Intelligent Drug Box System Based On Cloud Computing Under The Internet Of Things Platform

At present,the ageing problem of the chinese population has caused widespread concern.Therefore,the design of medical aids for the elderly has gradually become a research focus.The problem that the elderly often forget to take medications can be solved by the existing smart kits.However,the processing of data on medication plans needs to be improved.With the increase in the number of users on the use of smart kits,the processing of their medication data becomes more and more complex.The use of cloud computing under the Internet of Things platform can effectively accomplish the task which has huge data,and efficient task scheduling can be achieved by adopting a suitable optimization algorithm.Greatly shortening the task completion time and improving the resource utilization efficiency of cloud computing have always been goals which are pursued by cloud computing researchers.A cloud-based intelligent kit system under the Internet of Things platform is designed in this dissertation.The kit can be connected with a mobile phone via Bluetooth,the mobile phone sends a medication kit medication plan via Bluetooth,and then the medication record is transmitted to the mobile phone APP via Bluetooth by the smart kit,and then the mobile phone uploads plans to the cloud server,and then the data is processed by cloud computing in the cloud server.In cloud computing task scheduling,an improved particle swarm optimization algorithm is proposed to shorten the task completion time.The main work of the dissertation and its achievements are as follows:(1)The hardware is designed that nRF52832 Bluetooth chip is the processing core,and the hardware also includes the power supply module,switch module,antenna module,storage module,vibration module,voice module,and GPRS module,the structure and the function of each module are introduced in detail.At the same time,on the basis of the existing Bluetooth SDK,a set of software which is compatible with various modules of the smart kit is developed.The task scheduling method is used in the software,the corresponding task will be executed once the event triggers,and the kit will enter the sleep state at other times.As a result,the usage time of the smart kit is greatly improved.(2)In order to solve the complex issues of data processing for user-uploaded medication records,An IoT application system in this dissertation is built for modern medical care through the research of a new generation of information technology,the system is deployed on a unified server cluster with the cloud computing platform.Virtual technology is used to reduce the need for server hardware,network security devices,and the support for software upgrades and maintenance.At last,users can obtain the software and services they need through personal computers and the Internet.The amount of data on this server cluster is particularly large,so it is necessary to select a more appropriate cloud scheduling algorithm in cloud computing.Various dynamic scheduling algorithms are compared and analyzed,and the particle swarm algorithm is choosed in this dissertation,this algorithm has good convergence and more reliable degree of freedom,but it is easy to fall into local convergence,so the algorithm needs to be improved.(3)Aiming at the problem that the particle swarm algorithm is easy to fall into local convergence,an improved particle swarm optimization algorithm with decreasing linear inertia weight is proposed.First of all,constant perturbation,on the basis of linear decreasing of inertia weight,is added in this dissertation to increase the inertia weight so that it can jump out of local search and perform global search to prevent local convergence.Secondly,To avoid the high concentration of particles around the optimal particle in particle swarm algorithm,which causes the particles to tend to be the same and the diversity of the particle swarm is greatly lost,inertia weight is changed adaptively and mixed in random individuals in order to better maintain population diversity.Finally,several different particle swarm algorithms are used for simulation experiments under Matlab platform.Simulation results show that the improved particle swarm algorithm can find a more accurate solution under the same conditions.