Abstract:Because of the advantages of computation complexity compared with traditional localization algorithms, fingerprint based localization is getting increasing demand. Expanding the fingerprint database from the frequency domain by channel reconstruction can improve localization accuracy. However, in a mobility environment, the channel reconstruction accuracy is limited by the time-varying parameters. In this paper, we proposed a system to extract the time-varying parameters based on space-alternating generalized expectation maximization (SAGE) algorithm, then used variational auto-encoder (VAE) to reconstruct the channel state information on another channel. The proposed scheme is tested on the data generated by the deep-MIMO channel model. Mathematical analysis for the viability of our system is also shown in this paper.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to improve the reconstruction accuracy of Channel State Information (CSI) in a mobile environment in multi - band fingerprint positioning, thereby enhancing the positioning accuracy. Specifically, the article proposes solutions to the following problems: 1. **Efficient Characterization of Mobile Conditions**: Traditional positioning methods assume that all entities share the same mobile environment, but in practical applications, the mobile conditions of different entities may be different. Therefore, a method is required to efficiently extract and characterize these different mobile conditions. 2. **Cross - band Mobile - independent Feature Extraction**: In order to reduce the complexity of multi - band sampling, the author proposes a method based on Variational Auto - Encoder (VAE) to infer CSI in other bands. However, the non - linear correlation between different bands and the change of mobile conditions make this task more complex. Therefore, a method is needed to extract cross - band mobile - independent features. 3. **Using Fingerprint Diversity to Improve Positioning Accuracy**: By jointly using CSI in multiple bands, the positioning accuracy can be improved. However, how to effectively use this fingerprint diversity is a challenge. To solve the above problems, the author proposes a novel cross - band CSI prediction scheme, which combines the Space - Alternating Generalized Expectation - Maximization (SAGE) algorithm and Variational Auto - Encoder (VAE). The specific steps are as follows: - **Mobile Parameter Estimation**: Use the improved SAGE algorithm to extract mobile parameters from channel information. - **Cross - band CSI Reconstruction**: Use the VAE network to reconstruct CSI in other bands and remove the influence of mobility through a neural network. - **Fingerprint Diversity Utilization**: Through cross - band CSI reconstruction, increase the diversity of the fingerprint database, thereby improving the positioning accuracy. Through these methods, the author has verified the effectiveness of this scheme in experiments, especially being able to maintain high positioning accuracy at different moving speeds.

A Novel Cross-band CSI Prediction Scheme for Multi-band Fingerprint based Localization

A Variational Auto-Encoder Enabled Multi-Band Channel Prediction Scheme for Indoor Localization

Channel Estimation and Localization for Mmwave Systems: A Sparse Bayesian Learning Approach

Time-Varying Channel Prediction for RIS-Assisted MU-MISO Networks Via Deep Learning

Broad Learning System for Indoor CSI Fingerprint Localization

CSI of Each Subcarrier is a Fingerprint: Multi-Carrier Cumulative Learning Based Positioning in Massive MIMO Systems

Fingerprint-based Localization using Commercial LTE Signals: A Field-Trial Study

A CSI-Based Indoor Fingerprinting Localization with Model Integration Approach

CSI-based Fingerprinting for Indoor Localization: A Deep Learning Approach

Multi-Fingerprints Indoor Localization for Variable Spatial Environments: A Naive Bayesian Approach

Multi-channel Fingerprint Localisation Algorithm for Wireless Sensor Network in Multipath Environment

CSI Fingerprinting for Device-Free Localization: Phase Calibration and SSIM-Based Augmentation

A real-time algorithm for fingerprint localization based on clustering and spatial diversity

Indoor Localization Based on CSI Fingerprint by Siamese Convolution Neural Network

A novel real-time channel prediction algorithm in high-speed scenario using convolutional neural network

Toward Long-Term Effective and Robust Device-Free Indoor Localization via Channel State Information

Vehicular Localization Based on CSI-Fingerprint and Vector Match

A Novel Approach for Fingerprint Positioning Based on Spatial Diversity

CRISLoc: Reconstructable CSI Fingerprintingfor Indoor Smartphone Localization

Transformable Fingerprinting with Deep Metric Learning Approach for Indoor Localization

CRISLoc: Reconstructable CSI Fingerprinting for Indoor Smartphone Localization