. 12 and FIG. 14 (1), the node WR5 does not receive the route maintenance reply packet from the node WR2 within a predetermined time (s84) and indicates that a failure has occurred with the node WR2 in the route management table. Recognizing that the route check is still in the route investigation, the adjacent route management table changes the route of WR2 during route construction (s86). Since the route being constructed is in the direction of the specific node AP, the failure notification packet is broadcast for one hop (s87). This failure notification packet is set to packet type: failure information, destination: broadcast, neighbor: broadcast, hop count: 1, transmission source: WR5, transmission source: WR5. As a result, the node WR5 reconstructs the route to the specific node AP (s88), and the method according to the present invention described above or a conventional route construction method such as AODV may be used as the method.
[0068]
In FIG. 14 (3), the node WR7 receives the failure notification packet (s93), and when both the neighbor to the destination AP in the route management table and the transmission source of the failure notification packet match in WR5 (s94), the route management The route adjacent to the WR2 is changed during route construction (s95), and the failure notification packet is broadcast for one hop (s96). This failure notification packet is set to packet type: failure information, destination: broadcast, neighbor: broadcast, hop count: 1, transmission source: WR5, transmission source: WR7. As a result, the node WR7 reconstructs the route to the specific node AP (s97), and the method according to the present invention described above or the conventional route construction method such as AODV may be used as the method.
[0069]
Further, when receiving the failure notification packet, the node WR4 ignores this failure notification packet because the adjacent to the destination AP in the path management table is WR1 and the source of the failure notification packet does not match WR5.
[0070]
【The invention's effect】
As described above, according to the present invention, in a wireless network in which each wireless node frequently accesses a specific wireless node and access between other wireless nodes is small, the specific wireless node A wireless node transmits a route maintenance packet to a specific wireless node by unicast, a node that receives the route maintenance packet transmits a route maintenance reply packet by unicast, and a transmission node of the route maintenance packet receives a route maintenance reply packet By the confirmation, a route between the specific wireless node and each wireless node can be maintained. Therefore, for route maintenance with a specific wireless node, delay associated with route maintenance can be reduced, and wasteful route construction processing, processing for route maintenance, and waste of radio bandwidth associated with these processing can be omitted. it can.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a configuration example of a wireless network.
FIG. 2 is a diagram showing a route management table of a specific node AP in the case of AP initiative.
FIG. 3 is a diagram showing a control sequence when a route maintenance packet is transmitted in a specific node AP.
FIG. 4 is a diagram showing a control sequence in nodes WR1 to WR7.
FIG. 5 is a diagram showing a control sequence when a route maintenance reply packet is received by a specific node AP.
FIG. 6 is a diagram showing a configuration example of a route maintenance packet.
FIG. 7 is a diagram showing a configuration example of a route maintenance reply packet.
FIG. 8 is a diagram showing a control sequence of a specific node AP.
FIG. 9 is a diagram showing a control sequence in nodes WR1 to WR7.
FIG. 10 is a diagram showing a configuration example of a route maintenance packet.
FIG. 11 is a diagram showing a configuration example of a route maintenance reply packet.
FIG. 12 is a diagram showing a path management table of nodes WR1 to WR7 in the case of WR initiative.
FIG. 13 is a diagram showing a control sequence in nodes WR1 to WR7.
FIG. 14 is a diagram showing a control sequence in nodes WR1 to WR7.
[Explanation of symbols]
AP specific wireless node (specific node)
WR wireless node (node)

Claims

In a wireless network composed of a wireless node that transmits and receives packets via a wireless interface and a specific wireless node that serves as a gateway to an external network,
The specific wireless node includes a destination node (hereinafter referred to as “destination”), an adjacent node to which a packet is transferred to reach the destination node (hereinafter referred to as “adjacent”), and a metric of a route required until the packet reaches the destination node ( (Hereinafter referred to as “the number of hops”), a route management table in which each wireless node receives a packet within a predetermined time (hereinafter referred to as “route check”) is maintained, and “adjacent” is the same. A route maintenance packet describing all the “destinations” to be generated is generated, this route maintenance packet is unicast to the wireless node corresponding to “adjacent”, and a route maintenance reply packet transmitted by the wireless node that has received the route maintenance packet is transmitted. Make sure that each route is received,
The wireless node maintains a route management table that describes “destination”, “adjacent”, and “hop count”, and when the route maintenance packet is received, all “destination” described in the route maintenance packet. And generate a route maintenance packet in which all “destinations” having the same “adjacent” in the route management table with respect to the “destination” are generated, and the route maintenance packet is defined as a radio corresponding to the “adjacent”. When a unicast relay is performed to a node and it is not necessary to relay the route maintenance packet or when a predetermined time has elapsed after relaying the route maintenance packet, a route maintenance reply packet to the specific wireless node is generated and unicast is generated. When the route maintenance reply packet is received, information indicating that the node has received the route maintenance packet is added to the route maintenance reply packet to add a unicast. Collected by relay,
A method for maintaining a route in a wireless network, characterized by constantly confirming that a packet arrives between the wireless node and the specific wireless node.

In a wireless network composed of a wireless node that transmits and receives packets via a wireless interface and a specific wireless node that serves as a gateway to an external network,
The specific wireless node includes a destination node (hereinafter referred to as “destination”), an adjacent node to which a packet is transferred to reach the destination node (hereinafter referred to as “adjacent”), and a metric of a route required until the packet reaches the destination node ( (Hereinafter referred to as “the number of hops”), a route management table in which a value indicating whether or not each wireless node has received a packet within a predetermined time (hereinafter referred to as “route check”) is maintained, and a route maintenance packet is generated. The route maintenance packet is broadcast to an adjacent wireless node in a range that can be directly communicated with the wireless link, and the route maintenance reply packet transmitted from the wireless node that has received the route maintenance packet is received to maintain each route. Make sure that
The wireless node maintains a route management table that describes “destination”, “adjacent”, and “hop count”, and when receiving the route maintenance packet, confirms the source of the route management packet and sends it from the specific wireless node. If it is the first route maintenance packet, a route maintenance reply packet to the specific wireless node is generated and unicasted,
A method for maintaining a route in a wireless network, characterized by constantly confirming that a packet arrives between the wireless node and the specific wireless node.

In a wireless network composed of a wireless node that transmits and receives packets via a wireless interface and a specific wireless node that serves as a gateway to an external network,
The wireless node includes a destination node (hereinafter referred to as “destination”), an adjacent node (hereinafter referred to as “adjacent”) for transferring a packet to the destination node, and a metric (hereinafter referred to as “route”) required for the packet to reach the destination node. It holds a route management table with a value indicating whether or not each wireless node has received a packet within a predetermined time (hereinafter referred to as “route check”), and maintains a route maintenance packet at a predetermined time. Is generated and unicasted to the specific radio node, and when the route maintenance packet is received, information that can uniquely identify the own node is added to the route maintenance packet and relayed to the specific radio node, and the specific radio node When the route maintenance reply packet transmitted by the receiver is received, the route to the specific wireless node is recorded in the route management table, and the route maintenance reply packet If addressed to discard it, if the other wireless nodes addressed relays it,
The specific wireless node maintains a route management table in which “destination”, “adjacent”, and “hop count” are recorded. When the route maintenance packet is received, information on the wireless node that has passed from the route maintenance packet is stored. Take out, record the route information in the route management table, take out the transmission source from the route maintenance packet, unicast the route maintenance reply packet to the transmission source,
A method for maintaining a route in a wireless network, characterized by constantly confirming that a packet arrives between the wireless node and the specific wireless node.

In a wireless network composed of a wireless node that transmits and receives packets via a wireless interface and a specific wireless node that serves as a gateway to an external network,
The wireless node includes a destination node (hereinafter referred to as “destination”), an adjacent node (hereinafter referred to as “adjacent”) for transferring a packet to the destination node, and a metric (hereinafter referred to as “route”) required for the packet to reach the destination node. It holds a route management table with a value indicating whether or not each wireless node has received a packet within a predetermined time (hereinafter referred to as “route check”), and maintains a route maintenance packet at a predetermined time. When the route maintenance packet is received, the route maintenance reply packet is unicast to the transmission source when the route maintenance packet is received, and when the route maintenance reply packet cannot be received from the adjacent wireless node, A wireless network characterized by deleting information corresponding to the adjacent wireless node from the route management table and reconstructing the route. Route maintenance method in.

In the route maintenance method in the wireless network according to claim 1 or 2,
When the specific wireless node does not send a packet from the wireless node within the predetermined time for the predetermined time required for receiving the packet used for the “route check” described in the route management table, the specific wireless node A route maintenance method in a wireless network, characterized by reconstructing a route between nodes.

In the route maintenance method in the wireless network according to claim 3 or 4,
If the wireless node does not receive a reply packet even after a predetermined time has elapsed since the packet was transmitted to the other wireless node through the route based on the route management table, the “adjacent” of the route management table Among them, the same route as the “adjacent” of the route without the return packet is deleted, the failure notification packet is broadcast to the adjacent wireless node in the range where the wireless link can be directly communicated, and the route between the wireless node is A route maintenance method in a wireless network, characterized by reconstructing.

In a wireless network composed of a wireless node that transmits and receives packets via a wireless interface and a specific wireless node that serves as a gateway to an external network,
The specific wireless node includes a destination node (hereinafter referred to as “destination”), an adjacent node to which a packet is transferred to reach the destination node (hereinafter referred to as “adjacent”), and a metric of a route required until the packet reaches the destination node ( (Hereinafter referred to as “the number of hops”), a route management table in which each wireless node receives a packet within a predetermined time (hereinafter referred to as “route check”) is maintained, and “adjacent” is the same. Means for generating a route maintenance packet describing all the “destinations” to be transmitted, means for unicasting the route maintenance packet to a wireless node corresponding to “adjacent”, and a route transmitted by the wireless node that has received the route maintenance packet Receiving a maintenance reply packet and confirming that each route is maintained,
The wireless node maintains a route management table that describes “destination”, “adjacent”, and “hop count”, and when the route maintenance packet is received, all “destination” described in the route maintenance packet. Means for generating a route maintenance packet describing all “destinations” that have the same “adjacent” in the route management table with respect to the “destination”, and this route maintenance packet corresponds to the “adjacent” Unicast relay means to a wireless node to generate a route maintenance reply packet to the specific wireless node when the route maintenance packet does not need to be relayed or when a predetermined time has elapsed after relaying the route maintenance packet Unicast transmission means, and when the route maintenance reply packet is received, indicates that the route maintenance packet has been received by the own node in the route maintenance reply packet. And means for unicast relay by adding information,
A route maintenance system in a wireless network, which constantly confirms that a packet arrives between the wireless node and the specific wireless node.

In a wireless network composed of a wireless node that transmits and receives packets via a wireless interface and a specific wireless node that serves as a gateway to an external network,
The specific wireless node includes a destination node (hereinafter referred to as “destination”), an adjacent node to which a packet is transferred to reach the destination node (hereinafter referred to as “adjacent”), and a metric of a route required until the packet reaches the destination node ( (Hereinafter referred to as “the number of hops”), and a route management table in which a value indicating whether or not each wireless node has received the packet within a predetermined time (hereinafter referred to as “route check”) is maintained, and a route maintenance packet is generated. Means, a means for broadcasting this route maintenance packet to an adjacent wireless node in a range where it can be directly communicated with a wireless link, and a route maintenance reply packet transmitted from the wireless node that has received the route maintenance packet, A means of confirming that the route is maintained,
The wireless node maintains a route management table that describes “destination”, “adjacent”, and “hop count”, and when receiving the route maintenance packet, confirms the source of the route management packet and sends it from the specific wireless node. If it is the first route maintenance packet, comprising a means for generating a route maintenance reply packet to the specific wireless node and unicasting,
A route maintenance system in a wireless network, which constantly confirms that a packet arrives between the wireless node and the specific wireless node.

In a wireless network composed of a wireless node that transmits and receives packets via a wireless interface and a specific wireless node that serves as a gateway to an external network,
The wireless node includes a destination node (hereinafter referred to as “destination”), an adjacent node (hereinafter referred to as “adjacent”) for transferring a packet to the destination node, and a metric (hereinafter referred to as “route”) required for the packet to reach the destination node. It holds a route management table with a value indicating whether or not each wireless node has received a packet within a predetermined time (hereinafter referred to as “route check”), and maintains a route maintenance packet at a predetermined time. Means for generating and unicasting to the specific wireless node; means for adding to the path maintenance packet information that can uniquely identify the own node when the path maintenance packet is received; and relaying to the specific radio node; Means for recording a route to the specific wireless node in the route management table when receiving a route maintenance reply packet transmitted by the specific wireless node; When lifting return packet addressed to own node discards it, in the case of other wireless nodes addressed and means for relaying it,
The specific wireless node maintains a route management table in which “destination”, “adjacent”, and “hop count” are recorded. When the route maintenance packet is received, information on the wireless node that has passed from the route maintenance packet is stored. Means for taking out and recording the route information in the route management table; and means for taking out a transmission source from the route maintenance packet and unicasting a route maintenance reply packet to the transmission source,
A route maintenance system in a wireless network, which constantly confirms that a packet arrives between the wireless node and the specific wireless node.

In a wireless network composed of a wireless node that transmits and receives packets via a wireless interface and a specific wireless node that serves as a gateway to an external network,
The wireless node includes a destination node (hereinafter referred to as “destination”), an adjacent node (hereinafter referred to as “adjacent”) for transferring a packet to the destination node, and a metric (hereinafter referred to as “route”) required for the packet to reach the destination node. It holds a route management table with a value indicating whether or not each wireless node has received a packet within a predetermined time (hereinafter referred to as “route check”), and maintains a route maintenance packet at a predetermined time. Means for generating and broadcasting to the adjacent wireless node, means for unicasting the route maintenance reply packet to the transmission source when the route maintenance packet is received, and the route maintenance reply packet cannot be received from the adjacent wireless node And means for deleting the information corresponding to the adjacent wireless node from the route management table and reconstructing the route. Route maintenance system in a wireless network, wherein the door.

In the route maintenance system in the wireless network according to claim 7 or 8,
The specific wireless node, when a predetermined time required for receiving a packet used for “route check” described in the route management table is not sent from the wireless node within the predetermined time, the wireless node A system for maintaining a route in a wireless network, comprising means for reconstructing a route to and from the wireless network.

In the route maintenance system in the wireless network according to claim 10 or 9,
If the wireless node does not receive a reply packet even after a predetermined time has elapsed since the packet was transmitted to the other wireless node through the route based on the route management table, the “adjacent” of the route management table Among these, a means for deleting the same route as the “adjacent” of a route without a reply packet, a means for broadcasting a failure notification packet to an adjacent wireless node in a range where direct communication is possible via a wireless link, and the wireless node A route maintenance system in a wireless network, comprising: means for reconstructing a route between the wireless networks.

In a wireless network composed of a wireless node that transmits and receives packets via a wireless interface and a specific wireless node that serves as a gateway to an external network,
12. A wireless node device comprising means for a specific wireless node constituting the route maintenance system in the wireless network according to claim 7.

In a wireless network composed of a wireless node that transmits and receives packets via a wireless interface and a specific wireless node that serves as a gateway to an external network,
A wireless node device comprising each means of a wireless node constituting the route maintenance system in the wireless network according to claim 7.