Research On Channel Allocation In The Cognitive Radio-based Special Wireless Communication Network

In recent years, with the prevailing of the wireless communication services, theallocated spectrum resources is becoming increasingly tense, which seriously restrict thedevelopment of the special wireless communication network. The current fixedallocation scheme of spectrum resources results in low spectrum efficiency, and furtherintensifies the spectrum resources shortage. The emergence of cognitive radio networkbrings the hope of addressing these contradictions. However fixed channel allocationcannot satisfy the requirements of this new technology, so it has become an urgentproblem to obtain efficient channel allocation algorithms. This dissertation focuses onthe special wireless communication network for the Armed Police Force, aiming atcompleting diverse missions. we studied the channel allocation problem for three typicalwireless network scenarios and furthermore proposed three improved efficientallocation algorithms respectively.Firstly, due to the anti-fading ability, high frequency utilization and fitness forhigh-speed data transmission of OFDM technology, it will be of great importance in theconstruction of the Armed Police Force broadband wireless communication system. Inthis paper, considering wireless networks using OFDM transmission technology, thisthesis studies the fine-grained channel allocation algorithm based on the subchannelallocation, and proposes a priority-based proportional fairness algorithm. Byintroducing a fair degree index, the user priority requirements are satisfied, while thefairness of the system is ensured and the spectrum efficiency is significantly improved.Secondly, combining the high convergence speed of maximal independent set withthe convenient search of the optimal solution of genetic algorithm, a centralized channelallocation algorithm is proposed. The basic idea of the algorithm is that users establishand initialize communication links based on the maximal independent set criterion, andthen select the appropriate channels by utilizing genetic algorithm to obtain the optimalsolution. The proposal can not only significantly increase the channel allocationefficiency, but also reduce the convergence time, providing hierarchical services forusers with different priorities.Finally, considering a distributed network under the condition that the networktopology information utilized by the above maximal independent set algorithm isunavailable, a distributed channel allocation algorithm based on game theory is studiedand an improved game algorithm based on user priorities is proposed. Taking thepriorities of different links and their utilization efficiency on different channels intoaccount, each player employs appropriate strategy to select channels via backoffmechanism until reaching the Nash Equilibrium, thus obtaining maximal efficiency ofeach link as well as a balanced allocation around the whole network. Analysis and simulation results indicate that, in the case of the absence of network topology, theproposal significantly improve the network throughput and the fairness through thecomparison with the random allocation algorithm.