Abstract— Iterative detection and decoding (IDD) in multiple-input multiple-output (MIMO) wireless systems is known to achieve near channel capacity. The high computational complexity of IDD, however, poses significant challenges for practical implementations (in terms of circuit area, latency, throughput, and power consumption). While the implementation of the involved detector and decoder circuits have received attention in the literature, only little is known about the efficient combination of both blocks in an IDD architecture. In this paper, we propose a novel iterative receiver schedule, which simultaneously performs detection and decoding on the same code block. This novel IDD approach is referred to as layered detection and decoding (LDD) and achieves lower latency and better performance compared to conventional solutions. Moreover, LDD is able to automatically match the decoding effort to the wide range of different modulation schemes and code rates specified in modern MIMO wireless standards. To demonstrate the efficiency of LDD, we present an extensive case study based on the characteristics of existing reference designs for a soft-input soft-output MMSE detector and an LDPC decoder.
Introduction
In high-speed wireless communication, Multiple-Input Multiple-Output (MIMO) technology is essential for achieving high spectral efficiency. To approach theoretical channel capacity, modern receivers often employ Iterative Detection and Decoding (IDD). While mathematically effective, IDD presents significant hardware implementation challenges, particularly regarding latency and circuit area. This paper introduces Layered Detection and Decoding (LDD), a novel architecture that breaks the traditional sequential bottleneck of iterative receivers.
The Challenge of Iterative Detection and Decoding
In a standard iterative receiver, the MIMO detector (which separates overlapping antenna signals) and the channel decoder (which corrects bit errors) operate in a sequential loop. The detector produces soft information in the form of Log-Likelihood Ratios (LLRs), which are passed to the decoder. The decoder refines this information and feeds it back to the detector for the next iteration.
This "ping-pong" exchange creates two primary hardware issues:
- Latency: The sequential nature means one unit is often idle while the other processes data, leading to high processing delays that can violate standard-specific timing requirements (e.g., Wi-Fi acknowledgments).
- Hardware Mismatch: Modern wireless standards use various Modulation and Coding Schemes (MCS). A hardware design optimized for one scheme (like QPSK) may become inefficient for others (like 64-QAM) because the relative runtimes of the detector and decoder shift.
Proposed Solution: Layered Detection and Decoding (LDD)
The core innovation of LDD is a schedule that allows the MIMO detector and the channel decoder to access a shared LLR memory concurrently. Instead of waiting for a full iteration to complete, both units work on the same block of data simultaneously.
To manage concurrent access without race conditions, the authors implement an incremental LLR-value feedback strategy. Rather than overwriting absolute LLR values, each unit calculates and writes back only the change ($\Delta$) in reliability based on its latest computation. This approach, inspired by layered LDPC decoding, ensures that all refined information is integrated fluidly as soon as it becomes available.
Performance Evaluation and Results
The authors evaluated LDD using a case study based on 90nm CMOS reference designs for a soft-input soft-output MMSE-PIC detector and a Quasi-Cyclic LDPC decoder. Simulations were performed using the TGn type-C channel model for 4x4 MIMO configurations.
Key findings include:
- Latency Reduction: For 16-QAM, LDD reduced processing latency to less than 70% of conventional sequential schedules.
- Area-Delay Efficiency: LDD demonstrated superior complexity metrics ($mm^2/Mbps$), providing a better trade-off between silicon area and throughput than traditional architectures.
- Adaptive Throughput: LDD naturally adapts to different MCS. If a complex modulation slows down the detector, the decoder continues processing available LLRs in the background without needing strict hardware synchronization.
- SNR Gains: LDD often achieved target Packet Error Rates (PER) at lower Signal-to-Noise Ratios (SNR) due to the more frequent exchange of reliability information.
Conclusion
Layered Detection and Decoding (LDD) offers a significant advancement for MIMO receiver design. By enabling concurrent operation and incremental updates, LDD effectively mitigates the latency and efficiency bottlenecks of traditional iterative receivers. This architecture is particularly well-suited for modern wireless standards that demand high throughput and low latency across a wide range of operating conditions.
Summary updated