機(jī)會(huì)網(wǎng)絡(luò)中節(jié)點(diǎn)激勵(lì)機(jī)制研究
發(fā)布時(shí)間:2018-09-11 15:05
【摘要】:機(jī)會(huì)網(wǎng)絡(luò)是由無(wú)線自組織網(wǎng)絡(luò)和延遲容忍網(wǎng)絡(luò)演化而來(lái)的一種新型網(wǎng)絡(luò),它是一種源節(jié)點(diǎn)與目的節(jié)點(diǎn)可能不存在完整的通信鏈路、利用節(jié)點(diǎn)移動(dòng)帶來(lái)的相遇機(jī)會(huì)進(jìn)行數(shù)據(jù)交換的移動(dòng)自組織網(wǎng)絡(luò)。與傳統(tǒng)的無(wú)線自組織網(wǎng)絡(luò)不同,機(jī)會(huì)網(wǎng)絡(luò)節(jié)點(diǎn)稀疏、分布不均勻,節(jié)點(diǎn)間都是機(jī)會(huì)性的相遇,所以采用的是“存儲(chǔ)-攜帶-轉(zhuǎn)發(fā)”的路由機(jī)制。這種路由機(jī)制完全依賴于中繼節(jié)點(diǎn)的合作,現(xiàn)有的路由轉(zhuǎn)發(fā)協(xié)議也都是假設(shè)節(jié)點(diǎn)會(huì)選擇合作即幫助其他節(jié)點(diǎn)轉(zhuǎn)發(fā)數(shù)據(jù)。然而,組成機(jī)會(huì)網(wǎng)絡(luò)的大部分移動(dòng)設(shè)備是資源有限的,比如電池、內(nèi)存、CPU等資源。如果節(jié)點(diǎn)被理性個(gè)體控制的話,它們可能會(huì)為了節(jié)省資源而不幫其他節(jié)點(diǎn)轉(zhuǎn)發(fā)消息,這些節(jié)點(diǎn)被稱為“自私節(jié)點(diǎn)”,目前,大量文獻(xiàn)研究發(fā)現(xiàn)這些自私節(jié)點(diǎn)將嚴(yán)重影響整個(gè)網(wǎng)絡(luò)的性能,并且導(dǎo)致數(shù)據(jù)傳輸率下降。 本文提出的激勵(lì)機(jī)制就是為了解決自私節(jié)點(diǎn)影響網(wǎng)絡(luò)性能的問(wèn)題,現(xiàn)有的一些激勵(lì)機(jī)制基本上都是應(yīng)用在傳統(tǒng)的無(wú)線自組織網(wǎng)絡(luò)當(dāng)中,然而由于機(jī)會(huì)網(wǎng)絡(luò)的一些固有特性如網(wǎng)絡(luò)間歇性斷開(kāi)、延遲長(zhǎng)等,使得這些激勵(lì)機(jī)制不太適用于機(jī)會(huì)網(wǎng)絡(luò)。本文在討論現(xiàn)有的激勵(lì)機(jī)制基礎(chǔ)上,提出了一種基于交換的聲譽(yù)激勵(lì)機(jī)制(ERIS),該機(jī)制的主要思想是:相遇的兩個(gè)節(jié)點(diǎn)是基于感興趣的消息出發(fā),聲譽(yù)高的兩個(gè)相遇節(jié)點(diǎn)彼此交換消息,本文認(rèn)為任何交換的消息都是有價(jià)值的,雖然有可能交換到的消息不是自身感興趣的,但是以后可以用它們來(lái)交換自身感興趣的消息,這樣的交互使得消息在網(wǎng)絡(luò)中能夠順利的傳輸,數(shù)據(jù)傳輸延遲下降。不交換消息(不合作)的節(jié)點(diǎn)不僅可能失去自身感興趣的消息也會(huì)導(dǎo)致聲譽(yù)下降,聲譽(yù)下降到比閾值低時(shí),它將被隔離出網(wǎng)絡(luò),不會(huì)再有節(jié)點(diǎn)為它轉(zhuǎn)發(fā)消息。 本文利用博弈論理論分析ERIS機(jī)制,并在ONE模擬器上進(jìn)行仿真實(shí)驗(yàn),實(shí)驗(yàn)結(jié)果表明該激勵(lì)機(jī)制可以減少節(jié)點(diǎn)的自私行為,提高數(shù)據(jù)的傳輸率。自私行為在長(zhǎng)期看來(lái)對(duì)節(jié)點(diǎn)自身是不利的,節(jié)點(diǎn)為了獲取自身感興趣的消息和提高自身的聲譽(yù)會(huì)盡量選擇合作,實(shí)驗(yàn)證明在本文的激勵(lì)機(jī)制中節(jié)點(diǎn)選擇納什均衡的策略組合時(shí),整個(gè)網(wǎng)絡(luò)的性能和數(shù)據(jù)傳輸率均得到很大的改善。
[Abstract]:Opportunistic network is a new network evolved from wireless ad hoc network and delay tolerance network. It is a kind of source node and destination node which may not have a complete communication link. A mobile ad hoc network that uses the encounter opportunity brought by node mobility to exchange data. Different from the traditional wireless ad hoc networks, the nodes of the opportunistic networks are sparse, unevenly distributed, and the nodes meet each other opportunistically, so the routing mechanism of "store-carry-forward" is adopted. This routing mechanism depends entirely on the cooperation of relay nodes. Existing routing and forwarding protocols also assume that nodes will choose cooperation to help other nodes forward data. However, most of the mobile devices that make up the opportunity network have limited resources, such as batteries, memory processors, and so on. If nodes were controlled by rational individuals, they might not forward messages for other nodes, known as "selfish nodes," in order to save resources. A large number of literatures have found that these selfish nodes will seriously affect the performance of the entire network and lead to a decrease in the data transmission rate. The incentive mechanism proposed in this paper is to solve the problem that selfish nodes affect the performance of the network. Some existing incentive mechanisms are basically applied in the traditional wireless ad hoc networks. However, due to some inherent characteristics of opportunistic networks, such as intermittent disconnection, long delay and so on, these incentive mechanisms are not suitable for opportunistic networks. In this paper, based on the discussion of the existing incentive mechanism, a reputation incentive mechanism based on exchange (ERIS),) is proposed. The main idea of the mechanism is that the two nodes met are based on the interested message. Two reputable meeting nodes exchange messages with each other. This paper holds that any messages exchanged are valuable. Although the messages that may be exchanged are not of interest to them, they can be used to exchange messages of interest to them in the future. Such interaction enables messages to be transmitted smoothly in the network, and data transmission delays decrease. The node that does not exchange messages (uncooperative) may not only lose the information of its own interest but also lead to the loss of reputation. When the reputation falls below the threshold it will be isolated from the network and no more nodes will forward messages for it. In this paper, the game theory is used to analyze the ERIS mechanism, and the simulation experiments are carried out on the ONE simulator. The experimental results show that the excitation mechanism can reduce the selfish behavior of the nodes and improve the data transmission rate. In the long run, selfishness is unfavorable to the node itself. In order to obtain the information of interest to the node and improve its reputation, the node will choose cooperation as far as possible. The experimental results show that in the incentive mechanism of this paper, the node chooses the Nash equilibrium strategy combination. The performance and data transmission rate of the whole network are greatly improved.
【學(xué)位授予單位】:大連理工大學(xué)
【學(xué)位級(jí)別】:碩士
【學(xué)位授予年份】:2014
【分類號(hào)】:TN92
本文編號(hào):2237041
[Abstract]:Opportunistic network is a new network evolved from wireless ad hoc network and delay tolerance network. It is a kind of source node and destination node which may not have a complete communication link. A mobile ad hoc network that uses the encounter opportunity brought by node mobility to exchange data. Different from the traditional wireless ad hoc networks, the nodes of the opportunistic networks are sparse, unevenly distributed, and the nodes meet each other opportunistically, so the routing mechanism of "store-carry-forward" is adopted. This routing mechanism depends entirely on the cooperation of relay nodes. Existing routing and forwarding protocols also assume that nodes will choose cooperation to help other nodes forward data. However, most of the mobile devices that make up the opportunity network have limited resources, such as batteries, memory processors, and so on. If nodes were controlled by rational individuals, they might not forward messages for other nodes, known as "selfish nodes," in order to save resources. A large number of literatures have found that these selfish nodes will seriously affect the performance of the entire network and lead to a decrease in the data transmission rate. The incentive mechanism proposed in this paper is to solve the problem that selfish nodes affect the performance of the network. Some existing incentive mechanisms are basically applied in the traditional wireless ad hoc networks. However, due to some inherent characteristics of opportunistic networks, such as intermittent disconnection, long delay and so on, these incentive mechanisms are not suitable for opportunistic networks. In this paper, based on the discussion of the existing incentive mechanism, a reputation incentive mechanism based on exchange (ERIS),) is proposed. The main idea of the mechanism is that the two nodes met are based on the interested message. Two reputable meeting nodes exchange messages with each other. This paper holds that any messages exchanged are valuable. Although the messages that may be exchanged are not of interest to them, they can be used to exchange messages of interest to them in the future. Such interaction enables messages to be transmitted smoothly in the network, and data transmission delays decrease. The node that does not exchange messages (uncooperative) may not only lose the information of its own interest but also lead to the loss of reputation. When the reputation falls below the threshold it will be isolated from the network and no more nodes will forward messages for it. In this paper, the game theory is used to analyze the ERIS mechanism, and the simulation experiments are carried out on the ONE simulator. The experimental results show that the excitation mechanism can reduce the selfish behavior of the nodes and improve the data transmission rate. In the long run, selfishness is unfavorable to the node itself. In order to obtain the information of interest to the node and improve its reputation, the node will choose cooperation as far as possible. The experimental results show that in the incentive mechanism of this paper, the node chooses the Nash equilibrium strategy combination. The performance and data transmission rate of the whole network are greatly improved.
【學(xué)位授予單位】:大連理工大學(xué)
【學(xué)位級(jí)別】:碩士
【學(xué)位授予年份】:2014
【分類號(hào)】:TN92
【參考文獻(xiàn)】
相關(guān)期刊論文 前4條
1 牛建偉;周興;劉燕;孫利民;馬建;;一種基于社區(qū)機(jī)會(huì)網(wǎng)絡(luò)的消息傳輸算法[J];計(jì)算機(jī)研究與發(fā)展;2009年12期
2 李云;于季弘;尤肖虎;;資源受限的機(jī)會(huì)網(wǎng)絡(luò)節(jié)點(diǎn)激勵(lì)策略研究[J];計(jì)算機(jī)學(xué)報(bào);2013年05期
3 熊永平;孫利民;牛建偉;劉燕;;機(jī)會(huì)網(wǎng)絡(luò)[J];軟件學(xué)報(bào);2009年01期
4 趙廣松;陳鳴;;自私性機(jī)會(huì)網(wǎng)絡(luò)中激勵(lì)感知的內(nèi)容分發(fā)的研究[J];通信學(xué)報(bào);2013年02期
相關(guān)博士學(xué)位論文 前2條
1 葉暉;機(jī)會(huì)網(wǎng)絡(luò)數(shù)據(jù)分發(fā)關(guān)鍵技術(shù)研究[D];中南大學(xué);2010年
2 范家璐;機(jī)會(huì)移動(dòng)網(wǎng)絡(luò)建模與應(yīng)用研究[D];浙江大學(xué);2011年
,本文編號(hào):2237041
本文鏈接:http://sikaile.net/kejilunwen/wltx/2237041.html
最近更新
教材專著