Multi-Dimensional Urban Sensing in Sparse Mobile Crowdsensing

Wenbin Liu, Yongjian Yang, En Wang, Leye Wang, Djamal Zeghlache, Daqing ZhangIEEE Access 2019 · OpenReview

Abstract

Sparse mobile crowdsensing (MCS) is a promising paradigm for large-scale urban monitoring, allowing for data collection from only a few areas (cell selection) to infer the state of the entire city (data inference). While traditional Sparse MCS focuses on single variables, modern urban applications often require multi-dimensional data, such as simultaneous temperature and humidity maps. This paper presents a novel framework that jointly addresses data inference and cell selection for multi-task scenarios by exploiting intra- and inter-task correlations. By combining multi-task compressive sensing with deep reinforcement learning (DRL), the proposed system significantly reduces sensing costs while maintaining high data accuracy.

Introduction

Mobile Crowdsensing (MCS) leverages the sensors in citizens' smartphones to monitor urban environments. However, recruiting enough users to cover every subarea (cell) of a city is prohibitively expensive. Sparse MCS addresses this by collecting data from only a fraction of cells and using algorithms to fill in the gaps.

A major limitation of current research is the focus on single-variable tasks. In reality, urban planners need multi-dimensional data. Because variables like temperature and humidity or PM2.5 and PM10 are physically correlated, treating them as independent tasks is inefficient. This paper explores how to leverage these correlations to optimize both the inference of missing data and the selection of which cells to sample.

The framework for multi-dimensional urban sensing, illustrating the two-stage process of cell selection and data inference across multiple variables.

Methodology

The authors introduce a two-pronged approach to handle multi-task urban sensing: multi-task compressive sensing for inference and a reinforcement learning (RL) model for cell selection.

1. Multi-Task Compressive Sensing (Inference)

Instead of traditional matrix completion for a single task, the framework uses Collective Matrix Factorization. This allows different tasks to share a "factor matrix," enabling the model to learn cross-task relationships—for example, high temperature in a cell might imply lower humidity, even if humidity wasn't directly measured there. The model incorporates three key constraints:

  • Temporal: Values change smoothly over time.
  • Spatial: Nearby cells are likely to have similar readings.
  • Value: Cells with similar environmental surroundings (e.g., two parks) likely have similar data profiles.

2. RL-Based Cell Selection (Optimization)

To determine which cells and tasks to prioritize, the authors utilize Deep Reinforcement Learning (DRL).

  • Multi-Branch Architecture: The system uses a unique neural network where separate branches extract "intra-task" features (patterns within one variable), which are then fused to capture "inter-task" correlations (patterns across variables).
  • Reward Mechanism: The RL agent is rewarded based on how much a specific selection improves the overall inference accuracy of the entire urban map.
  • Two-Stage Framework: To handle the cold-start problem of RL, the authors propose an online framework. A Training Phase uses uncertainty-based sampling to gather initial data, followed by a Running Phase where the trained RL agent takes over the selection process.

Experimental Evaluation

The framework was evaluated on two real-world urban sensing datasets:

  • Sensor-Scope: Environmental data (Temperature and Humidity) from 57 sensors at the EPFL campus.
  • U-Air: Air quality data (PM2.5 and PM10) from 36 sensors in Beijing.

Key Results

  • Efficiency Gains: The RL-assisted algorithm required 16% to 25% fewer sensed values than the state-of-the-art SPACE-TA algorithm to achieve the same target accuracy.
  • Inference Accuracy: Multi-task compressive sensing consistently yielded lower error rates than single-task methods, especially in high-sparsity scenarios where very few samples were available.
  • Real-time Performance: While training the RL model takes 10–30 minutes, the actual decision-making for cell selection takes less than 1ms, making it highly suitable for live deployments.

Key Insights

  • Correlations as a Force Multiplier: Inter-task correlations are often stronger than spatial ones. By measuring one variable (like PM2.5), the system gains significant information about another (PM10) without needing additional sensors in that same cell.
  • RL vs. Uncertainty: Most systems pick cells where the model is "most confused" (highest variance). This paper argues that being "confused" doesn't mean a cell is useful. The RL approach learns to identify "anchor cells"—roughly 30% of the grid—that are most representative of the whole system.

Conclusion

By exploiting the inherent relationships between different environmental variables, this multi-dimensional framework offers a more efficient path for large-scale urban monitoring. Future work will focus on integrating user mobility models and privacy-preserving mechanisms to make the system more robust for real-world crowdsensing deployments.

Summary updated

Abstract

Sparse mobile crowdsensing (MCS) is a promising paradigm for the large-scale urban sensing, which allows us to collect data from only a few areas (cell selection) and infer the data of other areas (data inference). It can significantly reduce the sensing cost while ensuring high data quality. Recently, large urban sensing systems often require multiple types of sensing data (e.g., publish two tasks on temperature and humidity respectively) to form a multi-dimensional urban sensing map. These multiple types of sensing data hold some inherent correlations, which can be leveraged to further reduce the sensing cost and improve the accuracy of the inferred results. In this paper, we study the multi-dimensional urban sensing in sparse MCS to jointly address the data inference and cell selection for multi-task scenarios. We exploit the intra- and inter-task correlations in data inference to deduce the data of the unsensed cells through the multi-task compressive sensing and then learn and select the most effective 〈cell, task〉 pairs by using reinforcement learning. To effectively capture the intra- and inter-task correlations in cell selection, we design a network structure with multiple branches, where branches extract the intra-task correlations for each task, respectively, and then catenates the results from all branches to capture the inter-task correlations among the multiple tasks. In addition, we present a two-stage online framework for reinforcement learning in practical use, including training and running phases. The extensive experiments have been conducted on two real-world urban sensing datasets, each with two types of sensing data, which verify the effectiveness of our proposed algorithms on multi-dimensional urban sensing and achieve better performances than the state-of-the-art mechanisms.
BibTeX
@inproceedings{liu2019multi,
  author = {Wenbin Liu and Yongjian Yang and En Wang and Leye Wang and Djamal Zeghlache and Daqing Zhang},
  title = {{Multi-Dimensional Urban Sensing in Sparse Mobile Crowdsensing}},
  booktitle = {IEEE Access},
  year = {2019},
  url = {https://openreview.net/forum?id=Oku4Ax9ZJO},
}
Figure