[Paper Review] Pilot-Based Unsourced Random Access with a Massive MIMO Receiver, MRC and Polar Codes.

Alexander Fengler, Peter Jung+1 more|arXiv (Cornell University)|Dec 6, 2020

Advanced MIMO Systems Optimization31 references4 citations

TL;DR

This paper proposes a pilot-based unsourced random access scheme for massive MIMO uplink systems over Rayleigh block-fading channels, using non-orthogonal pilots, MRC combining, and polar codes with successive cancellation-list decoding. It demonstrates that with sufficiently long coherence blocks, the approach achieves scalable, energy-efficient performance with closed-form performance approximation and strong finite-blocklength results via low-complexity detection.

ABSTRACT

In this work we treat the unsourced random access problem on a Rayleigh block-fading AWGN channel with multiple receive antennas. Specifically, we consider the slowly fading scenario where the coherence block-length is large compared to the number of active users and the message can be transmitted in one coherence block. Unsourced random access refers to a form of grant-free random access where users are considered to be a-priori indistinguishable and the receiver recovers a list of transmitted messages up to permutation. In this work we show that, when the coherence block length is large enough, a conventional approach based on the transmission of non-orthogonal pilot sequences with subsequent channel estimation and Maximum-Ratio-Combining (MRC) provides a simple energy-efficient solution whose performance can be well approximated in closed form. For the finite block-length simulations we use a randomly sub-sampled DFT matrix as pilot matrix, a low-complexity approximate message passing algorithm for activity detection and a state-of-the-art polar code as single-user error correction code with a successive-cancellation-list decoder. These simulations prove the scalability of the presented approach and the quality of the analysis.

Motivation & Objective

• Address the unsourced random access problem in massive MIMO systems where users are indistinguishable and the receiver must recover a list of messages up to permutation.
• Design a low-complexity, energy-efficient solution for grant-free uplink transmission in slowly fading Rayleigh block-fading channels.
• Enable scalable performance in finite blocklength regimes using practical coding and detection techniques.
• Provide a closed-form performance approximation for the proposed scheme under long coherence blocks.
• Demonstrate the feasibility and robustness of the approach using low-complexity activity detection and state-of-the-art polar codes.

Proposed method

• Employ non-orthogonal pilot sequences for user identification and channel estimation at the massive MIMO base station.
• Apply Maximum-Ratio Combining (MRC) at the receiver to coherently combine signals from multiple antennas, improving signal-to-noise ratio.
• Use a randomly sub-sampled DFT matrix as the pilot matrix to enable efficient and structured pilot design.
• Implement a low-complexity approximate message passing (AMP) algorithm for joint user activity detection and channel estimation.
• Integrate a state-of-the-art polar code with successive-cancellation-list decoding for reliable single-user error correction.
• Leverage the long coherence block length to enable accurate performance approximation in closed form.

Experimental results

Research questions

• RQ1 Can a pilot-based, non-orthogonal transmission scheme with MRC combining achieve reliable unsourced random access in massive MIMO systems over Rayleigh block-fading channels?
• RQ2 How does the performance of the proposed scheme scale with increasing numbers of active users under finite blocklength?
• RQ3 To what extent can the system performance be approximated in closed form when the coherence block length is large?
• RQ4 What is the impact of using a randomly sub-sampled DFT pilot matrix and low-complexity AMP detection on system performance?
• RQ5 How effective is the integration of polar codes with successive-cancellation-list decoding in achieving reliable message recovery in this unsourced access framework?

Key findings

• The proposed scheme achieves scalable performance in finite blocklength regimes, demonstrating robustness to increasing numbers of active users.
• With sufficiently long coherence blocks, the system performance can be well-approximated in closed form, enabling analytical performance evaluation.
• The use of a randomly sub-sampled DFT matrix as the pilot matrix enables efficient pilot design with favorable correlation properties.
• The low-complexity approximate message passing algorithm enables effective joint user activity detection and channel estimation.
• The integration of polar codes with successive-cancellation-list decoding ensures high reliability and near-optimal error correction performance.
• The overall system achieves energy efficiency and scalability, making it suitable for massive grant-free uplink access.