Analysis Of Overflow And Idleness Of AP Buffer In Home IoT Based On SFM

Nowadays,with the rapid development of the Internet and information technology,the re-search on the Internet of Things(IoT)has received extensive attention.China has included the IoT as one of the country's five emerging strategic industries.The IoTs connect objects to the Internet according to a specific network protocol,interacts with data,and connects things.With the im-provement of people's living standards,the IoT is gradually being applied to the family,such as the security system,lighting control system,temperature monitoring system,etc.in the family.The application of the IoT technology in the family makes people's lives more safety,convenience,and greatly improved people's quality of life.In the home IoT network,multiple IoT devices share a local area network(WLAN),and the data generated by the IoT device communicates with the server through a WiFi Access Point(AP)buffer.Since the capacity of the AP buffer is finite and Internet traffic is bursty,if the capacity of the AP buffer is too small and the buffer does not have enough space to accommodate more upcoming data,the data cannot be processed in time by the system,it will cause the buffer overflow,i.e.,the loss of data;if the buffer capacity is too large,it will affect the data transmission speed,and easily cause the buffer to be idle,which will reduce the efficiency of the buffer equipment.The occurrence of these two situations will have a negative impact on the system and will present a difficulty in meeting the quality of service(QoS)requirements of the home IoT network,so the home IoT system still face many challenges.This work proposes a home IoT system model based on the framework of Stochastic Fluid Model(SFM),analyzes the overflow and idle probabilities of the AP buffer in the home IoT,and proposes the exact algorithm for the overflow and idle probabilities.We first truncate the sample path of the SFM,and then the transition rate matrix of the truncation process is derived by using the probability matrix of First Passage Time(FPT),the overflow and the idle probabilities of the AP buffer are obtained by the Matrix Analytic Method.and then numerical example is performed,as an application,given the maximum tolerable overflow probability,the optimal numerical settings for the number of devices and buffer capacity are obtained.Since the transient dynamics of the system are very important to the system,this work also considers two time dependent performance metrics.Based on the SFM framework,FPTs of the AP buffer in the home IoT system are analyzed:The FPT when the data in the AP buffer overflows and the FPT when the AP buffer is idle,respectively.we obtain the Laplace-Stieltjes Transforms(LSTs)of the two FPTs,from which n-order moments and the probability density functions of the FPTs can be further derived.By some numerical examples,we validate the theoretical results,and further discuss the effects of the system parameters on the FPTs in design.The theoretical results in this paper have practical significance for the design and management of the home IoT system.