Throughput of Cellular Systems with Conferencing Mobiles and Cooperative Base Stations

Osvaldo Simeone, Oren Somekh, Gerhard Kramer, H. Vincent Poor, Shlomo ShamaiEURASIP J. Wirel. Commun. Netw 2008 · DOI · OpenReview

Introduction

In traditional cellular networks, interference from neighboring cells is a primary bottleneck for performance. While "Multicell processing" (Network MIMO) addresses this by jointly decoding signals at base stations (BSs) via a high-capacity backbone, it only tackles cooperation at the receiver side. This paper, published in the EURASIP Journal on Wireless Communications and Networking, explores a hybrid approach: enabling cooperation at both ends of the wireless link.

The authors propose "conferencing" for mobile stations (MSs), allowing nearby handsets to exchange data over short-range links (e.g., Bluetooth or Wi-Fi) before transmitting to the cellular network. This effectively transforms a group of independent phones into a distributed antenna array, significantly boosting system throughput.

System Model and Methods

The study employs the linear Wyner model, a standard information-theoretic abstraction where cells are arranged linearly and interference is restricted to immediate neighbors. The architecture consists of two layers: the standard uplink channel and a dedicated conferencing layer.

The proposed system model featuring both uplink transmission and inter-mobile conferencing channels.

Two primary scenarios are analyzed:

  1. Intercell Conferencing: Mobiles in adjacent cells exchange information across cell boundaries. This is modeled as a series of "conferencing rounds" where messages propagate across the network.
  2. Intracell Conferencing: Multiple mobiles within the same cell coordinate their transmissions.

The core technical strategy is rate-splitting. Each mobile divides its message into a "private" part, sent directly, and a "common" part, which is shared with neighbors during a conferencing phase. In the subsequent transmission phase, multiple mobiles cooperatively transmit these common messages. For the intercell case, this cooperation acts as a convolutional spatial pre-equalizer, allowing mobiles to "pre-distort" signals to better navigate inter-cell interference.

Key Insights

  • Spatial Filtering: Intercell conferencing is mathematically equivalent to applying a Finite Impulse Response (FIR) filter across the spatial domain. The number of conferencing rounds ($K$) determines the number of "taps" in this spatial filter.
  • Coherent Power Combining: Unlike simple models where TDMA is often optimal, the authors find that simultaneous transmission with intracell conferencing is superior. It enables signals from different phones to add up constructively at the base station, providing a power gain proportional to the number of cooperating users.
  • Optimality: The rate-splitting strategy is shown to be capacity-achieving, provided the conferencing links have sufficient capacity.
  • Low-SNR Advantages: Benefits are most pronounced in low-signal environments, where conferencing allows the system to approximate "waterfilling" power allocation across the entire network.

Key Results

Numerical simulations validate the theoretical findings, demonstrating that even a few rounds of conferencing can significantly close the gap between standard multicell processing and the theoretical upper bound of perfect mobile cooperation.

Throughput gains as a function of inter-cell gain α for varying numbers of conferencing rounds K.

As shown in the figure above, increasing the number of conferencing rounds ($K$) progressively improves the achievable rate ($R$) towards the upper bound ($R_{upper}$). However, there is an inherent trade-off: excessive conferencing rounds over limited-capacity links can introduce overhead that diminishes net gains.

Limitations and Future Work

While promising, the linear Wyner model is a 1D simplification. Real-world 2D hexagonal cell layouts present more complex message propagation challenges. Additionally, the strategy assumes mobiles have accurate channel state information; in fast-fading environments, the overhead of sharing this state might outweigh throughput benefits. Future research could explore scenarios with limited backhaul capacity at both the mobile and base station ends, as well as the application of network coding to improve conferencing efficiency in multi-dimensional topologies.

Summary updated

Abstract

This paper considers an enhancement to multicell processing for the uplink of a cellular system, whereby the mobile stations are allowed to exchange messages on orthogonal channels of fixed capacity (conferencing). Both conferencing among mobile stations in different cells and in the same cell (inter- and intracell conferencing, resp.) are studied. For both cases, it is shown that a rate-splitting transmission strategy, where part of the message is exchanged on the conferencing channels and then transmitted cooperatively to the base stations, is capacity achieving for sufficiently large conferencing capacity. In case of intercell conferencing, this strategy performs convolutional pre-equalization of the signal encoding the common messages in the spatial domain, where the number of taps of the finite-impulse response equalizer depends on the number of conferencing rounds. Analysis in the low signal-to-noise ratio regime and numerical results validate the advantages of conferencing as a complementary technology to multicell processing.
BibTeX
@article{simeone2008throughput,
  author = {Osvaldo Simeone and Oren Somekh and Gerhard Kramer and H. Vincent Poor and Shlomo Shamai},
  title = {{Throughput of Cellular Systems with Conferencing Mobiles and Cooperative Base Stations}},
  journal = {EURASIP J. Wirel. Commun. Netw.},
  year = {2008},
  doi = {10.1155/2008/652325},
}
Figure