Abstract
This paper proposes a hybrid D2D-cellular scheme that transforms idle "standby" smartphones into a distributed, virtual antenna array for active users. By leveraging high-speed millimeter-wave (mmWave) links for local coordination, the system creates a "Distributed MIMO" environment. This architecture significantly boosts data rates and mitigates inter-cell interference, providing substantial gains for disadvantaged cell-edge users.
Introduction
In dense 5G and 6G networks, the primary bottleneck is often inter-cell interference rather than simple signal strength. As base stations (BS) operate in close proximity, their signals overlap, frequently drowning out data for users at the cell edge. While Multiple-Input Multiple-Output (MIMO) technology can mitigate this through beamforming, mobile handsets are physically limited in the number of antennas they can house.
This research explores utilizing the idle Device-to-Device (D2D) capabilities of nearby "standby" users to assist active cellular users. By forming a cooperative network, these standby devices act as a virtual antenna array, overcoming the physical limitations of individual handsets and improving overall network efficiency.
Proposed Methodology
The authors introduce a two-phase cooperative transmission protocol designed to create a distributed MIMO environment:
- Phase I (Broadcast): The Base Station transmits data to an intended active user. Nearby "standby" users also overhear this signal.
- Phase II (Relaying): These standby users act as relay nodes, compressing their overheard observations and transmitting them to the active user via high-bandwidth mmWave D2D links (operating at 73 GHz).
By combining these multiple observations, the active user effectively functions as a multi-antenna receiver, gaining diversity and interference-nulling capabilities that a single device could not achieve alone.
Orthogonal Access Protocol
A critical challenge in this architecture is managing interference between D2D links and the primary cellular network. The authors propose an Orthogonal Access Protocol using a hybrid approach: Frequency Division Multiple Access (FDMA) is used between different user clusters, while Time Division Multiple Access (TDMA) is employed within each cluster.
Furthermore, the study provides a closed-form "water-filling" solution for optimal time allocation to each relay. This ensures that resources are prioritized for relays with high-quality links, making the system computationally efficient for real-time implementation.
Evaluation and Results
The proposed scheme was evaluated using extensive system-level simulations based on a standard 19-cell "wrapped-around" hexagonal model. Channel models incorporated 3GPP LTE-A standards for 2 GHz cellular links and specific Nakagami fading models for 73 GHz mmWave D2D links.
Key Findings
- Significant Rate Improvements: For cell-edge users (10th percentile), the scheme achieved a 2x to 2.5x improvement in data rates when assisted by 9 standby users.
- Diversity Gain: Unlike traditional relaying which focuses on power gain, this method leverages "diversity gain," allowing users to mathematically cancel out interference.
- Optimal BS Loading: The results indicate that as the number of virtual antennas increases, the Base Station can optimally schedule more users simultaneously ("full loading"), increasing overall network capacity.
Discussion and Future Work
While promising, the implementation of distributed MIMO faces practical hurdles. The overhead required for Channel State Information (CSI) acquisition in dynamic environments could impact performance. Additionally, real-world deployment would require an incentive system to encourage standby users to share their battery and processing resources.
Future research directions include integrating this D2D-assisted approach with Coordinated Multi-Point (CoMP) at the base station level to further enhance network performance and reliability.
Summary updated