CAJ | 학술논문

对无线传感器网络目标跟踪中的协同任务分配机制进行了研究，针对一般任务分配算法中优化目标单一的缺陷，提出了一种基于多目标优化的任务分配算法。首先，建立了基于动态联盟的具有跟踪精度、系统能耗、负载均衡等多个目标参数的优化模型，并采用多目标进化算法NSGA-Ⅱ对模型进行求解；然后，提出了一种基于折中度的决策精选策略，从最优解集中决策出最终的任务分配方案。针对跟踪精度、能耗、负载均衡的仿真结果表明：所提算法可以对多个目标并行优化，较快收敛到全局最优解；与一般任务分配算法相比，该算法可获得更佳的调度结果。
대무선전감기망락목표근종중적협동임무분배궤제진행료연구，침대일반임무분배산법중우화목표단일적결함，제출료일충기우다목표우화적임무분배산법。수선，건립료기우동태련맹적구유근종정도、계통능모、부재균형등다개목표삼수적우화모형，병채용다목표진화산법NSGA-Ⅱ대모형진행구해；연후，제출료일충기우절중도적결책정선책략，종최우해집중결책출최종적임무분배방안。침대근종정도、능모、부재균형적방진결과표명：소제산법가이대다개목표병행우화，교쾌수렴도전국최우해；여일반임무분배산법상비，해산법가획득경가적조도결과。
The collaborative task allocation mechanism of target tracking in wireless sensor networks is studied.To solve the problem that the optimization objectives of the general task allocation algo-rithms are single, a new task allocation algorithm based on multi-objective optimization is proposed. First, an optimization model based on dynamic coalition with multiple objectives such as tracking ac-curacy, energy consumption and load balancing is established.And a multi-objective evolutionary algorithm named as NSGA ( non-dominated sorting genetic algorithm)-Ⅱis adopted to solve optimi-zation model.Then, a novel decision-making strategy based on the degree of compromise is presen-ted to give the final task allocation scheme from the set of optimal solutions.The simulation results aiming at the tracking accuracy, energy consumption and load balancing show that the proposed al-gorithm can optimize multi-objectives in parallel and converge to the global optimal solution quickly. Compared with the general task allocation algorithms, the proposed algorithm can obtain better scheduling results.