Coverage Analysis in Two-tier 5G Hetnet Based on Stochastic Geometry with Interference Coordination Strategy

Abstract

To meet the coverage requirement in the 5G cellular network, small cells are conceived as an emerging technology to increase coverage and satisfy the traffic demand. However, modeling and analysis coverage is the most important step in cell planning to explore the performance of system. Furthermore, with a large deployment of small cells in a Heterogeneous Network (Hetnet), the cross-tier interference management is a complex problem that needs to be studied. The main contribution of this paper is to analyze the downlink coverage performance for 5G Hetnet where the infrastructure is composed of Macro cell and Small cell. We model the received Signal to Interference plus Noise Ratio SINR at the user and we derive coverage probability according to Stochastic Geometry under different cognitive interference with and without coordination. Simulation result are provided to validate the proposed model.

Appendix A

The PDF of distance $d$ between the target UE and the closet Small Base Station can obtained in both LOS and NLOS cases by deriving the Cumulative Density Function CDF.

$$p_{r}^{NLOS} \left( d \right) = 2\pi \lambda_{m} d{\mathbb{P}}_{NLOS} \left( d \right)\exp \left( {\pi \lambda_{m} d^{{2{\mathbb{P}}_{NLOS} }} \left( d \right)} \right)$$

(18)

where ${\mathbb{P}}_{LOS} \left( d \right) = e^{ - \beta d}$ ${\mathbb{P}}_{NLOS} \left( d \right) = 1 - e^{ - \beta d}$ are the probability of a propagation path being LOS and NLOS respectively.

$$\begin{aligned} {\mathbb{P}}\left( {r > d} \right) = &\, \frac{{\exp \left( { - \pi \lambda_{m} d^{2} {\mathbb{P}}_{NLOS} \left( d \right)} \right) \times \left( {\pi \lambda_{m} d^{2} {\mathbb{P}}_{nLOS} \left( d \right)} \right)^{0} }}{0!} \\ = &\, {\text{exp}}\left( { - \pi \lambda_{m} d^{2} {\mathbb{P}}_{LOS} \left( d \right)} \right) \\ \end{aligned}$$

(19)

$$p_{r}^{NLOS} \left( d \right) = \frac{{\text{d}}}{{{\text{d}}d}}\left\{ {1 - {\mathbb{P}}(r > d} \right\}$$

(20)

$$= 2\pi \lambda_{m} d{\mathbb{P}}_{NLOS} \left( d \right)exp\left( {\pi \lambda_{m} d^{{2{\mathbb{P}}_{NLOS} }} \left( d \right)} \right)$$

(21)