Spectrum Management And Routing In Multi-hop Cognitive Radio Networks

Cognitive Radio Technology enables unauthorized radio devices to actively discover unoccupied licensed spectrum bands and make them for proper use, which improves spectrum utilization and avoids interferences and collisions via selecting appropriate working frequency band, and thus improves efficiency of communication. Therefore, Cognitive Radio Technology can alleviate the contradiction between the resource exhaustion at open spectrum bands and the underutilization of licensed spectrum bands.The multi-hop Cognitive Radio Network is a wireless network formed by a large number of nodes with cognitive radio transceivers, in an ad hoc manner. The difference between multi-hop cognitive radio networks and traditional ones lies in that: the traditional wireless networks are based on fixed frequency bands, while the cognitive radio networks have dynamic spectrum access, whose network communications are based on a set of available frequency bands under specific limit of time, region and rules. Cognitive Radio Network can radically improve the unbalanced usage of spectrum resources. It is the future direction of wireless communication development, and is entering industry through standardizations.This dissertation focuses on the scenario of multi-hop cognitive radio networks with multiple flows. Taking into account the spectrum heterogeneity, variation of available frequency bands and the inconsistency of spectrum opportunities between network nodes, we mainly investigate the following 3 challenges:1. Because of the dynamic spectrum accessing feature in cognitive radio network,nodes keep changing their working frequency bands while serving for multipleflows. Different serving orders result in different orders of frequency band switching, and thus affect the forwarding efficiency of nodes.2. Cognitive radio network employs multiple frequency bands for communication,which results in changing network topology. Therefore, routing schemes should specify the position of node together with its working frequency band, so as to derive effective multi-hop paths. On the other hand, the efficiency of network transmission is affected by different spectrum assignments at nodes along the path and the existing data flows.3. In multi-hop cognitive radio networks with multiple existing flows, intersecting relaying nodes are facing increasing work load and decreasing processing ability due to frequent spectrum switching, which results in network bottleneck. Neighborhood nodes don't have effective cooperations, which makes the spectrum usage in the neighboring area of intersecting nodes unapplicable with the requirement of multiple data flow transmission.To resolve the problems above, this dissertation carries out research on the schemes of spectrum management and routing, from node's, path's and network's point of view. Its work are based on three issues: (1) spectrum management and scheduling at nodes. (2) on-demand routing in cognitive radio networks. (3) neighborhood routing in cognitive radio networks. The contribution of this dissertation includes:1. Multiple-frequency-band scheduling scheme for network nodes: Thisdissertation analyzes the performance change at cognitive radio network nodeswhen serving for multiple flows, and addresses the existence and causes of variousdelays. A node analytical model with multi-frequency-band scheduling module,together with multi-frequency-multi-flow scheduling scheme of "scan among active bands" are proposed. It effectively reduces the extra delays incurred bymultiple data flows and improves efficiency of relaying.2. Joint scheme of spectrum assignment and on-demand routing in cognitive radio networks: A joint spectrum assignment and on-demand routingprotocol is proposed according to the multi-hop and dynamic spectrum accessnature on cognitive radio networks. In the protocol, frequency band on everyhop is determined while routing. Various delays caused by existing flows andspectrum assignment along the path are taken into account, based on which a metric is proposed. This joint protocol incurs paths with higher transmission performance when compared to other routing schemes.3. Neighborhood routing scheme for cognitive radio networks: This dissertation carries out thorough analysis on the processing performance at intersecting relaying nodes. It addresses the limit of routing protocols when selecting next-hop nodes, and describes the possibility of the formation of network bottlenecks. It derives work load evaluation through queueing system modeling at relaying nodes, and proposes a neighborhood coordination based flow redirection mechanism, which distributed the work load from overloaded nodes to other lightloaded ones, and thus improves the efficiency of network transmission and avoids the forming of network bottlenecks.This work is supported by the National Science Foundation of China "Research on Key Issues of Cognitive Radio Networks"(No.60602029), "Information Processing and Transmission on Reconstructable MIMO-based Wireless Sensor Networks" (No.60572049) and the Foundation of Hubei Provincial Key Laboratory of Smart Internet Technology "Research on Key Issues of multi-hop Cognitive Radio Networks"( No.HSIT200605). This dissertation gives a framework of resolution to the multi-hop cognitive radio networks with multiple data flows, it is significant for dealing with spectrum resource exhaustion, blind spots of network coverage and cooperation between heterogeneous wireless networks in the future.