Congestion Control To Support M2M Communication In Cellular Networks

The Internet of Things is developing rapidly these years,under this background,Machine-to-Machine communication is applying widely,including vehicle services,smart meters,home and industrial monitoring,etc.M2M communication has become popular in communication industry and has broad market prospect.As the next generation cellular network LTE has a wide range of seamless coverage,as well as low deployment costs and fast speed,it will become the ideal platform for carrying M2M services.The existing cellular network protocols and architectures are based on Human-to-Human communication,but M2M communication has massive terminal devices and burst data,these characteristics are very different from H2H communication.M2M services will bring tremendous pressure and challenges to the cellular network,among them,network congestion,especially the congestion of LTE random access process is a serious problem.How to optimize LTE random access process for M2M services and solve the random access congestion is the keystone of this paper.The main research issues and innovations are as follows:For the scene of massive M2M devices accessing LTE network,the characteristics of M2M and the competitive random access process of LTE are studied.On this basis,this paper studies the impact of massive synchronous M2M services on LTE system and 3 GPP optimization of LTE random access congestion,including ACB,EAB,slot access,backoff scheme,Pull based scheme,resource allocation,and so on.Based on 3GPP research reports,the model of M2M accessing LTE network is illustrated in this paper,also the throughput,average access success rate and delay are derived.On this basis,simulation of random access process and comparison of several congestion control schemes,including ACB,dynamic access barring and grouping based on M2M burst data are analyzed.The congestion control algorithm is generally divided into two stages:conflict avoidance and conflict resolution.Most of the existing schemes are trying to avoid conflicts at the price of time,but there are few conflict solutions for delay-sensitive services.When urgent services arrive,it is very important to control access delay.It can't work well just by conflict avoidance,and solutions after the conflicts are the key to improve wireless network QoS.Based on the Tree-splitting algorithm of good delay characteristic in slotted Aloha system,we improve the fixed Tree-splitting algorithm,and the dynamic channel resource allocation algorithm is introduced by means of estimating the load of channel.An algorithm of dynamic random access conflict solution based on Tree-splitting is proposed in this paper.The simulation results verify that the dynamic Tree-splitting algorithm is suit for delay-sensitive services and fits the dynamic traffic better.Finally,the researches of this paper are summarized,and the future research directions related to this paper are also given.