论文部分内容阅读
为降低对温度传感器网络的性能评估成本,提出了一种更准确的传感器网络性能检测方法,避免因性能评估误差产生的负面影响。建立的传感器网络的模型,提出了基于鞍点近似估计原理的性能评估策略,详细阐述了鞍点近似算法的优化过程,并给出了其对应的误差处理方法。该方法通过采用鞍点逼近方法,并利用一组结果证明了该方法的高准确性和低复杂度。仿真测试结果表明,在某些情况下最优化联合规则等同于一个简单的多数决定原则。
In order to reduce the cost of evaluating the performance of temperature sensor networks, a more accurate method of sensor network performance testing is proposed to avoid the negative impact of performance evaluation error. The model of sensor network is established, and the performance evaluation method based on the saddle point approximate estimation principle is proposed. The optimization process of the saddle point approximation algorithm is expounded in detail and the corresponding error handling method is given. The method adopts the saddle point approximation method, and uses a set of results to prove the high accuracy and low complexity of the method. Simulation results show that in some cases the optimization of joint rules is equivalent to a simple principle of majority decision.