Joint Beamforming and Compressed Sensing for Uplink Grant-Free Access

Guoqing Xia, Pei Xiao, Bohan Li, Yue Zhang, Huiyu Zhou

TL;DR

This paper tackles the challenge of efficient communication in massive IoT (mMTC) by proposing a novel framework that combines Spatial Division Multiple Access (SDMA) with Compressed Sensing (CS)-based grant-free Non-Orthogonal Multiple Access (NOMA). Their proposed joint adaptive beamforming and subspace pursuit (J-ABF-SP) algorithm, along with an interference cancellation enhancement (J-ABF-SP-IC), significantly improves user detection and data recovery performance, especially in overloaded scenarios.

What Problem Does This Paper Address?

The explosion of Internet of Things (IoT) devices, characteristic of mMTC, presents a significant challenge for wireless networks. These devices often transmit data sporadically and in small packets, making traditional grant-based access methods inefficient due to high latency and signaling overhead. Grant-free NOMA offers a solution by allowing devices to transmit without explicit permission, but it introduces the problem of blind user activity detection and data recovery at the base station (BS), especially when many users are active simultaneously. Furthermore, existing CS-based grant-free NOMA techniques often struggle with overloading and require prior knowledge of noise levels. This paper addresses these limitations by proposing a framework that leverages multi-antenna BS capabilities (SDMA) to manage interference and improve the efficiency of CS-based grant-free NOMA for massive connectivity.

Methods

The core of this work lies in a novel framework that synergistically combines SDMA with CS-based grant-free NOMA. Key methodological contributions include:

  • SDMA Integration: The paper proposes integrating SDMA with grant-free NOMA to manage inter-cluster interference. This is achieved through two beamforming schemes:
  • Statistical Beamforming (SBF): A closed-form solution designed to suppress inter-cluster interference by optimizing beamforming weights based on statistical channel information.
  • Dynamic Beamforming (DBF): A scheme that adapts beamforming weights based on the least-squares (LS) criterion, offering a more dynamic approach to interference suppression.
  • Joint Adaptive Beamforming and Subspace Pursuit (J-ABF-SP): This algorithm jointly optimizes beamforming weights and signal estimation. It iteratively updates beamforming and uses a subspace pursuit (SP) algorithm for compressed sensing-based multi-user detection (MUD) and data recovery (DR). A key innovation is a novel method for determining user sparsity level without requiring prior knowledge of the noise level.
  • Interference Cancellation (J-ABF-SP-IC): To further enhance performance, an interference cancellation scheme is introduced. This method uses initial signal estimates from J-ABF-SP to reconstruct and cancel interference, leading to improved signal recovery.

Datasets

Not discussed in the paper. The paper relies on simulated data generated based on the described signal model.

Key Insights and Ideas

  • SDMA as an Interference Manager: The paper demonstrates that SDMA is not just for multiplexing but can be effectively used to suppress inter-cluster interference in grant-free NOMA, a crucial step for handling massive numbers of users.
  • Joint Optimization is Key: The J-ABF-SP algorithm highlights the benefit of jointly optimizing beamforming and signal estimation, rather than treating them as separate problems. This iterative approach leads to better overall performance.
  • Sparsity Level Decision without Noise Knowledge: The development of a sparsity level decision method that doesn't require accurate noise level information is a significant practical advantage, as noise levels can be difficult to estimate precisely in real-world scenarios.
  • Interference Cancellation Boost: The J-ABF-SP-IC scheme shows that even after sophisticated beamforming, further gains can be achieved by explicitly canceling residual interference, leading to substantial improvements in signal recovery.
  • Robustness to Overloading: The proposed methods show strong performance even in highly overloaded scenarios (e.g., NQ/K = 600%), which is critical for mMTC applications.

Key Results

  • Superior Performance: Simulations show that the proposed J-ABF-SP and J-ABF-SP-IC algorithms significantly outperform existing CS-based grant-free NOMA techniques in terms of both Detection Error Rate (DER) and Symbol Error Rate (SER).
  • Outperforming Oracle: The J-ABF-SP algorithm achieves better DER than the "Oracle-BSASP" method, which assumes prior knowledge of the user sparsity level. This highlights the effectiveness of the proposed interference suppression techniques.
  • Significant SER Gains: The J-ABF-SP algorithm demonstrates a remarkable SER gain of over 8 dB compared to OAMP-MMV algorithms, and the J-ABF-SP-IC further improves this.
  • Efficiency with Fewer Antennas: The methods achieve satisfactory performance even with a small number of BS antennas, provided that the number of user clusters and their angular distribution match the antenna count.
  • Dynamic Beamforming Importance: The results emphasize the importance of dynamic beamforming updates, showing that the proposed J-ABF-SP is less sensitive to varying signal-to-noise ratios (SNRs) across clusters compared to static beamforming approaches.

Limitations and Future Work

  • Complexity: While the proposed methods offer performance gains, their computational complexity is noted to be higher than some benchmark algorithms, particularly those employing complexity reduction schemes. The paper mentions that future work will investigate computationally efficient CS methods.
  • Dynamic User Sparsity: The current work focuses on frame-wise block sparsity. The authors state that in the future, they will study the amalgamation of SDMA and CS for dynamic user sparsity-based grant-free NOMA.
  • Real-world Deployment: The paper relies on simulations. Future work could involve testing these algorithms in real-world testbeds to validate their performance under practical channel conditions and hardware constraints.

Summary updated

Abstract

Compressed sensing (CS)-based techniques have been widely applied in the grant-free non-orthogonal multiple access (NOMA) to a single-antenna base station (BS). In this paper, we consider the multi-antenna reception at the BS for uplink grant-free access for the massive machine type communication (mMTC) with limited channel resources. To enhance the overloading performance of the BS, we develop a general framework for the synergistic amalgamation of the spatial division multiple access (SDMA) technique with the CS-based grant-free NOMA. We derive a closed-form statistical beamforming and a dynamic beamforming scheme for the inter-cluster interference suppression when applying SDMA. Based on this, we further develop a joint adaptive beamforming and subspace pursuit (J-ABF-SP) algorithm for the multiuser detection and data recovery, with a novel sparsity level decision method without the accurate knowledge of the noise level. To further improve the data recovery performance, we propose an interference cancellation-based J-ABF-SP scheme (J-ABF-SP-IC) by using the initial signal estimates generated from the J-ABF-SP algorithm. Illustrative simulations verify the superior user detection and signal recovery performance of our proposed algorithms in comparison with existing CS-based grant-free NOMA techniques.
BibTeX
@article{xia2024joint,
  author = {Guoqing Xia and Pei Xiao and Bohan Li and Yue Zhang and Huiyu Zhou},
  title = {{Joint Beamforming and Compressed Sensing for Uplink Grant-Free Access}},
  journal = {IEEE Trans. Wireless Communications},
  year = {2024},
  doi = {10.36227/techrxiv.22723736.v2},
}
Figure