Spatially Common Sparsity Based Adaptive Channel Estimation and Feedback for FDD Massive MIMO (1604.03695v3)

Abstract: This paper proposes a spatially common sparsity based adaptive channel estimation and feedback scheme for frequency division duplex based massive multi-input multi-output (MIMO) systems, which adapts training overhead and pilot design to reliably estimate and feed back the downlink channel state information (CSI) with significantly reduced overhead. Specifically, a non-orthogonal downlink pilot design is first proposed, which is very different from standard orthogonal pilots. By exploiting the spatially common sparsity of massive MIMO channels, a compressive sensing (CS) based adaptive CSI acquisition scheme is proposed, where the consumed time slot overhead only adaptively depends on the sparsity level of the channels. Additionally, a distributed sparsity adaptive matching pursuit algorithm is proposed to jointly estimate the channels of multiple subcarriers. Furthermore, by exploiting the temporal channel correlation, a closed-loop channel tracking scheme is provided, which adaptively designs the non-orthogonal pilot according to the previous channel estimation to achieve an enhanced CSI acquisition. Finally, we generalize the results of the multiple-measurement-vectors case in CS and derive the Cramer-Rao lower bound of the proposed scheme, which enlightens us to design the non-orthogonal pilot signals for the improved performance. Simulation results demonstrate that the proposed scheme outperforms its counterparts, and it is capable of approaching the performance bound.

• The paper presents a compressive sensing-based approach that adaptively acquires CSI by exploiting spatial sparsity, reducing training overhead by up to 92%.
• It develops a closed-loop channel tracking and non-orthogonal pilot design to achieve near-oracle LS performance with efficient compressive recovery.
• The study integrates a DSAMP algorithm within a generalized MMV framework, using CRLB derivation to guide pilot design for enhanced FDD MIMO feedback.

Overview of Spatially Common Sparsity Based Adaptive Channel Estimation and Feedback for FDD Massive MIMO

This paper introduces a novel approach to channel estimation and feedback for Frequency Division Duplex (FDD) massive Multiple Input Multiple Output (MIMO) systems by exploiting spatially common sparsity. The proposed method addresses the challenge of high training overhead associated with acquiring accurate downlink Channel State Information (CSI), a critical requirement for effective beamforming and resource allocation in massive MIMO systems.

Key Contributions

The research presents several significant contributions to the field:

1. Compressive Sensing (CS) Based Adaptive CSI Acquisition:
   • The paper introduces a compressive sensing-based adaptive CSI acquisition scheme that adjusts the training time slot overhead according to the sparsity level of the channels. This method significantly reduces the required channel estimation overhead compared to conventional schemes, which typically scale with the number of base station (BS) antennas.
2. Closed-Loop Channel Tracking:
   • The research proposes a closed-loop channel tracking scheme that leverages the spatially common sparsity of massive MIMO channels across multiple time blocks. This method further reduces training overhead by designing the pilot signals based on previously acquired CSI.
3. Non-Orthogonal Downlink Pilot Design:
   • Distinct from standard orthogonal pilots, the proposed non-orthogonal pilot design is optimized for compressive sensing, providing efficient compression and reliable recovery of sparse signals. The design enhances performance by ensuring the measurement matrices satisfy the Restricted Isometry Property (RIP).
4. Distributed Sparsity Adaptive Matching Pursuit (DSAMP) Algorithm:
   • The tailored DSAMP algorithm is proposed for joint estimation of multiple sparse channel vectors. Compared to conventional algorithms like Orthogonal Matching Pursuit (OMP) and Subspace Pursuit (SP), DSAMP efficiently reduces the time slot overhead while maintaining similar computational complexity.

Implications and Results

The proposed method achieves substantial reductions in the overhead required for reliable channel estimation in FDD massive MIMO systems. Notably, the impact is illustrated by empirical results showing a potential overhead reduction of approximately 92% compared to traditional techniques. The simulations also validate that the CS based adaptive scheme closely approaches the oracle LS performance bound, affirming its robustness and efficacy.

Theoretical Underpinnings and Future Directions

The authors extend the conventional Multiple-Measurement Vector (MMV) problem to a Generalized MMV (GMMV) framework, demonstrating superior recovery performance for sparse signals. Moreover, they derive the Cramer-Rao Lower Bound (CRLB), guiding the non-orthogonal pilot design and validating the channel estimation accuracy. This work sets the stage for future exploration of adaptive pilot designs in massively distributed antenna systems and further optimization of feedback mechanisms in complex MIMO environments, potentially incorporating machine learning to dynamically predict and adjust to channel conditions.

In summary, this paper contributes a comprehensive and practical approach to channel estimation and feedback in massive MIMO systems, leveraging spatially common sparsity for improved efficiency and performance, with strong theoretical foundations and promising empirical results.