Using the NI PXI Platform for LiFi-Enabled Intelligent Transportation Systems

Author(s):

Murat Uysal - Ozyegin University, Centre of Excellence in Optical Wireless Communication Technologies
Burak Kebapci - Ozyegin University, Centre of Excellence in Optical Wireless Communication Technologies
Omer Narmanlioglu - Ozyegin University, Centre of Excellence in Optical Wireless Communication Technologies

About Us

The Centre of Excellence in Optical Wireless Communication Technologies (OKATEM) is Turkey’s first R&D center that specializes in the emerging area of optical wireless communication (OWC). Established at Ozyegin University with generous government funding in September 2015, OKATEM carries out fundamental and applied research in OWC systems operating at infrared, visible, or ultraviolet bands. With participation from other universities, telecommunication companies, and related NGOs in the Istanbul region, OKATEM aims to create a collaborative research environment and promote technical innovation and entrepreneurship in OWC technologies and application areas.

One of our major research areas is on visible light communication, also known as LiFi. This form of OWC uses the illumination infrastructure as wireless access points. LiFi systems are based on the principle of modulating light emitting diodes (LEDs) at very high speeds without any adverse effects on the human eye and illumination levels. LEDs are increasingly used both indoors (home and office lighting, and more) and outdoors (street lights, traffic lights, vehicle front/rear lights, and more). The dual use of LEDs for both lighting and communication purposes is a revolutionary solution and has the potential to open a new era in wireless communications.

Before we started to use the NI platforms, we built LiFi testbeds that consisted of standard benchtop instruments. With NI’s PXI platform, the available communication blocks, and the LabVIEW FPGA Module as the supporting HDL layer, we saw the opportunity to achieve superior results with a shorter development time. We have also enjoyed continuous support through the extensive ecosystem around the NI platform, including users from both academia and industry, with a supporting layer of partners and NI local branch offices delivering technical support.

About the Project Challenge

In an effort to improve road safety, traffic flow, and environmental concerns, there has been a growing interest in the field of intelligent transportation systems (ITSs). Vehicular networking, an essential component of ITSs, requires vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications so vehicles can communicate with each other and with roadside units installed along the road. V2V and V2I communications gather real-time information on traffic and road conditions including collisions, congestion, surface conditions, traffic signal violations, emergency brakes, and more to provide to drivers. Besides these road safety functionalities, potential in-vehicle applications have recently emerged such as high-speed internet access, multiplayer gaming, and mobile commerce as a result of ever-increasing dependency on the Internet and multimedia services.

Research activities and standardization efforts in ITSs have so far mainly focused on the deployment of RF-based communication techniques for vehicular networking. The impact of V2V and V2I communications on the amount of RF spectrum usages is currently low, but we expect this to significantly increase in the near future with the widespread adoption of ITSs. The allocated small portion of RF bands can quickly suffer from interferences when hundreds of vehicles located in the same vicinity try to communicate simultaneously.

As a potential alternative or complementary technology to RF-based solutions, we considered the use of LiFi for V2V and V2I communications. The fact that LEDs are becoming common in automotive lighting and roadside infrastructure provides unique opportunities for the widespread use of LiFi in ITSs. Vehicles fitted with LED-based front and back lights can communicate with each other and with the roadside units (for example, street lights and traffic lights) through LiFi technology. Recent experimental works have used some simple modulation techniques such as on/off keying, pulse place modulation, and more to investigate the performance of LiFi-based V2V and V2I links. Such works have demonstrated the feasibility of the vehicular LiFi systems. However, in many respects, this technology is still in its infancy and requires further research efforts in several areas, including the development of high-performance physical layer techniques.

To achieve high data rates over a multipath optical channel, we should use advanced multicarrier waveforms such as orthogonal frequency-division multiplexing (OFDM). Different from its RF counterparts, the deployment of OFDM in a LiFi system requires certain modifications to ensure the non-negativity of the intensity-modulated optical signal. Therefore, we need computationally efficient and customizable FPGA platforms for the real-time implementation of OFDM-based vehicular LiFi systems.

Application Overview

At OKATEM, we built an OFDM-based LiFi demonstrator for V2V applications using NI PXI solutions. After convolutional channel coding, we mapped the input bit stream to the 4-QAM modulation symbols. These then pass through the N-IFFT block to form the OFDM waveform. We applied Hermitian symmetry and added a DC bias voltage to ensure that the resulting OFDM waveform was real and non-negative. We then fed the waveform to the LED through a bias-tee and sent it through the optical channel. At the receiver side, a photodetector (PD) captured the optical signal and baseband processing took place in the electrical domain.

In our project, we implemented a V2V communication system using NI PXI Express FlexRIO FPGA boards as baseband signal processors and FlexRIO adapter modules (NI 5781 and NI 5772) with onboard high-speed wideband A/D converter and D/A converter channels to form the baseband path of the optical receiver and transmitter. By using LabVIEW and LabVIEW FPGA with proper parameter selection, we demonstrated real-time transmission, which gave superior performance results over traditional CPU-based transceivers and simple on/off keying based systems.

At OKATEM, we also built a custom-designed atmospheric emulator with dimensions of 60 cm x 40 cm x 300 cm. We equipped it with adjustable heaters, coolers, fans, a fog generator, and a droplet watering system to generate different weather conditions (sunny, windy, rainy, hazy, and more). We monitored the conditions within the emulator through temperature, humidity, visibility, altitude, pressure, dust, and wind speed sensors. We made all controls through a touch screen. This emulator provided a controlled environment to test the LiFi system under different weather conditions.

Our group is currently working on transferring the CPU-based modulation and decision algorithms to FPGA space. Our initial results of using FPGAs are positive. By off-loading the CPU, and thanks to the true-parallel processing of FPGAs, we can gather more data throughput and transmit more sub-channels within the same band.

Conclusion

We presented an initial performance evaluation study of a LiFi-based V2V system using NI platforms. Our results demonstrate that depending on the PD location in the car (particularly its height above ground level), we can achieve a data rate of 50 Mb/s for a distance up to 70 m between two cars. We plan to deliver more content and insights for the real-time performance of this topic. Particularly, we are working to address the FPGA implementation issues and improve the total system performance by integrating advanced capabilities such as multiple-input, multiple-output (MIMO) communication and link adaptation on the top of OFDM architecture.

Author Information:

Murat Uysal
Ozyegin University, Centre of Excellence in Optical Wireless Communication Technologies
Ozyegin University
Tel: (216) 5649329
murat.uysal@ozyegin.edu.tr