Two-User Relay Protocol Based on Energy Harvesting and Cognitive Radio Techniques

Abstract

Energy harvesting (EH) from radio frequency realizes simultaneous wireless information delivery and energy transfer, which greatly reduces energy consumption in the energy-constrained Internet of things (IoT) networks. On the other hand, cognitive radio (CR) enables the unlicensed secondary user (SU) to access the spectrum resource authorized to the licensed primary user (PU) to improve spectrum efficiency. Thererfore, the integration between EH and CR can address two critical problems of energy limitation and spectrum scarcity in massive IoT wireless networks. In the state-of-art power splitting-based overlay spectrum sharing strategy, the secondary transmitter superimposes its own signals on the primary signals received and forwards the combined signals with the harvested power from the primary transmitter. However, due to low energy harvested only from one primary transmitter for the combined signal, PU system performance e.g., outage probability (OP) and system throughput, declines seriously. Moreover, a perfect direct link between the secondary transmitter and the secondary receiver must also be required in the scenario. An alternative spectrum sharing strategy based on time switching can increase harvested energy, but cooperative communication cannot be conducted at the EH period, which also reduces the overall system throughput. Based on the consideration, a Two-User Relay protocol in this paper is presented that an energy-constrained relay node, as a center hub, harvests energy from two transmitter, i.e., primary transmitter and secondary transmitter, and perform cooperative communication concurrently for PU and SU systems satisfying the PU’s performance requirements (e.g., required target rate and OP). The expressions of OP and system throughput are derived and numerical simulations are made under the different system parameters, e.g., power splitting factor, power assignment factor, transmitting signal-to-noise ratio, channel fading coefficient, and required target rate. The results prove that our proposed Two-User Relay protocol improve spectrum sharing performance. Besides, the relay node can coordinate the harvested energy distribution between PU’s signals and SU’s signals to achieve the maximum system performance. Generally, the paper provides a new relaying concept for two-user communications in the future IoT wireless networks.

Appendix

Here we derive the PDF of Z = Z₁ + Z₂.

Because Z₁ and Z₂ are independent, we have

$$\begin{aligned} f_{Z} (z) & = \int_{ - \infty }^{\infty } {f_{{Z_{1} }} (z_{1} )f_{{Z_{2} }} (z - } z_{1} )dz_{1} \\ {\kern 1pt} & = \int_{0}^{z} {\frac{1}{{\rho_{1} \lambda_{1} \rho_{2} \lambda_{2} }}e^{{ - \frac{{z_{1} }}{{\rho_{1} \lambda_{1} }}}} e^{{ - \frac{{z - z_{1} }}{{\rho_{2} \lambda_{2} }}}} } dz_{1} \\ & = \frac{{e^{{ - \frac{z}{{\rho_{2} \lambda_{2} }}}} }}{{\rho_{1} \lambda_{1} \rho_{2} \lambda_{2} }}\int_{0}^{z} {e^{{ - \frac{{z_{1} }}{{\rho_{1} \lambda_{1} }}}} e^{{\frac{{z_{1} }}{{\rho_{2} \lambda_{2} }}}} } dz_{1} \\ \end{aligned}$$

If $\phi \lambda_{1} - \lambda_{2} = 0$, we have

$$f_{Z} (z) = = \frac{{ze^{{ - \frac{z}{{\rho_{2} \lambda_{2} }}}} }}{{\rho_{1} \lambda_{1} \rho_{2} \lambda_{2} }}$$

else

$$\begin{aligned} f_{Z} (z) & = \frac{{e^{{ - \frac{z}{{\rho_{2} \lambda_{2} }}}} }}{{\rho_{1} \lambda_{1} \rho_{2} \lambda_{2} }} \times \frac{{\rho_{1} \lambda_{1} \lambda_{2} }}{{\phi \lambda_{1} - \lambda_{2} }}\int_{0}^{z} {e^{{\frac{{(\phi \lambda_{1} - \lambda_{2} )z_{1} }}{{\rho_{1} \lambda_{1} \lambda_{2} }}}} } dz_{1} {\kern 1pt} \\ & = \frac{{e^{{ - \frac{z}{{\rho_{2} \lambda_{2} }}}} }}{{\rho_{2} (\phi \lambda_{1} - \lambda_{2} )}}\int_{0}^{z} {e^{{\frac{{(\phi \lambda_{1} - \lambda_{2} )z_{1} }}{{\rho_{1} \lambda_{1} \lambda_{2} }}}} } dz_{1} {\kern 1pt} \\ & = \frac{{e^{{ - \frac{z}{{\rho_{2} \lambda_{2} }}}} }}{{\rho_{2} (\phi \lambda_{1} - \lambda_{2} )}}\left( {e^{{\frac{{(\phi \lambda_{1} - \lambda_{2} )z}}{{\rho_{1} \lambda_{1} \lambda_{2} }}}} - 1} \right){\kern 1pt} \\ & {\kern 1pt} = \frac{{e^{{ - \frac{z}{{\rho_{1} \lambda_{1} }}}} - e^{{ - \frac{z}{{\rho_{2} \lambda_{2} }}}} }}{{\rho_{2} (\phi \lambda_{1} - \lambda_{2} )}} \\ \end{aligned}$$

This ends the proof.□