Enhanced Fingerprinting Based Indoor Positioning Using Machine Learning

Due to the inability of the Global Positioning System(GPS)signals to penetrate through surfaces like roofs,walls,and other objects in indoor environments,numerous alternative methods for user positioning have been presented.Amongst those,the Wi-Fi fingerprinting method has gained considerable interest in Indoor Positioning Systems(IPS)as the need for lineof-sight measurements is minimal,and it achieves better efficiency in even complex indoor environments.Offline and online are the two phases of the fingerprinting method.Many researchers have highlighted the problems in the offline phase as it deals with huge datasets and validation of Fingerprints without pre-processing of data becomes a concern.Machine learning is used for the model training in the offline phase while the locations are estimated in the online phase.Many researchers have considered the concerns in the offline phase as it deals with huge datasets and validation of Fingerprints becomes an issue.Machine learning algorithms are a natural solution for winnowing through large datasets and determining the significant fragments of information for localization,creating precise models to predict an indoor location.Large training sets are a key for obtaining better results in machine learning problems.Therefore,an existing WLAN fingerprinting-based multistory building location database has been used with 21049 samples including 19938 training and 1111 testing samples.The proposed model consists of mean and median filtering as pre-processing techniques applied to the database for enhancing the accuracy by mitigating the impact of environmental dispersion and investigated machine learning algorithms(kNN,WkNN,FSkNN,and SVM)for estimating the location.The proposed SVM with median filtering algorithm gives a reduced mean positioning error of 0.7959 m and an improved efficiency of 92.84%as compared to all variants of the proposed method for 108703 m^(2) area.

A Robust Wi-Fi Fingerprinting Indoor Localization Coping with Access Point Movement

A Wi-Fi fingerprinting localization approach has attracted increasing attention in recent years due to the ubiquity of Access Point( AP). However,typical fingerprinting localization methods fail to resist accidental environmental changes,such as AP movement. In order to address this problem,a robust fingerprinting indoor localization method is initiated. In the offline phase,three attributes of Received Signal Strength Indication( RSSI) —average,standard deviation and AP's response rate—are computed to prepare for the subsequent computation. In this way,the underlying location-relevant information can be captured comprehensively. Then in the online phase, a three-step voting scheme-based decision mechanism is demonstrated, detecting and eliminating the part of AP where the signals measured are severely distorted by AP 's movement. In the following localization step,in order to achieve accuracy and efficiency simultaneously,a novel fingerprinting localization algorithm is applied. Bhattacharyya distance is utilized to measure the RSSI distribution distance,thus realizing the optimization of MAximum Overlapping algorithm( MAO). Finally,experimental results are displayed,which demonstrate the effectiveness of our proposed methods in eliminating outliers and attaining relatively higher localization accuracy.

Received signal strength based indoor positioning algorithm using advanced clustering and kernel ridge regression

We propose a novel indoor positioning algorithm based on the received signal strength(RSS) fingerprint. The proposed algorithm can be divided into three steps, an offline phase at which an advanced clustering(AC) strategy is used, an online phase of approximate localization at which cluster matching is used, and an online phase of precise localization with kernel ridge regression. Specifically, after offline fingerprint collection and similarity measurement, we employ an AC strategy based on the K-medoids clustering algorithm using additional reference points that are geographically located at the outer cluster boundary to enrich the data of each cluster. During the approximate localization, RSS measurements are compared with the cluster radio maps to determine to which cluster the target most likely belongs. Both the Euclidean distance of the RSSs and the Hamming distance of the coverage vectors between the observations and training records are explored for cluster matching. Then, a kernel-based ridge regression method is used to obtain the ultimate positioning of the target. The performance of the proposed algorithm is evaluated in two typical indoor environments, and compared with those of state-of-the-art algorithms. The experimental results demonstrate the effectiveness and advantages of the proposed algorithm in terms of positioning accuracy and complexity.

一种WiFi混合滤波的RSSI室内定位算法

针对接收信号强度指示(RSSI)在基于WiFi的室内定位应用中易受到环境影响出现波动、突变进而导致定位精度降低的问题,提出了一种WiFi混合滤波的RSSI室内定位算法。首先,根据WiFi信号距离衰减模型设计一种自适应阈值滤波器剔除观测信号异常值;然后,根据自适应阈值滤波、卡尔曼(Kalman)滤波、均值(Mean)滤波设计了一种混合滤波算法(ATKM)对RSSI数据进行滤波;最后,提出了一种基于斯皮尔曼-欧几里得距离的加权K近邻算法(SEWKNN)进行位置估计。在真实环境下的实测结果表明,本文提出的ATKM滤波算法能显著降低RSSI信号的波动,提出的SEWKNN算法在室内环境中平均定位误差为1.17 m,在走廊环境中平均定位误差为1.53m,相比传统的WKNN算法平均定位误差分别下降18.18%和16.84%。

基于自适应距离ADWKNN室内定位算法

针对基于蓝牙接收信号强度(RSS)的加权K-近邻(WKNN)位置指纹室内定位算法中由于信号波动存在的单一距离估算标准导致定位精度偏低的问题,提出了一种基于自适应距离的ADWKNN定位算法。离线阶段利用K-means聚类算法对指纹数据库进行划分,减少数据查询量以保证定位的时效性。在线定位阶段首先对待定位点处采集到的RSS信号值进行卡尔曼滤波,减弱随机噪声的干扰;然后采用ADWKNN算法计算曼哈顿距离与欧氏距离的标准差,自适应的选择距离估算方法并实现K值的动态变化。实验结果表明,ADWKNN算法的平均定位精度为1.22 m,与使用余弦距离、曼哈顿距离、欧氏距离和Sorensen距离的单一距离的WKNN算法相比,ADWKNN算法的平均定位精度有明显提高。

基于聚类的快速Wi-Fi定位算法

Wi-Fi定位中用户移动终端计算能力差,提供的能量有限。为此,提出一种基于信号强度聚类的快速定位算法,依据接收信号强度的相似度对参考点分组再定位,从而减少定位计算量,提高Wi-Fi定位精度。在校园和市内2种不同室外环境下进行现场实验,数据显示定位均方根误差从16.71 m减小到9.68 m,定位时间从96.3 ms减小到5.2 ms,验证了该算法的高效性和有效性。

基于稳定AP选择的动态室内定位方法

在室内复杂多变环境下,基于接收信号强度指示(Received Signal Strength Indication,RSSI)的位置指纹算法得到了广泛研究。其中,在线阶段的匹配算法通常采用加权K近邻(Weighted K-Nearest Neighbor,WKNN)算法,但该算法往往采用固定k值方法存在较大的定位误差,具有一定的局限性,并且离线阶段构建位置指纹数据库时并没有考虑到无线接入点(Access Point,AP)信号的波动性。因此,存在大量不同AP的冗余信息,对定位效果产生较大影响。针对上述问题,提出一种基于稳定AP选择的动态室内定位方法。首先,通过高斯滤波对RSSI值进行预处理,滤除随机干扰;然后,通过优选AP算法计算AP的稳定度,筛选出关键AP用于定位;最后,利用距离阈值动态调整k值,并对权重系数进行改善,实现了对WKNN算法的改进。实验结果表明,基于稳定AP选择的动态室内定位方法可以有效去除冗余AP信息,并实现动态k值方案,在定位精度上优于K近邻(K-Nearest Neighbors,KNN)算法、加权K近邻算法和改进的加权K近邻算法,平均定位误差分别降低了26.13%、21.29%和9.89%,定位误差在1.5 m内的累积分布概率达到了60.41%,分别提升了25%、16.66%和8.33%,定位效果提升明显。

Enhancing Indoor User Localization:An Adaptive Bayesian Approach for Multi-Floor Environments

Indoor localization systems are crucial in addressing the limitations of traditional global positioning system(GPS)in indoor environments due to signal attenuation issues.As complex indoor spaces become more sophisticated,indoor localization systems become essential for improving user experience,safety,and operational efficiency.Indoor localization methods based on Wi-Fi fingerprints require a high-density location fingerprint database,but this can increase the computational burden in the online phase.Bayesian networks,which integrate prior knowledge or domain expertise,are an effective solution for accurately determining indoor user locations.These networks use probabilistic reasoning to model relationships among various localization parameters for indoor environments that are challenging to navigate.This article proposes an adaptive Bayesian model for multi-floor environments based on fingerprinting techniques to minimize errors in estimating user location.The proposed system is an off-the-shelf solution that uses existing Wi-Fi infrastructures to estimate user’s location.It operates in both online and offline phases.In the offline phase,a mobile device with Wi-Fi capability collects radio signals,while in the online phase,generating samples using Gibbs sampling based on the proposed Bayesian model and radio map to predict user’s location.Experimental results unequivocally showcase the superior performance of the proposed model when compared to other existing models and methods.The proposed model achieved an impressive lower average localization error,surpassing the accuracy of competing approaches.Notably,this noteworthy achievement was attained with minimal reliance on reference points,underscoring the efficiency and efficacy of the proposed model in accurately estimating user locations in indoor environments.

基于深度神经网络的室内定位算法

提出了一种基于深度学习的接收信号强度指示(RSSI)指纹定位算法。该算法将深度神经网络引入到指纹定位的2个阶段中:离线阶段对不同遮挡情况的指纹库进行特征训练,其中指纹数据作为输入,不同遮挡情况的指纹库编号作为标签;在线阶段将实时接收到的数据送入到网络中进行指纹库匹配,然后结合改进的加权K近邻(WKNN)算法进行定位。通过对比实验结果表明:所提算法的定位精度优于其他定位算法,有着很好的定位性能。

基于Wi-Fi指纹的改进型室内定位算法研究

针对复杂多变的室内环境,导致运用Wi-Fi位置指纹定位算法定位的确定性和抗干扰能力较差,且定位精度及定位效率无法得到保证的问题,提出了一种改进型的位置指纹识别算法,将手持设备的方向角和接入点的数量加入离线指纹数据库,在线定位阶段采用K-means聚类算法与加权KNN算法相结合进行位置估计。通过实验验证得出,改进后的位置指纹定位方法与常规的位置指纹定位算法相比,定位精度得到明显提高,效率显著加快。

An Indoor Localization Approach Based on Fingerprint and Time-Difference of Arrival Fusion

In this paper,an effective target locating approach based on the fingerprint fusion posi-tioning(FFP)method is proposed which integrates the time-difference of arrival(TDOA)and the received signal strength according to the statistical variance of target position in the stationary 3D scenarios.The FFP method fuses the pedestrian dead reckoning(PDR)estimation to solve the moving target localization problem.We also introduce auxiliary parameters to estimate the target motion state.Subsequently,we can locate the static pedestrians and track the the moving target.For the case study,eight access stationary points are placed on a bookshelf and hypermarket;one target node is moving inside hypermarkets in 2D and 3D scenarios or stationary on the bookshelf.We compare the performance of our proposed method with existing localization algorithms such as k-nearest neighbor,weighted k-nearest neighbor,pure TDOA and fingerprinting combining Bayesian frameworks including the extended Kalman filter,unscented Kalman filter and particle fil-ter(PF).The proposed approach outperforms obviously the counterpart methodologies in terms of the root mean square error and the cumulative distribution function of localization errors,espe-cially in the 3D scenarios.Simulation results corroborate the effectiveness of our proposed approach.

基于接收信号强度指示的AP簇测距方法

对移动目标的高精度测距是室内定位的关键。室内环境中无线接入点(Access Point, AP)采集的移动目标设备的接收信号强度指示(Received Signal Strength Indication, RSSI)受阴影衰落波动严重。受到RSSI采集频率和目标机动能力的限制,AP采集到的RSSI样本量少,导致基于RSSI测距精度差。为提高基于RSSI对移动目标测距的精度,本文提出了基于RSSI的AP簇测距(AP Cluster Ranging, APCR)方法。该方法通过对多个AP进行位置约束组成AP簇采集移动目标设备的RSSI,在相同采集频率下可获得更多的RSSI样本。利用RSSI波动特点,使用最大值选取和Dixon检验相结合的方式从AP簇采集的RSSI样本中筛选出高质量的RSSI样本,以提高对移动目标的测距精度。仿真和实验结果表明,与传统RSSI处理方法相比,本方法在室内环境简单或复杂时都具有更高的测距精度,在少量RSSI采集次数下同样能保持较高精度,更能满足对移动目标测距的需求。

一种基于GACNN改进的室内可见光指纹定位算法

为提高室内可见光定位系统性能,提出了基于遗传算法训练卷积神经网络(Genetic Algorithm Convolutional Neural Network,GACNN)的室内可见光指纹定位算法。该算法引入一维卷积神经网络学习模型,针对卷积神经网络的超参数设置,利用遗传算法对卷积神经网络进行训练,将超参数进行二进制编码后采用精英遗传算法对CNN进行训练,来解决卷积神经网络模型参数调节依靠经验和模糊最优化的过程。实验结果表明:在室内4 m×4 m×2.5 m的定位场景下,定位算法可以获得平均定位误差4.11 cm的定位精度。相较于卷积神经网络定位算法,平均定位误差降低了25%。对比分析了不同室内可见光定位算法的性能,验证了算法的技术优势。

基于随机森林算法的室内可见光指纹定位方法

为进一步提高动态目标室内可见光定位追踪系统性能,提出了一种基于随机森林(RF)算法的室内可见光指纹定位方法。利用发光二极管(LED)的光强信号作为特征构建指纹数据库,应用指纹库中的数据训练决策树,引入RF算法进行初始定位,再通过卡尔曼滤波对初始位置估计进行优化,从而获得更准确的定位轨迹。仿真结果表明:在5 m×5 m×3 m的室内场景下,通过所提定位方法能获得大部分采样点误差分布在4 cm之内的定位效果;此外,通过与不同室内可见光定位算法的性能进行对比,验证了所提算法的技术优势。

基于DBSCAN与KNN的室内定位算法

针对位置指纹室内定位算法中定位精度不足、计算量大、实时性较差的问题,提出了一种将聚类算法(DBSCAN)应用于传统定位算法的解决方案。离线阶段通过各参考点的接收信号强度进行聚类,将定位区域划分为多个子区域,在每个子区域选择一个中心节点,构造新型位置指纹数据库存储数据。在现阶段,通过各个子区域中心节点进行粗定位,根据参考标签所在的子区域对传统定位算法进行权重优化。实验结果表明,平均定位误差为1.63 m。该算法在提升定位精度的同时,提升了定位的实时性。

基于空间位置约束的稀疏指纹室内定位方法

针对基于位置服务的实际应用需求,分析了现有室内定位技术的局限性,提出一种基于空间位置约束的稀疏指纹定位方法,在数据层有效融合惯导和无线局域网(WLAN)定位信息,充分发挥二者优势协同完成定位任务。首先利用WLAN提供的接收信号强度(RSS)信息构建空间位置指纹数据库,并基于RSS构建稀疏指纹表征与定位模型;鉴于RSS数据易受环境干扰呈现多变性,利用惯导技术对位移状态进行初步估计,并以此作为约束条件构建基于空间位置约束的稀疏指纹定位模型。仿真实验结果表明,所提方法较惯导和稀疏指纹方法在定位精度方面分别提升58%和33%。

WiFi指纹定位中AP个数对定位精度的影响

基于RSSI的WiFi指纹定位算法离线建立指纹数据库阶段受AP个数影响,因此AP个数也将影响到指纹定位算法精度。为了探究AP个数对定位精度的影响,文中在室内环境下进行实验,选取不同的AP进行基于RSSI的WiFi指纹定位实验和分析,定位精度和可靠性作为定位结果的衡量指标。实验结果表明:在单个办公室内,5~6个AP时定位精度较高且定位结果可靠性达到最高。

基于Zigbee室内定位系统的指纹库优化算法

Zigbee无线传感器网络的接收信号强度指示(RSSI)可以用来提供位置服务,使用RSSI建立指纹数据库设计定位算法能够得到良好的定位效果,但指纹数据库的采集和建立精度会直接影响到最终的定位精度,而一般方法建立的指纹数据库误差较大。为此,通过对Zigbee网络指纹数据库采集和建立过程进行研究,提出一种Zigbee网络中指纹数据库采集、建立及优化的算法。在定位区域将各采样点采集到的指纹数据库源数据进行滤波处理,建立高精度的指纹数据库。利用Zigbee平台组建无线网络进行实验,分别使用针对该定位系统提出的限定区域最邻近算法、限定区域加权最邻近算法、限定区域贝叶斯算法进行定位计算。实验结果表明,运用优化后的指纹数据库在短距离范围内定位平均误差可限制在1.5 m以内。

终端异质下位置指纹的鲁棒性研究

针对终端硬件差异对接收信号强度(RSS)测量的影响,导致传统RSS指纹鲁棒性较差的问题,借鉴信号强度差(SSD)在室内定位中的应用,提出一种抗移动终端硬件异质的SSD位置指纹。从理论上对SSD、RSS及双曲位置指纹(HLF)3种指纹的鲁棒性进行分析,并在实际无线局域网环境中应用传统K最近邻法,对3种指纹在训练定位阶段使用相同终端与不同终端2种情况下进行实验。结果证明,与RSS和HLF指纹相比,SSD指纹在抗移动终端异质方面的鲁棒性更好。

基于MNN改进粒子滤波的指纹库定位算法研究

考虑到基于Zigbee传感器网络信号强度指示的指纹库定位算法只适用于静态条件下的定位,无法有效去除受噪声污染严重的有害数据,定位精度不尽人意,根据噪声的分布特点,引入粒子滤波技术,优化了该算法,有效去除了受噪声污染的有害数据,实现了目标的动态精确定位跟踪。传统粒子算法在应用过程中对重要性分布函数做了高斯分布假设,使得定位精度受到影响,将多层神经网络引入滤波算法进行补偿优化。实验结果验证了改进算法的优越性,其在静态条件下平均定位误差小于1.14m,动态跟踪定位精度也有很大幅度的提高,具有实际应用价值。