發(fā)布時間：2018-04-18 20:37

  本文選題：網(wǎng)絡編碼 + 弱安全��； 參考：《上海交通大學》2014年博士論文

【摘要】：網(wǎng)絡編碼是近年來通信領域國際研究的熱點方向之一,在網(wǎng)絡傳輸領域有較好的應用前景。在網(wǎng)絡編碼發(fā)展的同時,相應的安全問題也不容忽視,這些安全問題一定程度上也影響了網(wǎng)絡編碼方法的實際應用,阻礙了它在實際中的應用和推廣。現(xiàn)有的網(wǎng)絡編碼安全問題主要包括污染和竊聽,另外,也有過編碼、欠編碼、自私行為等一些網(wǎng)絡編碼特有面臨的安全問題。然而,到目前為止,網(wǎng)絡安全編碼方面的研究成果相對還較少,有待于進一步展開研究。對于網(wǎng)絡編碼所面臨的一類典型安全問題---搭線竊聽,現(xiàn)有的防搭線竊聽的安全網(wǎng)絡編碼方法主要是從編碼角度出發(fā),以編碼的復雜度換取傳輸?shù)陌踩��?雖然近年有一定研究成果考慮了網(wǎng)絡拓撲對安全傳輸?shù)闹匾�性并給出相應的解決方案,但是這方面研究成果不但少,而且也存在適用范圍小等問題。在此背景下,本文著重研究面向搭線竊聽的網(wǎng)絡安全編碼方法。此外,針對網(wǎng)絡編碼面臨的一類特殊安全問題----用戶自私行為,考慮到傳統(tǒng)自私行為解決方案并不適合在網(wǎng)絡編碼環(huán)境下使用,本文也對網(wǎng)絡編碼環(huán)境下的自私行為進行了探討研究。本文首先從路由選擇角度出發(fā),研究防搭線竊聽攻擊安全網(wǎng)絡編碼問題,提出了基于路由選擇的防搭線竊聽攻擊安全網(wǎng)絡編碼方法�；�于網(wǎng)絡拓撲結(jié)構(gòu),首先對單播網(wǎng)絡進行研究,在已知被竊聽鏈路位置(不可信鏈路位置)的情況下,對被竊聽鏈路的傳送消息進行分析。在保證網(wǎng)絡最大流不變的前提下,盡量移除較少被竊聽鏈路或者正常鏈路,以保證竊聽者無法得到完整的網(wǎng)絡源信息,而信宿節(jié)點能夠正常的接收到所有的信息。根據(jù)得到的安全網(wǎng)絡拓撲構(gòu)造新的系統(tǒng)傳輸矩陣,從而獲得安全網(wǎng)絡編碼,達到抵御搭線竊聽攻擊的目的;隨后將該方法由單播情形擴展到多播情形。更進一步,又在該方法的基礎上給出了一種改進的基于路由選擇的多播安全網(wǎng)絡編碼方法。仿真實驗表明了上述方法的有效性。然后本文又從網(wǎng)絡分割角度出發(fā),針對被竊聽節(jié)點位置已知和被竊聽節(jié)點位置未知兩種情況下的單播網(wǎng)絡搭線竊聽攻擊問題,提出了基于網(wǎng)絡分割的防搭線竊聽攻擊的網(wǎng)絡編碼安全方法。針對被竊聽節(jié)點位置已知情況下的網(wǎng)絡竊聽攻擊,安全編碼方法的核心在于:在網(wǎng)絡最大流沒有發(fā)生變化的前提下,找出合適的網(wǎng)絡分割方法,使網(wǎng)絡至多分割成三個子網(wǎng)絡,并保證子網(wǎng)絡中最少有一個沒有被竊聽者竊聽過,從而達到網(wǎng)絡的弱安全目標。針對被竊聽節(jié)點位置未知情況下的網(wǎng)絡搭線竊聽攻擊,安全編碼方法的核心為:在保證n?2m?l(n為被竊聽的中間節(jié)點數(shù)量,m為被分割的子網(wǎng)絡數(shù)量,l為子網(wǎng)絡最大流為1的數(shù)量)的前提下,便能找到合適的網(wǎng)絡分割安全編碼方法使網(wǎng)絡達到弱安全,仿真實驗證實該二方法的有效性。最后本文也對如何防范網(wǎng)絡編碼面臨的節(jié)點自私行為進行了研究,針對單播網(wǎng)絡傳輸中存在一定比例自私節(jié)點的情況,提出了一種防網(wǎng)絡節(jié)點自私行為的安全網(wǎng)絡編碼方法。該方法通過計算網(wǎng)絡存在一定數(shù)量自私節(jié)點時可能出現(xiàn)的傳輸失敗的概率,根據(jù)所得到的結(jié)果求解出網(wǎng)絡最大流的期望值,最后再由期望值對網(wǎng)絡進行重新編碼。在允許的誤差范圍下,找到最適合的編碼方法進行編碼,以此達到抵抗因節(jié)點自私行為而導致的網(wǎng)絡傳輸失敗的目的。仿真實驗證實了該方法的有效性。
[Abstract]:Network encoding is one of the hot topics of international research in the field of communications in recent years, there are better application prospects in the field of network transmission. In the network encoding development at the same time, the corresponding security issues can not be ignored, these security problems to a certain extent also affect the practical application of network encoding method, hinders its application and promotion in practice. The safety problem of the existing network. Encoding includes pollution and eavesdropping, in addition, there have been less encoding, encoding, selfish behavior and some network security problems facing the special encoding. However, so far, the research results of network security encoding also relatively less, need to be further studied for a typical class. Network security issues faced by wiretapping - encoding, anti wiretapping network security existing encoding method is mainly starting from the perspective of complexity for encoding, the encoding The safety of transmission, although in recent years, some research results, considering the importance of network topology on transmission security and gives the corresponding solutions, but the results are not only small, but there are also problems such as small application scope. Under this background, this paper focuses on the research of network security to wiretap encoding method. In addition, according to the a special kind of network security problems facing the selfish behavior of users - encoding, taking into account the solutions are not suitable for use in the network environment the traditional encoding of selfish behavior, the article also discussed the research on the network encoding environment selfish behavior. This paper from the Perspective of the routing problem, wiretapping attack network security protection encoding research, put forward the routing of the anti wiretapping attack security network encoding method based on network topology based on the first study of unicast network, Eavesdropping link at a known location (unreliable link position) under the condition of transmitted message interception were analyzed. The link under the premise of ensuring the maximum flow of network at the same time, try to remove less eavesdropping or normal link link, to ensure that the eavesdropper cannot get network complete source information, and the destination node can be normal receive all the information. According to the topological structure of the network security system transfer matrix are new, so as to obtain safe network encoding, to resist the wiretapping attack; then the method is extended to multicast unicast case. Further more, and on the basis of this method is presented based on an improved routing of multicast security network encoding method. Simulation results show the effectiveness of the proposed method. Then this paper from the perspective of network segmentation, the node location is known and eavesdropping Unicast network wiretapping attack eavesdropping node location unknown under two kinds of situations are problems, put forward the network security encoding method of anti wiretapping attack network based image segmentation. Aiming at the eavesdropping node position under a known network eavesdropping attacks, security encoding method core lies in: the premise of no change in the maximum flow of network and find the appropriate network segmentation method, enables the network to split into three sub networks, and to ensure that the network has at least one eavesdropper has not been bugged, so as to achieve the goal of weak security network. Aiming at the eavesdropping node location under unknown network wiretapping attack, core security encoding method is: in the guarantee n? 2m? L (n is the number of intermediate nodes for M sub network eavesdropping number, divided l into 1 sub network maximum flow quantity) of the premise, can find a suitable network segmentation The safety encoding method enables the network to achieve weak security, simulation experiments confirmed the validity of the two methods. Finally, this paper also makes a study on how to prevent the network node encoding face selfish behavior, according to a certain proportion of selfish nodes exist in unicast network transmission, put forward a kind of anti network nodes selfish behavior safety network encoding method. The method by calculating the probability of transmission network may appear a certain number of selfish nodes when the failure of the expected value of the maximum flow of network is calculated according to the obtained results, finally the expected value of re encoding of the network. In the allowed range of error, to find the most suitable method for encoding encoding, in order to achieve resistance network the transmission caused by the selfish behavior of nodes failure. Simulation results demonstrate the effectiveness of the proposed method.

【學位授予單位】：上海交通大學
【學位級別】：博士
【學位授予年份】：2014
【分類號】：TN915.08

【參考文獻】

相關期刊論文 前2條

1 李大霖;林雪紅;林家儒;吳偉陵;;安全網(wǎng)絡編碼的一個必要條件[J];北京郵電大學學報;2008年05期

2 李丹;吳建平;崔勇;;應用層組播用戶的自私性研究[J];軟件學報;2007年03期

，

本文編號：1769986