SISO APP Searches in Lattices with Tanner Graphs

Dumitru Mihai Ionescu,Haidong Zhu

DOI: https://doi.org/10.1109/TIT.2011.2178130

2011-09-23

Abstract:An efficient, low-complexity, soft-output detector for general lattices is presented, based on their Tanner graph (TG) representations. Closest-point searches in lattices can be performed as non-binary belief propagation on associated TGs; soft-information output is naturally generated in the process; the algorithm requires no backtrack (cf. classic sphere decoding), and extracts extrinsic information. A lattice's coding gain enables equivalence relations between lattice points, which can be thereby partitioned in cosets. Total and extrinsic a posteriori probabilities at the detector's output further enable the use of soft detection information in iterative schemes. The algorithm is illustrated via two scenarios that transmit a 32-point, uncoded super-orthogonal (SO) constellation for multiple-input multiple-output (MIMO) channels, carved from an 8-dimensional non-orthogonal lattice (a direct sum of two 4-dimensional checkerboard lattice): it achieves maximum likelihood performance in quasistatic fading; and, performs close to interference-free transmission, and identically to list sphere decoding, in independent fading with coordinate interleaving and iterative equalization and detection. Latter scenario outperforms former despite the absence of forward error correction coding---because the inherent lattice coding gain allows for the refining of extrinsic information. The lattice constellation is the same as the one employed in the SO space-time trellis codes first introduced for 2-by-2 MIMO by Ionescu et al., then independently by Jafarkhani and Seshadri. Complexity is log-linear in lattice dimensionality, vs. cubic in sphere decoders.

Information Theory,Data Structures and Algorithms

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is: **In multiple - input multiple - output (MIMO) communication systems, how to perform soft - output detection of lattice points efficiently and with low complexity**. Specifically, the author proposes an efficient, low - complexity soft - output detector based on Tanner graph representation for the nearest - point search problem of general lattice points.

### Specific description of the problem

1. **Background and challenges**:
- In MIMO communication systems, space - time coding utilizes spatial diversity and temporal diversity to achieve reliable transmission.
- The lattice structure plays an important role in constructing space - time lattice codes, especially when dealing with high - dimensional signals.
- Although the traditional sphere - decoding algorithm can achieve maximum - likelihood performance, its complexity is relatively high, especially in high - dimensional lattices, where the amount of calculation increases exponentially.

2. **Research objectives**:
- Propose a new soft - output detection algorithm based on Tanner graph representation to reduce complexity and improve detection efficiency.
- Perform the nearest - point search on the Tanner graph through non - binary belief propagation (BP), thereby naturally generating soft - information output.
- Avoid the backtracking problem in traditional sphere - decoding and extract extrinsic information.

3. **Key points of the solution**:
- Utilize the coding gain of lattice points to establish an equivalence relationship between lattice points, thereby dividing the lattice points into cosets.
- Use the total a posteriori probability (APP) and the extrinsic a posteriori probability (APP) to support the iterative detection scheme.
- Verify the effectiveness of the algorithm through two scenarios: one is the 32 - point super - orthogonal (SO) constellation in a quasi - static fading channel; the other is coordinate interleaving and iterative equalization and detection in an independent fading channel.

4. **Innovations**:
- The complexity of the proposed algorithm is log - linear, which is significantly reduced compared to the cubic complexity of sphere - decoding.
- The algorithm can optimize the extraction of extrinsic information through the inherent lattice - point coding gain without using forward error - correction coding (FEC).
- It is applicable to various MIMO communication scenarios, especially the detection problem of high - dimensional lattice points.

### Summary

The core of this paper lies in solving the problems of high efficiency and low complexity in lattice - point detection in MIMO communication systems. By introducing a soft - output detection algorithm based on Tanner graph, the goal of significantly reducing the computational complexity while maintaining high performance is achieved.