Optimal Power Allocation of Relay Sensor Node Capable of Energy Harvesting in Cooperative Cognitive Radio Network

Pham Ngoc Son, Dongsoo Har, Nam Ik Cho, Hyung Yun Kong

TL;DR

This paper proposes a novel cooperative cognitive radio (CR) scheme where a relay sensor node harvests energy from primary signals to simultaneously transmit its own data and relay the primary signal using power superposition coding. The research analytically investigates the impact of energy harvesting and power allocation strategies on network performance, aiming to optimize power splitting and sharing for improved reliability.

What Problem Does This Paper Address?

Wireless sensor networks (WSNs) often face energy constraints, limiting their operational lifespan and capabilities. Cognitive radio (CR) technology aims to improve spectrum utilization by allowing secondary networks (SNs) to opportunistically use licensed frequency bands of primary networks (PNs). This paper tackles the challenge of enabling WSNs to operate within CR frameworks while overcoming their energy limitations. Specifically, it addresses how a relay sensor node (ST) can effectively harvest energy from the primary signal and then utilize this harvested energy to both relay the primary signal and transmit its own data, all within a cooperative communication (CC) setup. The core problem is to find the optimal way to allocate this harvested energy between decoding the primary signal and transmitting both signals simultaneously, to minimize performance degradation (outage probability) for both the primary and secondary networks.

Methods

The proposed scheme operates in two phases. In the first phase, the secondary transmitter (ST) harvests energy from the primary transmitter's (PT) signal and decodes it. In the second phase, the ST uses the harvested energy to transmit a combined signal to both the primary receiver (PR) and a secondary receiver (SR) using power superposition coding. This coding allows the ST to transmit its own data alongside the relayed primary signal.

The key novelties lie in:

  • Energy Harvesting and Power Superposition Coding Integration: This is the first work to combine these two techniques in a cooperative cognitive radio relay scenario.
  • Power Splitting Ratio (ρ): The ST divides the received RF signal into two parts: one for charging its battery and the other for decoding the primary signal. The parameter ρ controls this split.
  • Power Sharing Coefficient (α): Within the power superposition coding, the ST allocates a fraction of its transmit power to the primary signal (for relaying) and the remaining fraction to its own signal. The parameter α governs this allocation.
  • Analytical Performance Evaluation: The paper derives mathematical expressions for the outage probabilities of both the PN and the SN. These expressions are functions of the power splitting ratio (ρ), power sharing coefficient (α), channel gains, and other system parameters.
  • Joint Optimization: The research aims to find the jointly optimal values of ρ and α that minimize the outage probabilities for both networks, subject to practical constraints.

Datasets

Not discussed in the paper. The evaluation relies on numerical analysis and Monte-Carlo simulations using synthesized channel models and system parameters.

Key Insights and Ideas

  • Trade-off between PN and SN Performance: There's an inherent trade-off. Optimizing for the primary network's outage probability doesn't necessarily yield the best performance for the secondary network, and vice-versa. The paper identifies distinct optimal values for ρ and α for each network.
  • Impact of Power Allocation:
    • High power sharing (large α) for the primary signal: This significantly benefits the PN, leading to lower outage probabilities, as more power is dedicated to relaying the primary signal.
  • High power splitting (large ρ) for energy harvesting: While beneficial for the ST's energy reserves, a very high ρ can lead to a weaker primary signal being decoded by the ST, potentially increasing the PN's outage probability. Conversely, a low ρ means less energy for the ST's transmission, impacting both PN and SN.
  • Cooperative Gain: The cooperative scheme, by leveraging the ST as a relay, offers better outage performance for the PN compared to a direct transmission (DT) scheme, especially when the ST allocates substantial power to the primary signal.
  • Location Matters: The positions of the ST and SR significantly influence the SN's performance, with increased distances leading to higher outage probabilities due to signal degradation. The PN's performance, however, is shown to be less sensitive to the SR's location.
  • Noise Impact: Higher composite noise (μ) at the RFBCUs degrades the performance of both networks, increasing outage probabilities.

Key Results

  • Optimal Values Identified: The paper identifies specific optimal values for the power splitting ratio (ρ) and power sharing coefficient (α) that minimize outage probabilities. For instance, for the PN, an optimal ρ of approximately 0.4 with α = 0.9 yields an outage probability around 0.06. For the SN, an optimal ρ of 0.1 with α = 0.1 results in an outage probability around 0.09.
  • Performance Improvement: The proposed cooperative scheme with energy harvesting and power superposition coding demonstrates improved outage performance for the PN compared to a direct transmission scheme, particularly when a significant portion of power is allocated to the primary signal.
  • Validation: Numerical results and Monte-Carlo simulations show good agreement, validating the derived analytical expressions for outage probabilities.

Limitations and Future Work

  • Idealized Assumptions: The analysis relies on assumptions such as Rayleigh block fading, perfect channel state information, and ideal decoding at the ST. Real-world scenarios might involve more complex channel models and imperfect decoding.
  • Single Relay: The system model considers a single relay sensor node. Future work could explore scenarios with multiple relays and relay selection mechanisms.
  • Limited Scope of Parameters: While the paper analyzes the impact of several parameters (ρ, α, Ψ, location, noise), further investigation into other factors like imperfect sensing, interference from other secondary users, or different energy harvesting models could be beneficial.
  • Future Directions: The authors suggest exploring scenarios with multiple relays and investigating the impact of imperfect sensing and interference. The analytical expressions derived can serve as a foundation for further research in this area.

Summary updated

BibTeX
@inproceedings{son2017optimal,
  author = {Pham Ngoc Son and Dongsoo Har and Nam Ik Cho and Hyung Yun Kong},
  title = {{Optimal Power Allocation of Relay Sensor Node Capable of Energy Harvesting in Cooperative Cognitive Radio Network}},
  booktitle = {Sensors},
  year = {2017},
  url = {https://openreview.net/forum?id=Cj1BPoa-mI65},
}
Figure