In Internet of Underwater Things (IoUT) scenarios, power-hungry devices need to periodically harvest energy to be alive. By dynamically controlling and redirecting the optical beam, reconfigurable intelligent surfaces (RIS) can enhance both the energy transfer and information reception processes, thereby improving the sustainability of low-power underwater devices. Simultaneous wireless information and power transfer (SWIPT) systems allows to harvest energy and receive data from the transmitted optical beam, thus feeding the low-power receiver (Rx) while downloading data. However, realistic underwater scenario can strongly affect the performance of such systems. This paper investigates the performance of ORIS-assisted SWIPT system under a realistic unified channel model for underwater optical wireless communications (OWC), considering both coherent and non-coherent detection schemes. Considering multiple attenuation losses, underwater turbulence, pointing error, and angle of arrival deviation, we derive the joint outage probability of an ORIS-aided SWIPT OWC system for different transmission protocols based on different resources that are allocated. Three approaches are investigatedi.e., (i)power-splitting (PS),(ii)time-switching (TS), and(iii)element-splitting (ES). While PS and TS are based on all-passive RIS elements, ES considers hybrid (active and passive) elements. Simulation results show enhanced performance obtained with ES approach for low signal-to-noise ratio (SNR), while PS outperforms ES for high SNR.

This paper presents a comprehensive performance study for the underwater optical wireless communication (UOWC) channel, accounting for the effects of attenuation, turbulence, pointing errors, and fluctuations in the angle-of-arrival (AOA), modeled by the Beer-Lambert model, the Fisher-Snedecor$\mathcal {F}$turbulence, Hoyt, and Rayleigh distributions, respectively. Considering the stochastic nature of UOWC, closed-form expressions for outage probability, average bit-error-rate (BER), and ergodic capacity are derived. The results revealed that increasing the field-of-view (FoV) reduces the impact of AOA fluctuations. Asymmetric two-dimensional beam misalignment exhibits better outage probability performance compared to symmetric misalignment. Furthermore, heterodyne detection consistently outperforms direct detection. When employing heterodyne detection, M-quadrature amplitude modulation (M-QAM) achieves superior BER performance. Additionally, as absorption increases with longer wavelengths, BER performance deteriorates. Narrower beamwidths improve ergodic capacity by enhancing received signal power.

In this experimental study, data transmission over step-index plastic optical fiber (SI-POF) is implemented with multilevel pulse amplitude modulation (PAM-M) scheme using a resonant cavity light emitting diode (RC-LED) as the optical transmitter. An artificial neural network (ANN) based equalizer is used in the RC-LED SI-POF system to jointly mitigate channel distortion and system non-linearities. Furthermore, the ANN equalizer's performance is compared to state-of-the-art equalizers used for the system - the Volterra equalizer and the decision feedback equalizer (DFE). The ANN equalizer offers the best bit rates compared to the others for the system with high non-linearities. For instance, at a BER of about 10−3, the ANN equalizer results in a bit rate of 780 Mbps, 710 Mbps, and 650 Mbps with PAM-2, PAM-4 and PAM-8 schemes, respectively. However, the Volterra equalizer results in a bit rate of 720 Mbps, 660 Mbps, and 580 Mbps with these PAM schemes, respectively. And with the DFE, the bit rates are 610 Mbps, 510 Mbps, and 120 Mbps, respectively.

There has been a rise of use of Autonomous Vehicles on public roads. With the predicted rise of road traffic accidents over the coming years, these vehicles must be capable of safely operate in the public domain. The field of pedestrian detection has significantly advanced in the last decade, providing high-level accuracy, with some technique reaching near-human level accuracy. However, there remains further work required for pedestrian intent prediction to reach human-level performance. One of the challenges facing current pedestrian intent predictors are the varying scales of pedestrians, particularly smaller pedestrians. This is because smaller pedestrians can blend into the background, making them difficult to detect, track or apply pose estimations techniques. Therefore, in this work, we present a novel intent prediction approach for multi-scale pedestrians using 2D pose estimation and a Long Short-term memory (LSTM) architecture. The pose estimator predicts keypoints for the pedestrian along the video frames. Based on the accumulation of these keypoints along the frames, spatio-temporal data is generated. This spatio-temporal data is fed to the LSTM for classifying the crossing behaviour of the pedestrians. We evaluate the performance of the proposed techniques on the popular Joint Attention in Autonomous Driving (JAAD) dataset and the new larger-scale Pedestrian Intention Estimation (PIE) dataset. Using data generalisation techniques, we show that the proposed technique outperformed the state-of-the-art techniques by up to 7%, reaching up to 94% accuracy while maintaining a comparable run-time of 6.1 ms.

In this study, we investigate the performance of split-complex multi-layer perceptron (SC-MLP) equalizer for a step index plastic optical fibre (SI-POF) communication link employing the carrierless amplitude and phase (CAP) modulation scheme. The SC-MLP equalizer is proposed to mitigate the effect of intersymbol interference (ISI) and non-linearity in the system. It is demonstrated that the SC-MLP offers significantly superior bit error rate (BER) performance compared to the complex-valued transversal equalizer and the complex-valued Volterra equalizer in a highly non-linear SI-POF system. For example, for a 10 m SI-POF, a data rate of 15 Gbps is attainable at a BER of 10-3with SC-MLP compared to 10 Gbps with the complex-valued transversal equalizer using CAP-64.

We propose and experimentally demonstrate for the first time, M-PAM transmission through a hybrid POF/VLC downlink using off-the-shelf components. The link consists of 60 m SI-POF and 1 m VLC and with the aid of a blue filter, the link has an effective bandwidth of 17 MHz. Conventional adaptive decision feedback equaliser (DFE) and multi-layer perceptron DFE (MLP-DFE) are also implemented and their bit error rate (BER) performances are measured and compared with offline (MATLAB) implementation. For the conventional DFE, the achievable data rates at a BER of 10-3are 60, 91.5 and 80 Mbps for M = 2, 4 and 8 respectively. Higher data rates are achieved for MLP-DFE at the same BER of 10-3as we obtained 61.5, 95 and 85 Mbps for M = 2, 4 and 8 respectively. Thus, we show that MLP-DFE offers superior BER performance than conventional DFE for higher PAM level.

In visible light communications (VLC) system, the linearity of the light emitting diode is an issue when being intensity modulated with an analogue signal. In optical orthogonal frequency division multiplexing (O-OFDM) based VLC systems, the source nonlinearity imposes a tight restriction on the achievable peak-to-average-power-ratio. To address this issue, we propose a hybrid OFDM-PTM (pulse time modulation) scheme, where a bipolar O-OFDM signal is converted into digital PTM formats for switching `on' and `off' the light source. The simulation results demonstrate that the proposed scheme offers an improved bit error rate performance compared to the traditional asymmetrically clipped O-OFDM (ACO-OFDM).

This paper experimentally investigates the use of a single input multiple output (SIMO) free space optical (FSO) communication system in an indoor laboratory controlled turbulence chamber. Two receiver circuits are used with a single laser input. The current outputs of two receivers were combined using equal gain combining (EGC) scheme and compared with single input single output (SISO) FSO system. A weak turbulence was generated within the atmospheric chamber and link performances are measured using ß-factor and bit error rate (BER) for On-Off Keying (OOK) modulation scheme. The results demonstrated that the simplest form of SIMO can improve the performance of the FSO communication channel under weak turbulent condition.

Visible light communications (VLC) has the potential to play a major part in future smart home and next generation communication networks. There is significant ongoing work to increase the achievable data rates using VLC, to standardize it and integrate it within existing network infrastructures. The future of VLC systems depends on the ability to fabricate low cost transceiver components and to realize the promise of high data rates. This paper reports the design and fabrication of integrated transmitter and receiver components. The transmitter uses a two dimensional individually addressable array of micro light emitting diodes (μLEDs) and the receiver uses an integrated photodiode array fabricated in a CMOS technology. A preliminary result of a MIMO system implementation operating at a data rate of 1 Gbps is demonstrated. This paper also highlights the challenges in achieving highly parallel data communication along with the possible bottlenecks in integrated approaches.

This paper presents the first ever experimental demonstration of a multiple-input multiple-output (MIMO) visible light communications system employing four silicon (Si) light emitting diodes (LEDs) and four organic photodetectors (OPDs) as transmitters and receivers, respectively. The proposed link is relatively low cost and it employs the on-off keying (OOK) modulation format offering a data rate of 200 kb/s without the need for equalization, which is a significant increase compared with previous non-equalized systems. In order to speed up date rates further, we implement an artificial neural network (ANN) to classify the signal and correct the error induced by the matrix inversion at the receiver, allowing a gross bit rate of 1.8 Mb/s in the best case.

Binary polarisation shift keying (BPOLSK) has been proposed to mitigate the atmospheric turbulence-induced fading in free space optical (FSO) communication systems. In this study, the Q-factors obtained for the BPOLSK-FSO system are verified in conjunction with theoretical results to confirm the validity of the proposed scheme. The analytical bit error rate (BER) for the BPOLSK and non-return-to-zero on–off keying (NRZ-OOK) schemes are presented. The authors show that the BPOLSK scheme with direct detection offers improved BER performances compared to NRZ-OOK in the presence of weak turbulence, which is inferred from the experimental Q-factor and theoretical BER. For a turbulence variance σ21 of 0.003 and the transmitted optical power of −16.8 dBm, values for Q-factor are ∼11 and ∼8.5 for BPOLSK and NRZ-OOK schemes, respectively. The authors show that the predicted signal-to-noise ratio (SNR) for BPOLSK and NRZ-OOK schemes are ∼13.5 and ∼15 dB, respectively, for a BER of 10−6 and σ21 of 0.01. When σ21 increases to 0.1, ∼8 dB lower values of SNR is required for BPOLSK compared with NRZ-OOK.