Performance of Opportunistic Fixed Gain Bidirectional Relaying With Outdated CSI

Fahd Ahmed Khan, Kamel Tourki, Mohamed-Slim Alouini, Khalid A. QaraqearXiv 2013 · OpenReview

Abstract

This paper studies the impact of using outdated channel state information (CSI) for relay selection on the performance of a network where two sources communicate with each other via fixed-gain amplify-and-forward (AF) relays. For a Rayleigh faded channel, closed-form expressions for the outage probability, moment generating function (MGF), and symbol error rate (SER) are derived. Simulation results are presented to corroborate the derived analytical results. It is shown that adding relays does not improve the performance if the channel is substantially outdated. Furthermore, relay location is taken into consideration, demonstrating that performance can be improved by placing the relay closer to the source whose channel is more outdated.

Introduction

In cooperative wireless communications, relaying is a vital strategy for extending coverage and improving link reliability, particularly when the direct path between two sources ($S_1$ and $S_2$) is obstructed or suffers from severe fading. Bidirectional relaying, where a relay facilitates the exchange of information between two nodes simultaneously, offers significant spectral efficiency. To maximize these gains, systems often employ Relay Selection (RS), choosing the "best" available relay from a pool to handle the transmission.

However, a critical gap exists between theoretical models and practical deployment: the assumption of perfect Channel State Information (CSI). In dynamic environments, there is an inevitable delay between the estimation of the channel and the actual data transmission. This results in "outdated" CSI. While previous literature has explored this impact on variable-gain relays, this research focuses on fixed-gain (FG) relays. FG relays are preferred in low-cost, low-complexity applications because they apply a constant amplification factor regardless of instantaneous channel conditions. However, the mathematical analysis of FG relays under outdated CSI is significantly more complex than their variable-gain counterparts.

System Model and Methodology

The study considers a dual-hop AF bidirectional relay network consisting of two sources and $R$ potential relays operating under Rayleigh fading. The system utilizes a Max-Min relay selection criterion, which identifies the relay that maximizes the minimum signal-to-noise ratio (SNR) of the two source-relay links.

To model the temporal correlation between the channel at the time of selection and the time of transmission, the authors use a correlation coefficient ($\rho$). A value of $\rho = 1$ represents perfect CSI, while $\rho = 0$ indicates completely outdated information. The research provides a rigorous mathematical framework by deriving new, closed-form expressions for three key performance metrics:

  1. Outage Probability: The probability that the end-to-end SNR falls below a predefined threshold.
  2. Moment Generating Function (MGF): A fundamental statistical tool used to derive various performance indicators.
  3. Symbol Error Rate (SER): Calculated for various modulation schemes, such as BPSK and QAM, to quantify data reliability.

These derivations utilize advanced calculus and Meijer-G functions to ensure exact analytical solutions, which are then validated through extensive Monte-Carlo simulations.

Key Insights and Performance Analysis

The Diversity Trap and the "More is Less" Paradox

In ideal scenarios, increasing the number of relays ($R$) provides spatial diversity, which significantly lowers the outage probability. However, this research reveals a "Diversity Trap": when CSI is outdated, the diversity gain diminishes. If the channel is sufficiently stale, the system's performance plateaus, and adding more relays provides no benefit.

More surprisingly, the authors identify a "More is Less" paradox. If the CSI is highly outdated, increasing the number of relays can actually degrade performance. This occurs because the selection process is more likely to choose a relay that appeared optimal during the selection phase but has since entered a deep fade by the time transmission occurs.

Strategic Relay Placement

Traditionally, placing a relay at the midpoint between two sources ($d=0.5$) is considered optimal for balanced links. This paper demonstrates that when CSI is outdated, the optimal geometry changes. If one source-to-relay link is more outdated than the other (for example, if one source is moving faster), the system performance can be partially recovered by moving the relay closer to the source with the more outdated channel. This asymmetric compensation helps balance the increased error probability caused by the stale CSI.

Results and Validation

The analytical formulas derived in the paper were validated using simulations with a path-loss exponent of $\upsilon = 3$ and varying relay counts ($R=2, 3, 6$). The numerical results perfectly matched the theoretical curves, confirming the accuracy of the Meijer-G based expressions. The results clearly show that as the correlation coefficient $\rho$ decreases, the SER and Outage Probability increase sharply, highlighting the sensitivity of fixed-gain bidirectional networks to feedback delays.

Conclusion

This work provides a comprehensive analytical foundation for understanding the limitations of fixed-gain bidirectional relaying in realistic, time-varying environments. The findings suggest that for high-mobility or high-delay applications, simply adding more relays is an ineffective strategy. Instead, network designers must account for CSI staleness through robust mathematical modeling and strategic physical placement of relay nodes to maintain communication reliability.

Summary updated

Abstract

This paper studies the impact of using outdated channel state information for relay selection on the performance of a network where two sources communicate with each other via fixed-gain amplifyand- forward relays. For a Rayleigh faded channel, closed-form expressions for the outage probability, moment generating function and symbol error rate are derived. Simulations results are also presented to corroborate the derived analytical results. It is shown that adding relays does not improve the performance if the channel is substantially outdated. Furthermore, relay location is also taken into consideration and it is shown that the performance can be improved by placing the relay closer to the source whose channel is more outdated.
BibTeX
@misc{khan2013performance,
  author = {Fahd Ahmed Khan and Kamel Tourki and Mohamed-Slim Alouini and Khalid A. Qaraqe},
  title = {{Performance of Opportunistic Fixed Gain Bidirectional Relaying With Outdated CSI}},
  year = {2013},
  eprint = {1305.5913},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1305.5913},
}
Figure