[Paper Review] Multi-User Transmissions for Relay Networks with Linear Complexity

Liangbin Li, Yindi Jing+1 more|arXiv (Cornell University)|Jul 20, 2010

Cooperative Communication and Network Coding3 citations

TL;DR

This paper proposes two low-complexity protocols—DSTC-ICRec and TDMA-ICRec—for multi-user transmission in a (J, R, N) multi-access relay network with linear processing at the relay and destination. By combining distributed space-time coding at the relay and zero-forcing interference cancellation at the receiver, the protocols enable concurrent transmission with full diversity gain under linear complexity constraints.

ABSTRACT

This paper considers transmission and detection schemes in a (J, R, N) multi-access relay network (MARN), where J single-antenna source nodes send independent messages to one N-antenna destination node through one R-antenna relay node. For complexity considerations, the network is under two linear constraints. The relay node linearly maps its received signals to generate the forwarded signals without decoding and the destination node has linear decoding complexity in the number of users. We propose two protocols to allow multi-user concurrent transmission in the network. Using distributed space-time coding (DSTC) at the relay and zero-forcing interference cancellation (IC) at the receiver, the protocol of DSTC-ICRec allows concurrent transmission in both the source-relay link and the relay-receiver link. When J = 2J-1, TDMA-ICRec achieves the maximum interference-free (int-free) diversity of R, at a lower symbol rate compared to DSTC-ICRec.

Motivation & Objective

• To enable efficient multi-user transmission in relay networks under strict linear complexity constraints.
• To address the challenge of interference in multi-source relay networks with limited processing at relay and destination.
• To design protocols that support concurrent transmission in both source-relay and relay-destination links without decoding at the relay.
• To achieve maximum diversity gain under linear detection complexity at the destination.
• To compare performance trade-offs between concurrent transmission (DSTC-ICRec) and TDMA-based transmission (TDMA-ICRec) in terms of diversity and spectral efficiency.

Proposed method

• The relay applies linear precoding to forward signals without decoding, maintaining linear complexity.
• Distributed space-time coding (DSTC) is employed at the relay to exploit spatial diversity across multiple sources.
• Zero-forcing interference cancellation (IC) is applied at the destination to suppress multi-user interference.
• The DSTC-ICRec protocol enables concurrent transmission over both hops using linear processing and interference cancellation.
• The TDMA-ICRec protocol schedules transmissions in time slots to achieve maximum diversity gain of R when J = 2J-1.
• Both protocols ensure linear decoding complexity at the destination in the number of users.

Experimental results

Research questions

• RQ1 How can multi-user transmission be efficiently supported in relay networks under linear processing constraints at the relay and destination?
• RQ2 What is the maximum achievable diversity gain under linear detection complexity and concurrent transmission?
• RQ3 How does the performance of concurrent transmission (DSTC-ICRec) compare to TDMA-based transmission (TDMA-ICRec) in terms of diversity and spectral efficiency?
• RQ4 Can distributed space-time coding at the relay effectively support interference suppression in multi-source relay networks?
• RQ5 What is the trade-off between diversity gain and symbol rate when using TDMA-ICRec versus DSTC-ICRec?

Key findings

• The DSTC-ICRec protocol enables concurrent transmission in both the source-relay and relay-destination links using linear processing and zero-forcing interference cancellation.
• The TDMA-ICRec protocol achieves the maximum possible interference-free diversity gain of R when J = 2J-1.
• Both protocols maintain linear decoding complexity at the destination, scaling with the number of users.
• DSTC-ICRec achieves full diversity gain with a higher spectral efficiency compared to TDMA-ICRec.
• TDMA-ICRec achieves optimal diversity but at the cost of lower spectral efficiency due to time-division multiplexing.
• The proposed protocols effectively balance diversity gain, spectral efficiency, and computational complexity in multi-user relay networks.