Indoor Localization Based on MSC Map
🤖 AI Summary
This paper addresses the low accuracy, hardware dependency, and requirement of prior maps in indoor cellular localization. We propose a lightweight localization paradigm based on Modulation and Coding Scheme (MCS) identification—introducing MCS as a fine-grained wireless environmental fingerprint for the first time. Leveraging deep learning models trained on raw IQ samples or spectrograms, we achieve robust MCS classification across heterogeneous multi-base-station cellular signals, enabling map-free, generalizable localization without prior site surveys. Localization is then performed via geographic mapping, interpolation, and weighted centroid or k-nearest neighbors (k-NN) estimation. Evaluated in typical indoor environments, our method reduces mean localization error by 37% compared to conventional RSSI-based approaches, while significantly improving robustness and generalizability—without requiring specialized hardware or pre-deployed radio maps.
Application Category
📝 Abstract
In this short paper, we propose a technique for AI-based identification of modulation and coding schemes (MCS) in surrounding cellular signals. Based on the created MCS map, we evaluate the performance of indoor localization techniques.
Problem
Research questions and friction points this paper is trying to address.
AI-based identification of cellular signal MCS
Creating MCS map for indoor localization
Evaluating indoor localization performance
Innovation
Methods, ideas, or system contributions that make the work stand out.
AI-based MCS identification in cellular signals
Creation of MCS map for localization
Performance evaluation of indoor localization
🔎 Similar Papers
No similar papers found.
