Research On Indoor Node Localization Algorithm Based On Optimized Neural Network

With the rapid development and application of 5g technology,the era of interconnection has come.At the same time,more and more software and hardware need to obtain location information through sensors and other devices,and the accuracy of location information is becoming more and more important.Because the traditional equipment positioning depends on the satellite signal,it has achieved good results in the case of outdoor without occlusion.But in the indoor scene,due to the barrier of various walls,many satellite positioning systems are difficult to meet the accuracy requirements of indoor positioning.Therefore,obtaining high-precision node location information and improving the accuracy of node location has become a research hotspot in the field of wireless sensor.Aiming at the problems of large error and poor stability of traditional positioning methods,this paper improves the accuracy and speed of indoor node positioning by using two methods of error back propagation(BP)neural network and intelligent optimization algorithm.Traditional schemes often use relatively poor performance intelligent optimization algorithm to optimize the indoor wireless signal attenuation model.Although it improves the positioning accuracy and has certain robustness,the convergence speed of the algorithm is slow and easy to fall into the local optimal value.In view of the above problems,this paper proposes to use whale optimization algorithm to optimize the indoor wireless signal attenuation model.Through the data in the self-built data set,we find the optimal parameters of the model,and optimize the conversion formula between the received signal strength indication(RSSI)and the corresponding distance in the indoor node positioning scene,Finally,the node position is calculated according to the optimization model,so as to improve the positioning speed and accuracy of the node.According to the simulation results,the localization effect of the whale optimization algorithm is compared with other optimization algorithms,and the effectiveness of the whale optimization algorithm in indoor node localization is verified.Although the signal attenuation model optimized by the optimization algorithm improves the positioning accuracy,it has the problems of insufficient accuracy and poor anti-noise ability.In view of the above problems,this paper proposes to use the combination of whale optimization algorithm and BP neural network to complete indoor positioning.By using the strong algorithm performance of whale optimization algorithm,adjust the parameters of BP neural network model,so as to improve the performance of BP neural network.Using the strong fitting ability of BP neural network for data,the mapping relationship between RSSI value and distance value of indoor nodes is fitted,so as to generate the corresponding wireless signal attenuation model,and calculate the node position according to the model,so as to improve the anti-interference ability of the model and reduce the indoor positioning error.The experimental results show that compared with the original neural network,the neural network optimized by whale optimization algorithm has a significant improvement in positioning accuracy and stability.