MIMO Visible Light Communications with Camera-Based Receiver for Intelligent Transport Systems

Bani Hassan, Navid (2019) MIMO Visible Light Communications with Camera-Based Receiver for Intelligent Transport Systems. Doctoral thesis, Northumbria University.

Text (Doctoral thesis)
Bani Hassan.Navid_phd.pdf - Submitted Version

Download (6MB) | Preview


The number of vehicle accidents in the world each year is considerable which yield to an approximately 1.2 million deaths. Besides, the car manufacturers are now investing significantly on autonomous vehicles, which highlights the need for communication between vehicles and the surrounding environment. Intelligent Transportation Systems (ITS) enables the vehicles to realize this need. In ITS, vehicles and infrastructure can communicated directly without the need of cellular networks. Currently, the developed technology for ITS is based on radio frequency (RF) under the name of dedicated short-ranged communication (DSRC), which shares the frequency spectrum with other RF applications such as fixed satellite and wireless services, mobile services, radiolocation, etc. Therefore, DSRC can be a potential interfering source for other communication services. Besides, the licence to use this spectrum makes this service costly in the frequency range of 5.855-5.925 GHz and in heavy traffics may cause severe packet collision. Alternatively, visible light communication (VLC) can be used to release the pressure on RF. The equipment required for VLC, such as light emitting diodes (LEDs), camera, and computer, is already available on most of the vehicles nowadays. Therefore, VLC can be considered as a less expensive option for ITS. Most of the works done in camera-based VLC, also known as optical camera communication (OCC), consider a line-of-sight (LOS) link from the transmitter (Tx) to the receiver (Rx). However, in some scenarios, the LOS link might not be available. In this thesis, a non-LOS (NLOS) link is considered to decrease the probability of outage. Accordingly, this thesis highlights another advantage of using camera compared to photodiodes, i.e., the Rx can extract the data information from the off-axis projection of the Tx on the road surface when the LOS link is blocked or is not available. An end-to-end NLOS OCC system is proposed which employs differential signalling and frame subtraction. Throughout the thesis, different detection techniques are proposed based on spatial division multiplexing (SDM) and time division multiplexing (TDM), hybrid selection/equal gain combining (HS/EGC), hybrid selection/maximal ratio combining (HS/MRC), and zero forcing (ZF) equalization, respectively. The thesis deals with different types of the road in terms of the density of the light sources. An experimental investigation of the proposed system shows that, using HS/EGC and simple detection zero forcing equalization (SDZFE), the system can achieve the 7% overhead forward error correction limit of 3.8×10-3 at a very low transmit power of 9 dBm over a link span of 5 m at a data rate of 30 bps, when the Tx is 2 m above the floor surface. It is also shown that by doubling the ISO level, exposure time and aperture size, the performance of the system improves by ~3 dB. In addition, by increasing the link span from 5 to 10 m, the power penalty is ~3 dB. This is because at very low light environment, the intensity of received light is close to the nonlinear region of the camera and since the footprint shrinks as the link span increases, the nonlinearity affects the signal less in longer link spans. The thesis states that the focal length and focusing distance of the camera does not make a significant impact on the performance of the system. Therefore, the camera can have a large field of view with a wide-open aperture. In addition, the thesis demonstrates that the ZF-based detection schemes outperform the HS/EGC schemes under severe level of interference. It is shown that, in ZF-based cases the normalized height of the eye diagram in is 70% higher compared to HS/EGC when the spacing between Txs is double the height of them.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: zero forcing, equal gain combining, non-line-of-sight, image sensor, interference
Subjects: H600 Electronic and Electrical Engineering
Department: Faculties > Engineering and Environment > Mathematics, Physics and Electrical Engineering
Depositing User: Paul Burns
Date Deposited: 05 Aug 2019 09:55
Last Modified: 02 Sep 2022 08:30
URI: https://nrl.northumbria.ac.uk/id/eprint/40257

Actions (login required)

View Item View Item


Downloads per month over past year

View more statistics