Research On The Improvement Of Load Balancing-based RPL Routing Protocol

The low power and lossy networks(LLN),which are composed of a large number of wireless sensor nodes with low power and limited processing capacity,are widely used in the Internet of things fields such as smart grid,smart community and industrial control.However,there are a large number of nodes and unstable links in LLN.Therefore,the Internet Engineering Task Force(IETF)has proposed an IPv6-based routing protocol for LLN(RPL)to meet a wider range of application requirements.In this thesis,based on the existing RPL routing algorithm,the load balancing mechanism in different scenarios are investigated and further improved.The main contents and working process are as follows:To address the problem that existing routing protocols in LLN cannot achieve load balancing well in the high load scenario due to the lossy characteristics of wireless links,a load balancing-based routing protocol for LLN under high-load scenario is proposed.This protocol mainly includes the following three aspects: firstly,parent node selection is optimized to avoid the alternative parent node with a larger number of expected transmissions being selected as the optimal parent node;secondly,the Trickle timer is improved when detecting the load of a node is heavy,the time interval for sending DIO control message of the node should be adjusted to reduce the probability of being selected as the best parent node by neighboring node;finally,a network congestion notification message(NCN)is designed to notify the current network congestion state timely when the network congestion occurs.To deal with the problems that existing routing protocols in LLN easily may cause network congestion and drastical control overhead increase in the process of load balancing adjustment when the load is unbalanced,a traffic adaptive assignment based multi-sink routing protocol for LLN(TAAM-RPL)is proposed.This protocol presents three new strategies: TAAM-RPL networking,zero additional control overhead for node load state publishment and adaptive data traffic allocation.TAAM-RPL protocol can reduce the probability of network congestion and improve the throughput of large-scale networks effectively without adding any additional control overhead.The simulation results show that,compared with the existing protocols,HSL-RPL protocol can achieve load balancing effectively,in which the network lifetime is increased by 23.57%,the average throughput of the sink-node is increased by 14.52%,and the normalized control overhead is lowered by 12.87%;for TAAM-RPL protocol,the probability of network congestion is reduced by 16.04%,the average throughput of the sink-node is increased by 14.89% and the normalized control overhead is lowered by 9.77%.