1. S. Lyu, A. Campello and C. Ling, "Ring Compute-and-Forward over Block-Fading Channels,"  IEEE Trans. Inf. Theory, under review, 2018.

2. S. Lyu and C. Ling, "Hybrid Vector Perturbation Precoding: The Blessing of Approximate Message Passing," IEEE Trans. Signal Process., under revision, 2018.

3. S. Lyu and C. Ling, “Boosted KZ and LLL algorithms,” IEEE Trans. Signal Process., vol. 65, no. 18, pp. 4784–4796, Sep. 2017.   MATLAB CODES FOR THIS PAPER

4. S. Lyu, C. Porter and C. Ling, "On Algebraic Lattice Reduction and its Application to Compute-and-Forward,"  to be submitted, the conference version has been accepted for presentation in ITW-2018.

5. Z. Wang, Y. Huang, and S. Lyu, "Markov chain Monte Carlo Methods For Lattice Gaussian Sampling: Lattice Reduction and Decoding Optimization," IEEE Trans. Signal Process., submitted, 2018.

Abstracts

1. On compute-and-forward: rather than decoding an integer linear combination of lattice codes, we propose to decode an algebraic-integer linear combination of lattice codes. The algebraic integers can have different values in different blocks.

2. On approximate message passing (AMP): When the signal prior is integers $\mathbb{Z}$, which is the case in MIMO precoding/detection, we try to use AMP in the "second round". That is, first use a sub-optimal algorithm (with promised performance) to obtain an estimation, and then use AMP to estimate the residual error.

3. On lattice reduction: We replace the size reduction steps in LLL or HKZ/KZ with stronger length-reduction steps. If the length-reduction involves a closest vector problem (CVP) oracle, the so called boosted KZ algorithm yields the strongest bounds on the lengths of basis vectors. The link for its MATLAB codes is HERE.