Abstract:
Most of the current wireless networks are radio frequency (RF) based, however, these networks are regularly confronted with growing demand for higher data rates. Watching high-definition live stream videos and accessing
cloud-based services are the main user activities that rapidly consume the data capacity. The higher data rates also lead to a significant increase in energy consumption. Further, the demand is much more intense for indoor
communication, where the maximum data usage occurs. This leads to extra load on the base station (BTS) and poor quality of services (QoS) at the end-users. Furthermore, the indoor radio channel heavily depends on
factors such as building structure, room layout, and construction materials used. Consequently, there is severe attenuation of RF waves at the receiver. Furthermore, for multi-storey buildings, floor-wise attenuation factor
(FAF) increases toward lower floors. FAF for each successive floor is reduced by a factor of 6 dB. Therefore, in conjunction with RF communication, visible light communication (VLC) can provide potential solutions to address the issues as mentioned above faced by RF communication in the indoor environment. Motivated by the above, this thesis first aims to reduce the power consumption at BTS caused due to strong attenuation of radio signals in an indoor environment by proposing a hybrid cellular-VLC link. This is achieved by replacing the end-user connection with a VLC link in the indoor environment. VLC access points (AP) act as a decode-and-forward (DF) relays, which decode the received signal transmitted from the BTS and forward it to the indoor user using LEDs. Earlier work on hybrid RF-VLC systems has not considered the amount of power saving achieved by using VLC as an indoor link. Further, the existing works have not optimized the Lambertian order for uniform delay spread and high average received optical power inside the room. The recent
research also ignored the effect of dimming on the VLC link. Additionally, the analysis so far has used a static VLC channel model and has not considered the impact of user movement, type of room, or shadow objects in the room on the performance of the hybrid RF-VLC system. Hence, in this thesis, we have proposed a hybrid cellular-VLC downlink framework and have analyzed its performance for indoor environments by considering all the above parameters. This work also analyze hybrid RF-VLC system performance for a single user between RF and VLC modulation schemes of the same order. For example, VLC on-off-keying (OOK) with RF binaryphase- shift-keying (BPSK) and VLC modified color-shift-keying (M-CSK) with RF M-ary quadrature amplitude modulation (M-QAM) schemes of the same order. Further, this work also calculates the amount of power saving in the VLC link as compared to the RF link.
Light-emitting-diode (LED) deployment also plays an essential role in the VLC system, as the received optical power distribution (ROPD) varies with respect to the LED’s placement and the user’s location inside the room. Consequently, ROPD can be improved by optimizing the placement of LEDs inside the room. Hence,in our subsequent work, this thesis utilizes the Matern hardcore point (MHCP) process to propose a random placement of LEDs in an indoor scenario to achieve uniform signal-to-noise-ration (SNR) and improve bit-error-rate (BER) performance at the receiver. It has been shown that MHCP is the desirable and more appropriate process for LED placement since it results in the minimum separation between any two LEDs for better coverage and reduced interference. The system is analyzed under two receiver configurations, i.e., with the non-imaging receiver (1-FOV and 2-FOV) and with the imaging receiver.
Studies show that the received optical power in the indoor VLC system depends on the distance between the transmitting LEDs and the desired user inside the room and the presence of static and dynamic blockages. In a
multi-user scenario, the other users inside the room act as blockages for the desired user. These blockages result in a sudden fall in the received optical power as it can block both the line-of-sight (LoS) and the non-line-of-sight
(NLoS) signals from the LED to the desired user. The amount of power reduction will depend on the height and width of the blockage. In order to analyze the impact of static and dynamic human blockages on indoor VLC
systems performance, as our third objective in this thesis, we adopt a stochastic-geometry-based approach to study the performance of an indoor VLC system in the presence of human blockages. In particular, we consider
two types of blockages: static and mobile, and characterize the impact of the density of the blockages on the received signal strength of a receiver uniformly placed inside the room. Contrary to the existing studies on indoor VLC systems, which typically ignore the impact of human blockages, our investigation reveals that the blockages considerably impact the propagation environment and significantly alter the system design insights. Moreover, unlike conventional RF communication systems, the optical wireless communication (OWC) channel is not isotropic, meaning that the user equipment (UE) orientation and obstacles significantly affect the gain
of VLC. Specifically, the performance of VLC may severely deteriorate when the LoS link gets blocked due to other users/obstacles. Further, the received power may also fluctuate due to the random orientation of UE as
NLoS power varies. In this thesis, in order to combat the shadowing due to obstacles and the UE orientation, as our fourth objective, we employ optical intelligent reflecting surfaces (OIRS) in indoor VLC systems in the
presence of multiple human blockages. Moreover, we propose the UE orientation model of users for an OIRS-aided indoor VLC system. The LoS channel gain statistics are calculated, and the orientation of UE is modeled
as a random process. By utilizing the above, the impact of the random UE orientation and human blockages on SNR performance of VLC systems is evaluated. In addition, the probability distribution function (PDF) of
received SNR with OIRS is also calculated. The proposed analysis also deduces the optimum LED semiangle and the receiver FOV for the OIRS-aided indoor VLC system.
The recent interest in using visible light as a means of communication has opened up possibilities for using visible light for other applications as well, such as indoor positioning. Visible light offers higher bandwidth, immunity from electromagnetic radiation, and most importantly, it can be seamlessly integrated into the existing lighting infrastructure. As this thesis’s fifth objective, we propose a visible light-based positioning system for estimating an object’s three-dimensional (3-D) parameters, such as height and radius, in addition to the location in an indoor environment. The model is built using neural networks, trained by simulating multiple object scenarios in an indoor environment as well. It also takes into account the shadowing effects so it can be implemented in a multiple obstacle environment. The proposed algorithm has numerous applications, such as positioning assisted communication, suspicious object monitoring, and surveillance in an indoor environment.
Further, in earlier work, utilization of this location information for indoor communication has not been explored. We believe that this position information can be exploited to improve communication performance
in the presence of different obstacles inside the room. Moreover, the LED power allocation can be optimized to maximize the data rate or minimize the BER by exploiting this location information. Hence in this thesis as a final objective, we propose a location-assisted indoor VLC system, wherein the location information is exploited to enhance the communication performance of the user. Specifically, we propose an optimal LED power allocation framework to maximize the average data rate across the room subject to predefined communication constraints as well as a number of blockages inside the room.