TL;DR
This paper introduces the Fluctuating Two-Ray (FTR) fading model, a new statistical channel model that better captures the complex amplitude fluctuations observed in modern wireless environments, particularly at millimeter-wave frequencies. The FTR model offers a more accurate representation than existing models like Rician and TWDP, providing closed-form expressions for key statistical functions and demonstrating superior fit to real-world measurement data.
What Problem Does This Paper Address?
Modern wireless communication systems, especially those utilizing millimeter-wave (mmWave) bands for 5G and beyond, face significant challenges in accurately modeling the radio channel. Existing models, such as Rayleigh and Rician, often fail to capture the nuanced amplitude variations experienced by signals, particularly in urban outdoor environments. Recent measurements at 28 GHz have shown that even established models like Rician don't fully account for the observed signal behavior, which exhibits a bimodality not present in simpler models. This gap hinders the accurate prediction of system performance and the design of robust communication systems. The paper aims to bridge this gap by proposing a more flexible and accurate channel model.
Methods
The core contribution is the introduction of the Fluctuating Two-Ray (FTR) fading model. This model is a generalization of the Two-Wave with Diffuse Power (TWDP) model. The key innovation is allowing the two specular components, which represent dominant signal paths, to have fluctuating amplitudes in addition to random phases. This is achieved by modulating the amplitudes of the specular components with a Gamma distributed random variable.
The paper then proceeds to:
- Statistically characterize the FTR model: Deriving closed-form expressions for the Moment Generating Function (MGF), Probability Density Function (PDF), and Cumulative Distribution Function (CDF) of the received signal envelope.
- Provide approximate expressions: For cases where the parameter 'm' (related to the fluctuation) is an integer, approximate closed-form expressions using finite sums of elementary functions are derived. This simplifies evaluation and analysis.
- Establish connections to existing models: The FTR model is shown to encompass several well-known fading models (Rician, Rayleigh, Nakagami-m, Hoyt, TWDP) as special cases by adjusting its parameters (K, Δ, m).
- Analyze system performance: The paper derives exact and asymptotic expressions for key performance metrics like Bit Error Rate (BER) and Outage Capacity for various modulation schemes under FTR fading.
Datasets
The paper uses empirical results from a 28 GHz outdoor measurement campaign presented in reference [7] to validate the FTR fading model. Specifically, they compare the model's theoretical CDFs against measured CDFs for both Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS) scenarios. No new datasets are introduced.
Key Insights and Ideas
- Amplitude fluctuation is crucial: The paper highlights that simply having random phases for specular components (as in TWDP) is insufficient. Allowing their amplitudes to fluctuate is essential for accurately modeling real-world mmWave channels.
- FTR model's versatility: By unifying several existing fading models, the FTR model provides a single framework to describe a wide range of propagation conditions, from better-than-Rayleigh to worse-than-Rayleigh scenarios.
- Bimodality explained: The FTR model naturally exhibits bimodality, which is observed in empirical data, unlike simpler models. This bimodality is influenced by the model's parameters (K and Δ).
- Mathematical tractability maintained: Despite increased flexibility, the FTR model retains analytical tractability, with closed-form expressions for its statistics and performance metrics. This is a significant advantage over some other complex models.
- Performance implications: The study reveals that dissimilar specular components (low Δ) and lighter fluctuations (high m) lead to better performance (lower BER, lower outage probability).
Key Results
- The FTR model provides a significantly better fit to 28 GHz mmWave fading measurements compared to the Rician model, with error factors (e) of 0.2246 (LOS) and 0.2681 (NLOS) for FTR, versus 0.3302 (LOS) and 0.3571 (NLOS) for Rician.
- Closed-form expressions for the MGF, PDF, and CDF of the FTR fading model are derived.
- Approximate expressions for the PDF and CDF are provided, simplifying calculations.
- Exact and asymptotic expressions for BER and outage capacity are derived, enabling performance analysis.
- Numerical results (Figures 2-11) illustrate the impact of FTR parameters (K, Δ, m) on signal statistics and system performance, showing the model's flexibility.
Limitations and Future Work
- Correlation between specular components: The paper acknowledges that the FTR model assumes the two specular components are fully correlated. A more general model considering partial correlation between these components is suggested for future work, though it is noted that this would significantly increase analytical complexity.
- Approximation accuracy: While approximate expressions are provided for computational ease, their accuracy is implicitly tied to the parameter choices and the nature of the approximation.
- Specific modulation schemes: While BER expressions are derived for a family of modulations, the numerical results focus on BPSK. Further analysis for other modulation types could be beneficial.
Summary updated