mD-Track: Leveraging Multi-Dimensionality in Passive Indoor Wi-Fi Tracking (1812.03103v2)

Abstract: Wi-Fi localization and tracking face accuracy limitations dictated by antenna count (for angle-of-arrival methods) and frequency bandwidth (for time-of-arrival methods). This paper presents mD-Track a device-free Wi-Fi tracking system capable of jointly fusing information from as many dimensions as possible to overcome the resolution limit of each individual dimension. Through a novel path separation algorithm, mD-Track can resolve multipath at a much finer-grained resolution, isolating signals reflected off targets of interest. mD-Track can localize human passively at a high accuracy with just a single Wi-Fi transceiver pair. mD-Track also introduces novel methods to greatly streamline its estimation algorithms, achieving real-time operation. We implement mD-Track on both WARP and cheap off-the-shelf commodity Wi-Fi hardware and evaluate its performance in different indoor environments.

• The paper introduces mD, a passive indoor Wi-Fi tracking system that leverages multi-dimensional signal parameters (AoA, ToF, Doppler, AoD) to enhance resolvability and accuracy.
• Experimental findings show mD achieved a 3.5x accuracy improvement over prior approaches like SpotFi and localized over ten paths concurrently in diverse indoor environments.
• mD's ability to use commodity Wi-Fi devices without specialized hardware makes it a practical solution for real-world applications like security surveillance and elder care.

Overview of mD: Leveraging Multi-Dimensionality in Passive Indoor Wi-Fi Tracking

This paper presents mD, a sophisticated system for passive indoor Wi-Fi tracking focusing on maximizing the usage of multi-dimensional information to improve target localization accuracy. Traditionally, Wi-Fi localization has been restricted by the limitations inherent in individual dimensions like angle-of-arrival (AoA) and time-of-flight (ToF), defined by the number of antennas and channel bandwidth, respectively. The mD system addresses these limitations by using a multi-dimensional approach to enhance resolvability and parameter estimation, significantly boosting tracking precision.

Key Contributions and Methodology:

1. Multi-Dimensional Signal Estimator: mD introduces a signal processing structure capable of integrating multiple dimensions of signal parameters, including AoA, ToF, Doppler shift, and angle-of-departure (AoD). This allows for a finer resolution compared to traditional methods that look at dimensions in isolation.
2. Iterative Path Parameter Refinement: To isolate signals from multiple paths, mD employs an iterative process that refines path parameters by iteratively reconstructing and subtracting signals. This strategy is similar to the Expectation-Maximization algorithm, which ensures convergence and improved accuracy over successive iterations.
3. Computational Efficiency for Real-Time Operation: While the multi-dimensional approach increases computational demands, mD uses a coordinate descent method and expectation maximization to reduce complexity, enabling real-time application. By breaking down the dimensional search into iteratively optimized components, mD effectively manages the computational overhead.

Experimental Findings:

The implementation of mD on both specialized hardware like WARP and commodity Wi-Fi devices demonstrated significant improvements in tracking accuracy. Notably, the system achieved a $3.5\times$ accuracy improvement in passive localization over prior approaches like SpotFi, with a notable enhancement when incorporating additional dimensions like Doppler, showcasing an approximate $3\times$ accuracy boost. The experimental setup successfully localized more than ten signal paths concurrently, proving the system's robustness across diverse indoor environments.

Implications and Future Directions:

From a practical standpoint, mD's ability to leverage off-the-shelf hardware without additional cost-intensive equipment makes it a viable solution for various real-world applications, including security surveillance and elder care. Theoretically, mD's framework lays the groundwork for further research into multi-dimensional signal processing, potentially integrating other signal parameters or exploring analogous applications in non-Wi-Fi contexts.

The authors highlight potential advancements in passive Wi-Fi sensing; however, future extensions could refine computational algorithms further, possibly incorporating machine learning for adaptive parameter tuning. Another avenue is expanding this methodology to dynamic environments with more fluctuating multipath profiles, which could broaden the applicability of mD to more complex scenarios.

In conclusion, this paper presents a comprehensive method for improving passive Wi-Fi tracking by exploiting multi-dimensional signal parameters. It provides a substantive leap forward in enhancing resolvability and accuracy without necessitating unwieldy hardware modifications, offering a promising new direction for advancements in indoor localization technologies.