Research On The Localization Algorithms And Their Performance Optimization In Wireless Sensor Networks

With the development of Internet of Things(IoT)and Internet of Everything(Io E)technology and continuous promotion of industrialization,location-based services are playing an increasingly important role in the field of intelligent medical,intelligent agriculture,intelligent internet of vehicles,military reconnaissance,building monitoring etc.Therefore,how to achieve the high precision,high coverage and low energy overhead location services is a hot research issues in the field of wireless networks,Internet of Things,internet of vehicles and so on.According to the practical problems of localization service in Wireless Sensor Networks(WSN),with different application scenarios as the starting point,we proposed optimization scheme of single location performance based on intelligent algorithm,optimization scheme of multiple location performance based on hybrid algorithm,optimization scheme of location performance based on dynamic parameter configuration.These schemes are adapted to different application scenarios with different performance requirements,which improve the application effect of WSN localization algorithm.Our contributions are as follows:1.For the application scenarios that requires high of single performance in location algorithm,we proposed two localization performance optimization solutions based on intelligent algorithms.They are PD-DVHop(Particle Swarm Optimization Differential Evolution DV-Hop)localization algorithm and SQPSODV-Hop(Standard Quantum Particle Swarm Optimization DV-Hop)localization algorithm.Firstly,the PD-DVHop localization algorithm is proposed based on the typical DV-Hop localization algorithm,and the differential evolution particle swarm optimization(PSODE)algorithm is applied to the DV-Hop localization algorithm.Secondly,the quantum particle swarm optimization(QPSO)algorithm is further applied to DV-Hop localization algorithm,and the SQPSODV-Hop localization algorithm is proposed.Finally,through theoretical analysis and simulation experiment,the localization error rate,communication overhead,computation complexity and iteration number are compared and analyzed between PD-DVHop localization algorithm,SQPSODV-Hop localization algorithm and the classical DV-Hop localization algorithm.The results show that PD-DVHop and SQPSODV-Hop localization algorithm achieve good results in single performance optimization,and can reduce the communication overhead of the algorithm.Moreover,compared with the PD-DVHop localization algorithm,the SQPSODV-Hoplocalization algorithm can converge globally in the case of very few iteration numbers without increasing communication overhead.The experimental results show that the localization algorithms based on the intelligent algorithm have good performance in the optimization of single performance.2.In view of the application scenarios of balancing the several performances of localization algorithms according to requires,a localization performance optimization solution based on hybrid algorithm(CCHL localization algorithm)is proposed.For a single localization algorithm,the results of multiple performance optimization are different under the same conditions,therefore,two algorithms are combined to obtain a hybrid localization algorithm that can be applied to application scenarios where performances are reqired to balance.Firstly,the centroid localization algorithm(Centroid)is improved,and the WCL(Weighted Centroid Localization)algorithm with higher localization accuracy is obtained.Then,WCL and CESLIA localization algorithms are combined to obtain the CCHL(CESILA and Weiged Centroid Algorithm Hybrid Range-free Localization)algorithm.The algorithm optimized several performances of the localization algorithm,and according to the principle of the algorithm,several performances of the algorithm are analyzed,including localization coverage,error rate and energy consumption.Finally,the simulation experiments verified that the proposed hybrid localization algorithm has good results in the optimization of several performances.3.To solve the application scenarios of nonuniform distribution of nodes,we proposed a localization performance optimization solution based on dynamic parameter configuration(HTCRL and DVQR localization algorithm).Compared with the classical APIT algorithm,HTCRL(Homothetic Triangle Cyclic Refinement Location)algorithm achieved higher localization accuracy in the condition of different parameter configurations including R,AH,ANR etc.without additional hardware equipment and increasing communication overhead.The performance of DVQR(DV-Hop Localization Algorithm based on QR Decompose)localization algorithm is better than that of classical DV-Hop algorithm under the same parameter condition,the accuracy of numerical calculation and the stability of operation is improved.The localization performance of HTCRL and DVQR localization algorithm can be optimized by dynamic parameter configuration.The experimental results show that the localization optimization algorithms of dynamic parameter configuration have good performance in the application scenarios of nonuniform distribution of nodes.In summary,in this paper,according to different application scenarios we studied optimization scheme of single location performance based on intelligent algorithm,optimization scheme of multiple location performance based on hybrid algorithm,optimization scheme of location performance based on dynamic parameter configuration.The simulation results show that the performances of the algorithms proposed in this paper have achieved the expected effect.