Table of Contents
Fetching ...

Performance Analysis of Indoor VLC Network with Secure Downlink NOMA for Body Blockage Model

Tianji Shen, Vamoua Yachongka, Yuto Hama, Hideki Ochiai

TL;DR

The paper tackles indoor VLC downlink security under unknown eavesdropper CSI by introducing a geometry-based body blockage model and a novel equilateral-triangle LED arrangement to minimize LED overlap. It develops three transmission strategies (broadcasting, simple LED linking, and smart LED linking) and two power allocation schemes (fixed and maximum-sum-rate) to maximize both transmission sum rate $R_T$ and secrecy sum rate $R_S$. Through Monte Carlo simulations, the authors show that the proposed LED linking strategies outperform broadcasting in both rate and secrecy across various user distributions, with the max-sum-rate scheme further boosting performance. The work provides practical designs for secure VLC in environments with body blockage and CSI uncertainty, and sets the stage for analytical bounds and optimized LED placement in future studies.

Abstract

In this work, we investigate the performance of indoor visible light communication (VLC) networks based on power domain non-orthogonal multiple access (NOMA) for mobile devices, where multiple legitimate users are equipped with photodiodes (PDs). We propose a body blockage model for both the legitimate users and eavesdropper to address scenarios where the communication links from transmitting light-emitting diodes (LEDs) to receiving devices are blocked by the bodies of all parties. Furthermore, we propose a novel LED arrangement that improves secrecy without requiring knowledge of the channel state information (CSI) of the eavesdropper. This arrangement reduces the overlapping areas covered by different LED units supporting distinct users. We also suggest two LED transmission strategies, i.e., simple and smart LED linking, and compare their performance with the conventional broadcasting in terms of transmission sum rate and secrecy sum rate. Through computer simulations, we demonstrate the superiority of our proposed strategies to the conventional approach.

Performance Analysis of Indoor VLC Network with Secure Downlink NOMA for Body Blockage Model

TL;DR

The paper tackles indoor VLC downlink security under unknown eavesdropper CSI by introducing a geometry-based body blockage model and a novel equilateral-triangle LED arrangement to minimize LED overlap. It develops three transmission strategies (broadcasting, simple LED linking, and smart LED linking) and two power allocation schemes (fixed and maximum-sum-rate) to maximize both transmission sum rate and secrecy sum rate . Through Monte Carlo simulations, the authors show that the proposed LED linking strategies outperform broadcasting in both rate and secrecy across various user distributions, with the max-sum-rate scheme further boosting performance. The work provides practical designs for secure VLC in environments with body blockage and CSI uncertainty, and sets the stage for analytical bounds and optimized LED placement in future studies.

Abstract

In this work, we investigate the performance of indoor visible light communication (VLC) networks based on power domain non-orthogonal multiple access (NOMA) for mobile devices, where multiple legitimate users are equipped with photodiodes (PDs). We propose a body blockage model for both the legitimate users and eavesdropper to address scenarios where the communication links from transmitting light-emitting diodes (LEDs) to receiving devices are blocked by the bodies of all parties. Furthermore, we propose a novel LED arrangement that improves secrecy without requiring knowledge of the channel state information (CSI) of the eavesdropper. This arrangement reduces the overlapping areas covered by different LED units supporting distinct users. We also suggest two LED transmission strategies, i.e., simple and smart LED linking, and compare their performance with the conventional broadcasting in terms of transmission sum rate and secrecy sum rate. Through computer simulations, we demonstrate the superiority of our proposed strategies to the conventional approach.
Paper Structure (26 sections, 2 theorems, 32 equations, 14 figures, 4 tables, 2 algorithms)

This paper contains 26 sections, 2 theorems, 32 equations, 14 figures, 4 tables, 2 algorithms.

Key Result

Lemma 1

Suppose that the vector $\bm{v}_{t} = (x_{t},y_{t},z_{t})$ in Euclidean space passes through the point $M(x_m,y_m,z_m)$ and the point $P(x_p,y_p,z_p)$ on the plane ${\cal W}$ as illustrated in fig:lemma1. If the plane ${\cal W}$ also includes the point $N(x_n,y_n,z_n)$ with one of its normal vectors where the inner product of ${\bm v}_s$ and ${\bm v}_{t}$ satisfies

Figures (14)

  • Figure 1: An illustration of the system model. Cyan dots represent LED transmitters, $D_1, \cdots, D_K$ denote the positions of PDs of the legitimate users, and $D_E$ is the position of the eavesdropper.
  • Figure 2: An illustration of the VLC path between the $n$th LED and the $k$th PD, where $\theta_{k,n}$ is the radiation angle between $S_n$ and $D_k$, $\theta_{1/2}$ is the half illuminance angle of the LED transmitter, $\psi_{k,n}$ is the incidence angle of link between $S_n$ and $D_k$, $\Psi$ is the received field of view (FoV) of the PD receiver, and ${\bm n}_k$ is the normal direction of the photosensitive surface of the $k$th PD.
  • Figure 3: The geometry of visible light communications with $S_n$, $D_k$ and $U_k$. $S_n'$ is the projection point of $S_n$ onto the plane that contains the top base of cylinder, ${\bm n}_k$ is the normal direction of the photosensitive surface of $D_k$ with its projection at XOY plane is given by ${\bm n}_k$, $\omega_k$ is the azimuth angle between ${\bm n}_k$ and $x$-axis, $\lambda_k$ is the polar angle of $D_k$, $\theta_{k,n}$ is the radiation angle from $S_n$ to $D_k$, $\psi_{k,n}$ is the incidence angle of the light beam from $S_n$ to $D_k$, and $\phi_{k,n}$ is the azimuth angle between the vector $\overrightarrow{S_n U_k}$ and $y$-axis.
  • Figure 4: A mathematical explanation of the body blockage model in this work. The $n$th LED light is positioned at $S_n(x_{S_n}, y_{S_n}, Z)$, while the PD receiver for the $k$th user is located at $D_k(x_{D_k}, y_{D_k}, z_{D_k})$. A cylinder, who is modeled as the receiver holder of $D_{l}$ with $l \in {\cal K}'$, is centered at $U_{l}(x_{U_{l}}, y_{U_{l}}, {\sf h}_{D_l})$ on its top plane, with an azimuth angle $\phi_{l,n}$ between itself and the $n$th LED source. The light vector ${\bm a}_{k, n}$ projects onto the XOY plane as $\tilde{\bm a}_{k, n}$. The plane ${\cal P}_{l, n}$, which passes through the point $U_{l}$, is perpendicular to the vector $\tilde{\bm a}_{k, n}$. Additionally, the vector ${\bm a}_{k, n}$ intersects with the planes ${\cal Q}_{l}$ and ${\cal P}_{l, n}$ at the points $Q_{l, n}$ and $P_{l, n}$, respectively.
  • Figure 5: Illustration for the relation of intersection point of vectors and plane ${\cal W}$ in Lemma 1. Points $N$ and $P$ are on plane ${\cal W}$, ${\bm v}_s$ is the normal vector to plane ${\cal W}$, and ${\bm v}_t$ is the vector have same direction as $\overrightarrow{MP}$ (i.e., ${\bm v}_t = a \overrightarrow{MP}$ with real factor $a$).
  • ...and 9 more figures

Theorems & Definitions (3)

  • Lemma 1
  • proof
  • Proposition 1