Li Yiliang, Feng Jun-E, Li Xiaodi, Xu Shengyuan
IEEE Trans Neural Netw Learn Syst. 2023 Dec;34(12):10838-10850. doi: 10.1109/TNNLS.2022.3171576. Epub 2023 Nov 30.
Considered the stimulation of tumor necrosis factor as an impulsive control, an apoptosis network is modeled as a state-dependent impulsive Boolean network (SDIBN). Making cell death normally means driving the trajectory of an apoptosis network out of states that indicate cell survival. To achieve the goal, this article focuses on the pinning controller design for set reachability of SDIBNs. To begin with, the definitions of reachability and set reachability are introduced, and their relation is illustrated. For judging whether the trajectory of an SDIBN leaves undesirable states, a necessary and sufficient condition is presented according to the criteria for the set reachability. In addition, a series of algorithms is provided to find all possible sets of pinning nodes for the set reachability. Note that attractors containing in all undesirable states are studied to make SDIBNs set reachable via controlling the smallest states. For the purpose of determining pinning nodes for one-step set reachability, the Hamming distance is presented under scalar forms of states. Pinning nodes with the smallest cardinality for the set reachability are derived by deleting some redundant nodes. Compared with the existing results, the state feedback gain can be obtained without solving logical matrix equations. The computation complexity of the proposed approach is lower than that of the existing methods. Moreover, the method of designing pinning controllers is used to discuss apoptosis networks. The experimental result shows that apoptosis networks depart from undesirable states by controlling only one node.
将肿瘤坏死因子的刺激视为一种脉冲控制,将凋亡网络建模为状态依赖脉冲布尔网络(SDIBN)。使细胞死亡通常意味着将凋亡网络的轨迹驱动到表示细胞存活的状态之外。为实现这一目标,本文重点研究了SDIBNs集可达性的牵制控制器设计。首先,介绍了可达性和集可达性的定义,并说明了它们之间的关系。为判断SDIBN的轨迹是否离开不良状态,根据集可达性准则给出了一个充要条件。此外,还提供了一系列算法来寻找集可达性的所有可能的牵制节点集。注意,研究包含所有不良状态的吸引子,以使SDIBNs通过控制最小状态实现集可达。为确定一步集可达性的牵制节点,在状态的标量形式下给出了汉明距离。通过删除一些冗余节点,得到了集可达性的最小基数的牵制节点。与现有结果相比,无需求解逻辑矩阵方程即可获得状态反馈增益。所提方法的计算复杂度低于现有方法。此外,利用牵制控制器的设计方法对凋亡网络进行了讨论。实验结果表明,凋亡网络仅通过控制一个节点就能离开不良状态。