Research On Some Key Techniques For Cognitive Wireless Mesh Networks

The past decade has witnessed a huge and fast development in the wireless communication techniques. Considering the periods and costs of lines of communication deployment, the area coverage, the restrictions of geographical surroundings and the maintenance of lines, the wireless communication performs much better compared with the wired one, which enables the former the most rapidly-developing and widely-applied fundamental technique in present information communication field. With the increasingly higher requirements on the quality applications of the wireless communication, the network mode is becoming more and more complex and the radio spectrum resource are becoming more and more precious. Thus it is so imperative a problem for the experts that how to meet the communication demands using the limited resources to the uttermost.The Wireless Mesh Network (WMN) is a brand new wireless broadband network which exceeds the characteristics of the traditional wireless network in aspects like the flexible networking, the network coverage rate raising, the network capacity expanding and the initial investment saving. Especially, the WMN helps realize the broadband wireless access and coverage in conditions lacking the cable and other wired resources. Since the traditional fixed spectrum allocation policies led to different degrees of idle in time and space in wireless spectrum resources, the Cognitive Radio Technology (CRT) can make full use of the idle spectrum and efficiently solve the scarcity problem of wireless spectrum resources by adopting dynamic access spectrum resources. Thus, it has become a present hotspot in research.By applying the CRT in the WMN, a new type of broadband wireless network is constructed with the cognitive ability, known as the Cognitive Wireless Mesh Networks (CWMN). It has a great potential in integrating the heterogeneous networks and improving the utilization efficiency of wireless resources, and has become a promising candidate for future network. This paper discusses the basic concept, the characteristics, the network architecture and the related technology research of the CWMN in depth, basing on which the research on the key technology of CWMN is conducted. The main work is listed as follows:1. It studies on the channel detection technology of the CWMN. It defines the concept of the cognitive channel throughput, and studies the multi-channel detection technology to pursue the cognitive channel throughput maximization as the goal. Firstly, for the ideal sensing, two kinds of simple multi-channel sequence detecting methods are put forward and research in depth. The methods are a decrement detection sequence according to the channel rate and a decrement detection sequence according to the free probability of channels. Secondly, it comes to the non-ideal sensing situation, the sole existence of the best sensing time is analyzed and a searching method for the best sensing time is provided. Then the simulation of best sensing time and cognitive throughput performance in the single user sensing and collaboration sensing shows that with the increase of the number of cooperative nodes, the best sensing time will reduce, and cognitive throughput will increase. Compared with multi-channel sequence detection technology, using parallel detection can improve the detection efficiency. For multi-channel parallel detection method, this paper studies the assignment problem of multiple nodes involved in cooperative detection and puts forward a multiple node assignment algorithm for multi-channel cognitive throughput maximum. The simulation results show that, through the cognitive node allocation, the detection performance can be significantly improved.2. It studies on cognitive WMN channel assignment and routing technology. Based on the restriction that the interference to authorized users created by the transmission of cognitive node shall not exceed certain limit, the available channel definition of the CWMN is presented, and a method to calculate the available channel set is proposed. The channel assignment and routing metrics of CWMN has been researched and we put forward some guidelines adapted to the features of CWMN. According to the dynamic changes of CWMN available channels, this paper presents a Hyacinth-model based routing and channel assignment algorithm which can establish spectrum-sensing tree based on the dynamic changes of node transmission power and the time-varying condition of communication distance, assign and update channels adaptively to reduce impact on network performance caused by the changes of available channels. The simulation results show that, this algorithm can make full use of the available channels to improve the throughput of network.3. It studies on the channel and spatial degree of freedom joint assignment technology in CWMN using MIMO technology. In CWMN, the CR technology can improve the utilization ratio of spectrum in the channel level, and the MIMO technology can improve the capacity of communication link through space and time processing. This paper elaborates the basic concept and principle of CWMN-MIMO system, such as data transmission, interference counteraction, channel assignment and assignment of spatial degree of freedom, establishes a mathematic model of channel and spatial degree of freedom allocation in CWMN-MIMO system, proposes a channel and spatial degree of freedom joint assignment algorithm combined with route selection. The algorithm complys to the criterion that maximize the link capacity, considering the minimum resource occupancy and load balance, adopt the widest and shortest routing algorithm to choose the optimal transmission path of network session, then according to the activate nodes determined by the transmission path, optimize jointly multi-channel's channel assignment and the spatial degrees of freedom. Simulation results reveal that the proposed algorithm can obtain great capacity gain through optimization of the adoption of CR in channel level and MIMO in the same channel.4. It studies on the security technology in CWMN. In view of the characteristics of CWMN and the security threats, it established CWMN security model based on trust mechanism. Because of the malicious node attacks to the cooperative spectrum sensing, the trust mechanism is introduced into cooperative spectrum sensing, and it proposes a cooperative spectrum sensing strategy based on the trust mechanism. The fusion center assign trust value to each node, and update trust value according to the consistency test of node's local test results and overall decision. Each node is divided into malicious nodes, pending nodes and trusted nodes according to node trust value, and it use different fusion processing method to eliminate the influence of malicious nodes to increase the security of cooperative sensing. The simulation results verify the validity of the method. As a dynamic and initiative security policy, intrusion detection can provide strong security guarantee to networks. This paper introduces trust mechanism into CWMN with intrusion detection, and presents an intrusion detection model based on trust mechanism, and proposes a channel allocation and routing strategy. This strategy establishs reliable routing by selecting the trustable node and assign the transmission channel with high trust degree to adjacent node, and adjust channel allocation and routing dynamically according to the network security state, so as to avoid suspicious nodes and unsafe channel, and effectively improves the safe delivery fraction of the packet.