Enhanced unipolar orthogonal frequency division multiplexing (eU-OFDM) and layered asymmetrically clipped optical OFDM (LACO-OFDM) are spectrally efficient modulation techniques for intensity modulated systems which layer multiple non-negative signals. In this paper, we propose absolute value layered asymmetrically clipped optical OFDM (ALACO-OFDM), which further improves the spectral efficiency while using a smaller number of layers and no explicit direct current (DC) bias. In ALACO-OFDM, asymmetrically clipped optical OFDM (ACO-OFDM) signals are sent at the first L layers and absolute value optical OFDM (AVO-OFDM) is used for the remaining subcarriers and transmitted simultaneously. Analysis indicates that ALACO- achieves higher spectral efficiency than eU- or LACO-OFDM while using a smaller number of layers. Bounds on achievable information rates of ALACO-OFDM and related layered ACO-OFDM techniques are also developed. Two optical power allocation schemes over the layers of ALACO-OFDM are developed with the objective of optimizing uncoded transmission performance and the achievable information rate respectively. Additionally, a theoretical bound on the uncoded BER of ALACO-OFDM is derived. Monte Carlo simulation results indicate ALACO-OFDM with the optimum power allocation achieves significant uncoded BER performance gains compared to its counterparts at the same spectral efficiency while having a smaller peak-to-average power ratio.

In this paper, antisymmetry-constructed clipped optical OFDM (AC- OFDM) is proposed for intensity modulation with direct detection (IM/DD) systems, in which an antisymmetry property is imposed directly in time domain. AC-OFDM has the same spectral efficiency as traditional asymmetrically clipped optical OFDM (ACO-OFDM) but is less complex to implement. Layered AC-OFDM (LAC-OFDM) is then proposed as an extension to further improve spectral efficiency, where different layers of AC-OFDM signals are added in the time domain and transmitted simultaneously. Computational complexity analysis and numerical results show that LAC-OFDM has the same spectral efficiency as layered asymmetrically clipped optical OFDM (LACO-OFDM) and enhanced unipolar OFDM (eU-OFDM) but is less complex in terms of both the number of real multiplications and real additions required. Specifically, LAC-OFDM requires less than half the multiplication and addition operations compared to the comparable LACO-OFDM scheme. Monte Carlo simulation results show that LAC- OFDM has the same bit error rate (BER) performance as LACO- and eU-OFDM.

Both enhanced unipolar orthogonal frequency division multiplexing (eU-OFDM) and layered asymmetrically clipped optical OFDM (LACO-OFDM) are among the most spectrally efficient modulation techniques for intensity modulated links that layer multiple non-negative signals. In this paper, we propose absolute value layered asymmetrically clipped optical OFDM (ALACO-OFDM) which further improves the spectral efficiency while using a smaller number of layers and no direct current (DC) bias. In ALACO-OFDM, asymmetrically clipped optical OFDM (ACO-OFDM) signals are sent at the first L layers and absolute value optical OFDM (AVO-OFDM) is used for the remaining subcarriers and transmitted simultaneously. Analysis indicates that ALACO-OFDM achieves higher spectral efficiency and smaller peak-to-average power ratio (PAPR) compared to eU-OFDM, LACO-OFDM and asymmetrically clipped absolute value optical OFDM (AAO-OFDM). Monte Carlo simulation results also indicate that ALACO-OFDM achieves significant bit error rate (BER) performance gains compared to its counterparts at the same spectral efficiency.

A 3D positioning method based on visible light communication is proposed. Compared to the previous methods, we only need three light emitting diodes (LEDs) with known coordinates to obtain the object position without the priori knowledge of the receiver's height and tile angle. The gradient descent method and vector method are used to obtain the coordinate and inclination of the object (i.e., camera). To validate the effectiveness of our methods, the relation of mean positioning error caused by the discrete camera sensor pixel and the system parameters (i.e., focal length of the camera, pixel density and height of the receiver) is analyzed. Simulation results show the quantization error is about 5 cm.