Mobile Location in Wireless Sensor Networks Based on Multi Spot Measurements Model

Chao Zheng, Wei Hu, Jiyan Huang, Pengfei Wang, Yufei Liu, Chenyu YangSensors 2022 · DOI

TL;DR

This paper introduces a novel approach to locating mobile sensors in wireless sensor networks (WSNs) by leveraging "multi-spot" measurements, which capture a sensor's position over time. This method significantly improves accuracy compared to traditional "one-spot" methods, offering both a computationally efficient closed-form solution and a highly accurate iterative solution that approaches the theoretical best possible accuracy.

What Problem Does This Paper Address?

Accurate localization of sensors is crucial for many WSN applications, from military surveillance to health monitoring. Existing methods, however, primarily rely on a "one-spot" measurement model, meaning they take a single snapshot of measurements to estimate a sensor's location. While these methods have been extensively studied, their accuracy is reaching a plateau, making it difficult to achieve further improvements. This paper addresses this limitation by proposing a "multi-spot" measurement model, which utilizes a sequence of measurements from a moving sensor to enhance localization precision.

Methods

The core innovation lies in the multi-spot measurements model. Instead of a single measurement, this model considers a series of range measurements (specifically, Time of Arrival - TOA) taken from multiple reference nodes (RNs) over a short period. This allows the system to not only estimate the sensor's current position but also its velocity, by assuming a simple linear motion model for the mobile sensor.

The paper proposes two main algorithmic approaches:

  • Closed-Form Solution (Two-Step WLS): This method linearizes the problem by introducing new variables representing combinations of position and velocity components. It then employs a two-step Weighted Least Squares (WLS) estimator to find a solution. This approach is computationally less demanding and suitable for real-time applications, though it might be slightly suboptimal due to the linearization.
  • Iterative Solution (MLE): This method aims for higher accuracy by using a Maximum Likelihood Estimator (MLE) and an iterative Newton-Raphson approach. It maximizes the likelihood function derived from the multi-spot measurements. This method is more computationally intensive but can achieve performance closer to the theoretical best.

A significant theoretical contribution is the derivation of a novel Cramer-Rao Lower Bound (CRLB) specifically for the multi-spot measurement model. This CRLB serves as a benchmark for the best possible accuracy achievable with this model.

Datasets

Not discussed in the paper. The evaluation relies on simulated data generated according to the described system model.

Key Insights and Ideas

  • Leveraging Motion for Accuracy: The fundamental insight is that by observing a sensor's movement over time (multi-spot measurements), we gain more information than a single snapshot. This additional information, particularly about the sensor's velocity, significantly boosts localization accuracy.
  • Theoretical Bound Improvement: The derivation of a new CRLB for the multi-spot model demonstrates that the theoretical limit of accuracy is indeed lower (i.e., better) than that of the traditional single-spot model. This provides strong theoretical backing for the proposed approach.
  • Trade-off Between Speed and Accuracy: The paper effectively presents two solutions catering to different needs: a fast, closed-form method for real-time scenarios and a more accurate, iterative method for applications where precision is paramount.
  • Initial Guess Matters for Iterative Methods: The simulations highlight the importance of a good initial guess for the iterative method. Using the closed-form solution to provide this initial guess leads to more stable and accurate convergence.

Key Results

  • Reduced RMSE: Simulations show that both the proposed closed-form and iterative methods achieve a lower Root Mean Square Error (RMSE) compared to traditional single-spot TOA localization methods.
  • Iterative Method Attains CRLB: The proposed iterative method demonstrates the ability to reach the derived CRLB, indicating it achieves near-optimal performance.
  • CRLB Comparison: The proposed CRLB for the multi-spot model is consistently lower than the traditional CRLB for the single-spot model, confirming the theoretical advantage of the new approach.
  • Impact of Measurement Moments (M): Increasing the number of measurement moments (M) in the multi-spot model leads to improved accuracy for the proposed methods, while the traditional methods show less improvement as they don't benefit from the additional temporal data.
  • Robustness: The RMSE of the proposed method is less volatile than the traditional method, suggesting better stability, especially with an increasing number of measurements.

Limitations and Future Work

  • Linear Motion Assumption: The proposed methods assume a simple linear motion model for the mobile sensor. This might not hold true for highly dynamic or erratic movements, which could limit accuracy in such scenarios.
  • Noise Distribution: The CRLB derivation relies on the assumption of Gaussian noise. While the methods are evaluated under different noise levels, the theoretical analysis might not directly apply to other noise distributions.
  • Computational Cost of Iterative Method: While the iterative method offers the highest accuracy, its computational demands might be a concern for resource-constrained WSN nodes.

The authors suggest that the proposed methods can be extended to other localization techniques beyond TOA, and future work could explore more complex motion models or different noise characteristics. The hybrid approach of using the closed-form solution to initialize the iterative method is also a promising direction for achieving both speed and accuracy.

Summary updated

Abstract

The localization of sensors in wireless sensor networks has recently gained considerable attention. The existing location methods are based on a one-spot measurement model. It is difficult to further improve the positioning accuracy of existing location methods based on single-spot measurements. This paper proposes two location methods based on multi-spot measurements to reduce location errors. Because the multi-spot measurements model has more measurement equations than the single-spot measurements model, the proposed methods provide better performance than the traditional location methods using one-spot measurement in terms of the root mean square error (RMSE) and Cramer-Rao lower bound (CRLB). Both closed-form and iterative algorithms are proposed in this paper. The former performs suboptimally with less computational burden, whereas the latter has the highest positioning accuracy in attaining the CRLB. Moreover, a novel CRLB for the proposed multi-spot measurements model is also derived in this paper. A theoretical proof shows that the traditional CRLB in the case of single-spot measurements performs worse than the proposed CRLB in the case of multi-spot measurements. The simulation results show that the proposed methods have a lower RMSE than the traditional location methods.
BibTeX
@inproceedings{zheng2022mobile,
  author = {Chao Zheng and Wei Hu and Jiyan Huang and Pengfei Wang and Yufei Liu and Chenyu Yang},
  title = {{Mobile Location in Wireless Sensor Networks Based on Multi Spot Measurements Model}},
  booktitle = {Sensors},
  year = {2022},
  doi = {10.3390/s22239559},
}
Figure