[Paper Review] Vector Bin-and-Cancel for MIMO Distributed Full-Duplex

Jingwen Bai, Chris Dick+1 more|arXiv (Cornell University)|Feb 4, 2014

Full-Duplex Wireless Communications3 references3 citations

TL;DR

This paper proposes a vector bin-and-cancel strategy in MIMO distributed full-duplex networks to mitigate inter-mobile interference using an out-of-band side-channel between mobiles. By leveraging Han-Kobayashi message splitting, the scheme achieves the capacity region to within a constant bit, significantly improving generalized degrees-of-freedom (GDoF) and diversity-multiplexing tradeoff (DMT), especially when side-channel bandwidth or receive antennas are increased, reducing reliance on channel state information at the transmitter (CSIT).

ABSTRACT

In a multi-input multi-output (MIMO) full-duplex network, where an in-band full-duplex infrastruc- ture node communicates with two half-duplex mobiles supporting simultaneous up- and downlink flows, the inter-mobile interference between the up- and downlink mobiles limits the system performance. We study the impact of leveraging an out-of-band side-channel between mobiles in such network under different channel models. For time-invariant channels, we aim to characterize the generalized degrees- of-freedom (GDoF) of the side-channel assisted MIMO full-duplex network. For slow-fading channels, we focus on the diversity-multiplexing tradeoff (DMT) of the system with various assumptions as to the availability of channel state information at the transmitter (CSIT). The key to the optimal performance is a vector bin-and-cancel strategy leveraging Han-Kobayashi message splitting, which is shown to achieve the system capacity region to within a constant bit. We quantify how the side-channel improve the GDoF and DMT compared to a system without the extra orthogonal spectrum. The insights gained from our analysis reveal: i) the tradeoff between spatial resources from multiple antennas at different nodes and spectral resources of the side-channel, and ii) the interplay between the channel uncertainty at the transmitter and use of the side-channel.

Motivation & Objective

• To analyze the generalized degrees-of-freedom (GDoF) in a MIMO full-duplex network enhanced by an out-of-band side-channel between mobiles.
• To characterize the diversity-multiplexing tradeoff (DMT) under slow-fading channels with varying channel state information at the transmitter (CSIT) availability.
• To investigate the tradeoff between spatial resources (antennas) and spectral resources (side-channel bandwidth) in interference mitigation.
• To determine the conditions under which side-channel assistance eliminates the need for CSIT, simplifying transceiver design.
• To quantify the bandwidth and SNR requirements of the side-channel to achieve optimal DMT performance without CSIT.

Proposed method

• Employs a vector bin-and-cancel strategy based on Han-Kobayashi message splitting to manage interference in MIMO full-duplex systems.
• Derives the GDoF region under time-invariant channels by optimizing power and rate allocation across main and side-channels.
• Uses a two-phase optimization framework: first minimizing degrees-of-freedom loss due to interference, then allocating resources based on CSIT availability.
• Introduces a dual-optimization problem with constraints on sum rate and power allocation, using ordered variables to minimize the objective function.
• Analyzes the DMT performance by modeling the side-channel as a fading link with SNR scaling, and derives the required side-channel bandwidth to compensate for lack of CSIT.
• Derives closed-form expressions for DMT under different CSIT assumptions, showing inverse proportionality between required side-channel bandwidth and BS antenna count.

Experimental results

Research questions

• RQ1 How does the side-channel bandwidth affect the generalized degrees-of-freedom (GDoF) in a MIMO full-duplex network with inter-mobile interference?
• RQ2 Under what conditions does the side-channel eliminate the need for channel state information at the transmitter (CSIT)?
• RQ3 What is the required side-channel bandwidth to achieve the optimal diversity-multiplexing tradeoff (DMT) when CSIT is unavailable?
• RQ4 How do the number of antennas at the base station and mobiles trade off against side-channel spectral resources in performance gain?
• RQ5 What is the impact of channel uncertainty at the transmitter on the performance gain from the side-channel?

Key findings

• The vector bin-and-cancel scheme achieves the capacity region to within a constant bit, demonstrating near-optimal performance.
• When $ N_{\mathrm{dl}} \geq M_{\mathrm{ul}} $, CSIT provides no benefit, as the GDoF is already maximized without it.
• When $ M_{\mathrm{ul}} > N_{\mathrm{dl}} $, CSIT reduces the required side-channel bandwidth to achieve no-interference performance.
• For slow-fading channels, the required side-channel bandwidth to compensate for lack of CSIT is inversely proportional to the number of base station antennas, i.e., $ W \propto \frac{1}{M} $.
• The side-channel SNR must scale with the number of base station antennas to maintain optimal DMT when CSIT is absent.
• The DMT performance is improved by increasing either the number of downlink receive antennas or the side-channel bandwidth, reducing the need for CSIT.