Research On Medical CPS Collaborative Computing Model And Related Problems

The medical information systems have entered the stage of intelligent,and medical CPS are the representative systems at this stage.Medical CPS are complex systems in which the complex medical process and complex life system interact closely.They combine medical knowledge with engineering applications to create personalized intelligent medical technology.However,with the improvement of the economic level,isolated smart devices and simple sharing of medical resources can no longer meet people's demand for high-quality medical services.Precision medicine has become the desired goal and has risen to the national strategy.Although medical CPS introduce intelligent units in the traditional medical closed loop,the lack of comprehensiveness and flexibility of its knowledge base often leads to misdiagnosis and mistreatment of complex diseases.At the same time,the refinement of medical specialties also led to the need for multi-disciplinary collaboration when specialists faced complex diseases.Therefore,the introduction of collaborative computing technology to deeply share and integrate the advantageous resources of the network space can reduce the disadvantages of single model,and improve the medical accuracy of medical CPS through collaborative computing.Aiming at the problem of medical CPS collaborative computing,this paper studies related issues such as system model and collaborative architecture.The main research work and innovations are summarized as follows:(1)Research on medical CPS collaborative computing model.The medical CPS collaborative computing system integrates the medical resources of multiple medical institutions through the network.The collaboration between medical CPS provides patients with continuous high-quality medical services.Collaborative computing nodes,the model of medical CPS collaboration,and the model of collaborative computing network are defined.The collaborative computing node is composed of a medical CPS cluster and a cooperative controller deployed in this cluster.Many collaborative computing tasks and calculation results are passed between the collaborative computing source node and the collaborative computing target node,the calculation results are returned to the collaborative computing source node according to the original path or the new communication path.Because the evolutionary trend of medical CPS collaborative computing network has the characteristics of scale-free networks,the dynamic equations of collaborative computing nodes are defined by combining the features of scale-free networks and the characteristics of two-way closed loops of medical CPS.(2)Research on medical CPS cooperation framework.Network is one of the important support for medical CPS collaborative computing.This paper combines the control architecture of SDN network with complex networks such as medical CPS collaborative computing,and proposes an SDN-based medical CPS collaborative computing network architecture.The network architecture is divided into three layers,which are the foundation device layer,control layer and application layer.In the control layer,there are control level and decision level.The corresponding basic equipment are cooperative controllers and decision controllers.The combination of a collaborative controller and a medical CPS cluster forms a medical CPS collaborative computing node.Therefore,the problem of generating medical CPS collaborative computing nodes translates into medical CPS clustering and the deployment of collaborative controllers.Aiming at this problem,the HCD algorithm is proposed in this paper,and the CCD algorithm is proposed for the deployment of decision controllers.Finally,simulation experiments show the effectiveness of HCD algorithm in generating medical CPS collaborative computing nodes,and the HCD algorithm greatly improves the load balancing of the cooperative controllers at the expense of less communication delay.With fewer cluster nodes,the CCD algorithm has a better optimization effect on the three indicators than the HCD algorithm,so it is more suitable for the deployment of decision controllers.(3)Research on collaborative computing targets search problem.In the collaborative computing model,collaborative tasks and calculation results are transmitted on the path between the source node and the target nodes.Solving the problem of collaborative computing target search is to establish communication paths from the collaborative computing source node to the collaborative computing target nodes.This paper proposes the LTS algorithm to establish communication paths from source node to multiple target nodes.In addition,in order to perform the synchronization stability analysis of the cooperative computing network,this paper solves the synchronization stability analysis parameters according to the communication paths,which are external coupling matrices.Considering the network topology change factor,this paper proposes a dynamic processing strategy(ECMCP)to ensure the Communication path updated in time and matrices have the same dimensions.The experiment proves the effectiveness of the LTS algorithm and the LT S & EC MCP strategy,and it has the advantages of less resource occupation and low communication cost.(4)Research on synchronization of collaborative computing.This paper studies the medical CPS collaborative computing network from the perspective of complex networks.The medical CPS collaborative computing network has the characteristics of two-way multiple closed loops.Synchronous stability analysis is performed to ensure the network service quality and ensure the successful cooperation between the collaborative computing nodes.Based on the three-node dynamic equation defined in the collaborative computing network model,this paper uses the Lyapunov synchronization stability theorem to solve the synchronization stability criterion of the collaborative computing network.Finally,the validity of the synchronization stability criterion is verified by experiments.