Abstract
We study a new linear physical-layer network coding (LPNC) scheme for fading two-way relay channels. In the uplink phase, two users transmit simultaneously. The relay selects some integer coefficients and computes a linear combination (in a size-q finite set) of the two users' messages, which is broadcast in the downlink phase. We develop a design criterion for choosing the integer coefficients that minimizes the error probability. Based on that, we derive an asymptotically tight bound, in a closed-form, for the error probability of the LPNC scheme over Rayleigh fading channels. Our analysis shows that the error-rate performance of the LPNC scheme becomes asymptotically optimal at a high SNR, and our designed LPNC scheme significantly outperforms existing schemes in the literature.
Original language | English |
---|---|
Article number | 6196153 |
Pages (from-to) | 1068-1071 |
Number of pages | 4 |
Journal | IEEE Communications Letters |
Volume | 16 |
Issue number | 7 |
DOIs | |
Publication status | Published - 2012 |
Externally published | Yes |
Keywords
- code design
- compute-and-forward
- physical-layer network coding
- relay channel