Abstract
This paper investigates adaptive signal shaping methods for millimeter wave (mmWave) multiple-input multiple-output (MIMO) communications based on the maximizing the minimum Euclidean distance (MMED) criterion. In this work, we utilize the indices of analog precoders to carry information and optimize the symbol vector sets used for each analog precoder activation state. Specifically, we firstly propose a joint optimization based signal shaping (JOSS) approach, in which the symbol vector sets used for all analog precoder activation states are jointly optimized by solving a series of quadratically constrained quadratic programming (QCQP) problems. JOSS exhibits good performance, however, with a high computational complexity. To reduce the computational complexity, we then propose a full precoding based signal shaping (FPSS) method and a diagonal precoding based signal shaping (DPSS) method, where the full or diagonal digital precoders for all analog precoder activation states are optimized by solving two small-scale QCQP problems. Simulation results show that the proposed signal shaping methods can provide considerable performance gain in reliability in comparison with existing mmWave transmission solutions.
Introduction
Millimeter-wave (mmWave) MIMO is a cornerstone technology for high-speed wireless communications. However, implementing fully digital MIMO systems at these frequencies is challenging due to the high cost and power consumption of radio frequency (RF) chains. Hybrid precoding architectures, which combine a limited number of RF chains with a large antenna array via analog phase shifters, offer a practical compromise.
Beamspace Modulation (BM) further enhances these systems by using the index of the active beam to carry additional information. While traditional BM often assumes uniform beam usage, real-world mmWave channels are sparse and non-uniform. This paper addresses the optimization of non-uniform beam-symbol combinations through "signal shaping" to maximize communication reliability.
Transmitter Architectures
The proposed signal shaping methods are applicable to various hybrid MIMO architectures. The two primary configurations are the Fully Connected Hybrid (FCH) and the Partially Connected Hybrid (PCH) structures. In FCH, each RF chain is connected to all transmit antennas, whereas in PCH, each RF chain drives a specific sub-array.
Proposed Signal Shaping Methods
The core objective is to maximize the Minimum Euclidean Distance (MMED) between received signal points. Three distinct strategies are proposed:
- Joint Optimization Based Signal Shaping (JOSS): This method jointly optimizes the symbol vector sets for all analog precoder states. It employs a recursive greedy design to build the signal set, solving a sequence of Quadratically Constrained Quadratic Programming (QCQP) problems. While achieving high performance, its complexity is significant.
- Full Precoding Based Signal Shaping (FPSS): To mitigate complexity, FPSS starts with a standard constellation and optimizes a full digital precoder for each beam activation state. This "refinement" approach significantly reduces the search space.
- Diagonal Precoding Based Signal Shaping (DPSS): This is the most computationally efficient variant, utilizing only diagonal digital precoders to adjust the phase and amplitude of individual streams.
Key Insights and Results
- Exploiting Non-Uniformity: By allowing different beams to carry varying amounts of information based on channel quality, the system achieves higher reliability than uniform modulation schemes.
- Shaping Philosophy: The entire transmit vector is treated as a point in high-dimensional space, optimized to resist noise effectively.
- Performance Gains: Simulation results using the Saleh-Valenzuela channel model demonstrate that JOSS and FPSS approach the performance of Fully Digital Signal Shaping (FDSS). In a 64-antenna massive MIMO setup, these methods significantly outperform Best Beamspace Selection (BBSS) and Uniform Beamspace Modulation (UBMSS).
- Complexity Trade-offs: FPSS offers a substantial reduction in computational complexity compared to JOSS (reducing the dependency on the number of signal vectors $N$ from $O(N^5)$ to $O(N^2)$), making it more suitable for real-time implementation.
Limitations and Future Work
The effectiveness of these methods relies on accurate Channel State Information (CSI) at the transmitter. In highly dynamic environments, the overhead of updating signal shapes could be a limiting factor. Additionally, the switching speed of analog phase shifters is a critical hardware requirement for beamspace modulation. Future research may focus on lower-complexity codebook-based shaping to further enhance scalability for ultra-low-latency applications.
Summary updated