Collaborative Spectrum Sensing from Sparse Observations in Cognitive Radio Networks

Jia Jasmine Meng, Wotao Yin, Husheng Li, Ekram Hossain, Zhu Han

Abstract

Spectrum sensing, which aims at detecting spectrum holes, is the precondition for the implementation of cognitive radio (CR). Collaborative spectrum sensing among the cognitive radio nodes is expected to improve the ability of checking complete spectrum usage. Due to hardware limitations, each cognitive radio node can only sense a relatively narrow band of radio spectrum. Consequently, the available channel sensing information is far from being sufficient for precisely recognizing the wide range of unoccupied channels. Aiming at breaking this bottleneck, we propose to apply matrix completion and joint sparsity recovery to reduce sensing and transmitting requirements and improve sensing results. Specifically, equipped with a frequency selective filter, each cognitive radio node senses linear combinations of multiple channel information and reports them to the fusion center, where occupied channels are then decoded from the reports by using novel matrix completion and joint sparsity recovery algorithms. As a result, the number of reports sent from the CRs to the fusion center is significantly reduced. We propose two decoding approaches, one based on matrix completion and the other based on joint sparsity recovery, both of which allow exact recovery from incomplete reports. The numerical results validate the effectiveness and robustness of our approaches. In particular, in small-scale networks, the matrix completion approach achieves exact channel detection with a number of samples no more than 50% of the number of channels in the network, while joint sparsity recovery achieves similar performance in large-scale networks.

Introduction

Spectrum is a scarce resource, yet much of it sits idle at any given time. Cognitive Radio (CR) technology aims to exploit these "spectrum holes" to improve spectral efficiency. However, CR nodes face two major hurdles: hardware constraints, where a single node cannot monitor the entire wideband spectrum simultaneously, and environmental fading, which can lead to the "hidden terminal" problem.

Traditional collaborative sensing requires every node to report on every channel, creating a massive communication overhead. This paper proposes a collaborative framework that uses Matrix Completion and Joint Sparsity Recovery to reconstruct a global view of spectrum occupancy from sparse, incomplete observations, significantly reducing the data that nodes need to sense and transmit.

Methods and System Architecture

The proposed framework equips CR nodes with frequency-selective filters that capture linear combinations of multiple channel powers. Instead of reporting raw power levels for every channel, nodes send these "compressed" reports to a central Fusion Center (FC).

System architecture for collaborative compressive spectrum sensing in cognitive radio networks.

The FC employs two primary decoding strategies:

  • Matrix Completion (MC): Treating sensing data as a low-rank matrix (since few primary users are typically active), the authors use the Fixed-Point Continuation Algorithm (FPCA) to fill in missing entries caused by intentional under-sampling.
  • Joint Sparsity Recovery (JSR): Leveraging the common sparsity pattern across nodes, the authors developed a custom Joint Detection Algorithm using truncated $\ell_1$ minimization. This iterative approach is specifically designed to handle the high dynamic range of signal strengths caused by wireless fading.

Key Results

The study utilized numerical simulations based on established wireless propagation models (AWGN, Rayleigh fading, and Log-normal shadowing). Small-scale tests involved 35 channels and 20 nodes, while large-scale tests scaled to 500 channels.

Probability of Detection (POD) vs Sampling Rate for different numbers of Primary Users (PU).

Key findings include:

  • Data Reduction: In small-scale networks, Matrix Completion achieved exact channel detection using less than 50% of the samples required by traditional methods.
  • High Sensitivity: As shown in the performance metrics, even with a sampling rate as low as 20%, the Probability of Detection (POD) remained above 95% for moderate occupancy levels.
  • Scalability: The Joint Sparsity approach proved superior for large-scale networks (500+ channels), maintaining high accuracy even in noisy environments with a Signal-to-Noise Ratio (SNR) as low as 15dB.

Conclusion

This research demonstrates that shifting the burden from "sensing more" to "computing smarter" can overcome hardware and communication bottlenecks in CR networks. While Matrix Completion is better suited for small networks with high data loss, Joint Sparsity Recovery excels in large-scale, low-complexity deployments. Future work will focus on adapting these algorithms for dynamic environments where primary users change state during the sensing interval.

Summary updated

Abstract

Spectrum sensing, which aims at detecting spectrum holes, is the precondition for the implementation of cognitive radio (CR). Collaborative spectrum sensing among the cognitive radio nodes is expected to improve the ability of checking complete spectrum usage. Due to hardware limitations, each cognitive radio node can only sense a relatively narrow band of radio spectrum. Consequently, the available channel sensing information is far from being sufficient for precisely recognizing the wide range of unoccupied channels. Aiming at breaking this bottleneck, we propose to apply matrix completion and joint sparsity recovery to reduce sensing and transmitting requirements and improve sensing results. Specifically, equipped with a frequency selective filter, each cognitive radio node senses linear combinations of multiple channel information and reports them to the fusion center, where occupied channels are then decoded from the reports by using novel matrix completion and joint sparsity recovery algorithms. As a result, the number of reports sent from the CRs to the fusion center is significantly reduced. We propose two decoding approaches, one based on matrix completion and the other based on joint sparsity recovery, both of which allow exact recovery from incomplete reports. The numerical results validate the effectiveness and robustness of our approaches. In particular, in small-scale networks, the matrix completion approach achieves exact channel detection with a number of samples no more than 50% of the number of channels in the network, while joint sparsity recovery achieves similar performance in large-scale networks.
BibTeX
@article{meng2011collaborative,
  author = {Jia Jasmine Meng and Wotao Yin and Husheng Li and Ekram Hossain and Zhu Han},
  title = {{Collaborative Spectrum Sensing from Sparse Observations in Cognitive Radio Networks}},
  journal = {IEEE J. Sel. Areas Commun.},
  year = {2011},
  url = {https://openreview.net/forum?id=g8zG0TACV3OM},
}
Figure