[Paper Review] Inband device-to-device relays in cellular networks

Bilal Sadiq, Saurabha Tavildar+1 more|arXiv (Cornell University)|Oct 8, 2015

Cooperative Communication and Network Coding10 references3 citations

TL;DR

This paper proposes a novel inband device-to-device (D2D) relay architecture in LTE networks that uses idle user equipment (UEs) as relays to improve downlink and uplink throughput by exploiting shadow-diversity and reusing uplink spectrum for access links. System simulations show a median throughput gain of 110% for downlink and 40% for uplink, enabled by power-controlled interference management and efficient relay association protocols.

ABSTRACT

A new design for two-hop opportunistic relaying in cellular networks is proposed, with the objective of throughput improvement. We propose using idle UEs in a cellular system with better channel to the base station to relay traffic for active UEs. One of the key ideas proposed is the use of (only) uplink spectrum for the Access links, and corresponding interference management schemes for managing interference between Access and Backhaul links. The proposed algorithms and architecture show a median throughput gain of 110% for downlink and 40% for uplink in system simulations performed according to the 3GPP methodology.

Motivation & Objective

• To improve system throughput in cellular networks by leveraging idle UEs as relays, especially in low- and median-throughput regimes.
• To address interference challenges arising from inband D2D access links and backhaul links in a shared spectrum environment.
• To enable efficient, low-overhead relay association and interference management using uplink spectrum for access links.
• To reduce power consumption at relay UEs by minimizing active modem time and enabling high-efficiency PA operation.
• To validate the proposed architecture using 3GPP-compliant system-level simulations.

Proposed method

• Uses idle UEs with better channel conditions to the base station as relays for active UEs, exploiting spatially uncorrelated shadowing over short distances.
• Reuses uplink spectrum for both uplink and downlink traffic on the D2D access link to simplify interference management and reduce spectral congestion.
• Employs power control at relay UEs to limit uplink-to-backhaul interference, with a single broadcast parameter signaling the maximum tolerable interference level to the base station.
• Relies on statistical multiplexing to manage uplink-to-access interference, avoiding explicit signaling for this link.
• Uses SRS-based measurement reports from UEs to the eNB to detect and coordinate interference between concurrent access links, enabling time/frequency orthogonality.
• Implements a relay discovery mechanism using two subframes to allow active UEs to discover and select optimal idle UEs based on DL SINR and path-loss constraints.

Experimental results

Research questions

• RQ1 Can idle UEs with better channel conditions to the base station significantly improve throughput for active UEs through two-hop relaying?
• RQ2 Is reusing uplink spectrum for D2D access links feasible and beneficial for interference management in a cellular network?
• RQ3 Can power control-based interference management effectively suppress uplink-to-backhaul interference while minimizing signaling overhead?
• RQ4 What is the impact of relay selection and association on system-level throughput and power consumption?
• RQ5 Can the proposed architecture achieve substantial median throughput gains without introducing significant complexity or power overhead?

Key findings

• The proposed architecture achieves a median downlink throughput gain of 110% and a median uplink throughput gain of 40% in 3GPP-compliant system-level simulations.
• Shadow-diverse relaying, leveraging uncorrelated shadowing over short distances, can yield over 10 dB improvement in median SINR and up to 200% increase in median throughput.
• Using uplink spectrum for access links enables simpler interference management, particularly for uplink-to-backhaul interference, which is effectively controlled via power control.
• The probability of any given UE being selected as a relay is low (~4%), minimizing the impact on user battery life.
• Power consumption at relay UEs is reduced due to low-duty-cycle modem operation and efficient power amplifier use on the access link, with additional gains from higher spectral efficiency.
• Interference between access links is effectively managed through SRS-based reporting and eNB coordination, enabling orthogonalization in time and/or frequency with minimal signaling.