End-to-end Variation-adaptive Transmission Mechanismsand Key Technologies For Cognitive Wireless Networks

While the demands of modern society for high-speed, ubiquitous, and diversifiedwireless applications are increasing fast, current communication systems are inefficientwith resource utilization and are consisting of heterogeneous networks everywhere be-cause of the traditional enclosed static networking manner. How to provide a ubiquitousaccess and high-quality communication performance for users in heterogeneous envi-ronments has become one of the most promising problems in wireless communications.The emergence of cognitive wireless technology provides an important approach tosolve these problems. Equipped with this technology, a wireless network is capable ofactively sensing the internal parameters and external environments of communicationnodes, intelligently making decisions, and dynamically allocating resources so as to sat-isfy the users’transmission demand and maximize the utilization of network resources.There exist three main theoretical issues for cognitive wireless networks, i.e., cog-nition, adaptation and autonomy. This dissertation is devoted to the adaptation issue ofcognitive wireless networks in a wireless multi-hop topology with the specific focus onend-to-end variation-adaptive transmission mechanisms and key technologies for cogni-tive wireless networks. We delve into the transmission capacity of cognitive wirelessnetworks, dynamic spectrum allocation, variation-adaptive access control, routing tech-nology, end-to-end variation-adaptive transmission control. Our research basically co-vers the main aspects of cognitive communication protocols from bottom layer to toplayer, thus forms a relatively complete set of cognitive wireless networks varia-tion-adaptive transmission system.First of all, Chapter2proposes an end-to-end variation-adaptive transmissionframework and studies the transmission capacity of cognitive wireless networks. Bystrengthening the ability of cross-layer interaction between different protocol layers anddesigning efficient transmission and control mechanism based on cognition information,the proposed transmission framework is capable of variation-adaptive to dynamic envi-ronment. Based on this framework, we then adopt the Markov chain to analyze thetransmission capacity of cognitive wireless network on both control and data channels. This analysis model can figure out the capacity of entire network and every pair oftransmission nodes.In the ensuing chapters, we then study a few key technologies of the end-to-endvariation-adaptive transmission. Chapter3delves into the issue of dynamic spectrumallocation. We propose a dynamic spectrum allocation algorithm under the end-to-endtransmission QoS demand. In this algorithm, nodes obtain spectrum in accordance withthe QoS requirements, and if they can not obtain enough spectrum resources to meet theQoS requirement are assigned with no spectrum resources. The allocation result willmake node obtained spectrum have a good ability to support the end-to-end QoS trans-mission.Chapter4study the access control technology in cognitive wireless networks. Wepropose a novel access control mechanism, called active cognition and passive alloca-tion and develop a variation-adaptive access control protocol based on this mechanism.In the active cognition and passive allocation access control mechanism, all communi-cation nodes require resources in accordance with the QoS demand just in data trans-mission and implement the active cognition to sense the dynamic resource changing in-formation otherwise. Through this way, all nodes could obtain accurate resource chang-ing information and hence have a good capability of QoS transmission supporting. Theaccess control protocol use frequency division duplex transceiver to continuously listenthe control channel during the transmission, and is capable of supporting various com-munication methods between nodes with different number of transceivers.In chapter5, we delve into the effect of node mobility on end-to-end transmissionperformance. We first propose a link duration probability calculation method forepoch-based mobility model and develop a node mobility cognitive routing protocolbased on this calculation method. This routing protocol can provide a stable route forend-to-end transmission by considering both the end-to-end maximum stable probabilityand the minimum number of hops. Then we develop a cross-layer control mechanismwith the following two specific functions. First, Active Route Switch can predict therouting break and initiate a new route discovery in advance for preventing the routingbreak occurrence. Second, during the routing break period, Save then Forward can forma de facto reliable transmission between the routing break node and the source nodewith the benefit of maintaining the packet order and integrity. Chapter6finally studies the end-to-end transmission control mechanism systemat-ically. We investigate the main factors that impact the end-to-end transmission perfor-mance of cognitive wireless networks and classify them into two categories: factorsbased on spectrum changes or factors based on node mobility. To handle these two kindsof factors separately, we propose two sets of control mechanisms, i.e., backpressuretransmission control and cross-layer transmission control interacting with link reliablerouting. The former mechanism takes full advantage of the interaction, storage and ad-justment ability between local nodes to solve various adverse network conditions in lo-cal area. The latter interact with the cross-layer control mechanism in chapter5serve asa good solution for adapting node mobility. In combination with the transmission tech-nologies developed in the forward chapters, this set of transmission control mechanismsconstitutes a relatively complete variation-adaptive transmission system for cognitivewireless networks.