Home Engineering Broadband Wireless Communications for Railway Applications: For Onboard Internet Access and Other Applications
Studies in UK
Other works in UK were realized in order to evaluate performance of an OWC system to obtain a broadband Internet access on board trains. All the following papers are written by Paudel et al. from the Northumbria University of Newcastle.
In , the possibility to use MIMO optical systems is mentioned, in order to increase throughputs. However, the emphasis is placed on the description of a SISO optical system. Laser diodes are used and tests are performed on the size of the lens and the transmission power. The link budget of the developed system on short distances is computed. Nonetheless, no mobile experiments with higher distances (several hundreds of meters) were performed.
In , the communication protocol of the system is described. A base station is placed at a distance equivalent to two carriages of a train (around 42 m). Two receivers are placed for each carriage. Two types of applications are considered: the outdoor case and the indoor case (tunnel). In the first case, receivers are put on the roof of the train and base stations are deployed along the track, at the same height than receivers. In the second case, base stations are put on the ceiling of the tunnel, the configuration on train is the same. A system of control of switch on and off of the base stations with the passage of the train is presented. First tests on received power are realized. Optimizations on received power level with optical concentrators are presented.
In , the effect of turbulence on OWC is studied. Tests are performed in laboratory using an atmospheric chamber. The system is based on a transmission device using a LED, an infrared LED, an optical lens and a data source. The LED is modulated with a Non-Return-to-Zero (NRZ) and On-Off Keying (OOK) pseudo-random signal. Results show a high resistance of the developed system to the turbulence.
In , tests on throughputs depending on BER are performed in simulation, by varying the distance between two consecutive transmitters. The envisioned system is as follows. Transmitters consist of LED. They are put every 75 m on high voltage electric pylons located at about 1 m from the track. One transmit can cover three carriages (length of a carriage around 21 m). The receivers, consisting of photodetectors are positioned on the side or on the roof of the train at a height of 4m approximately. A Lambertian model can represent this system. Simulations are performed with the Matlab tool in order to evaluate system performance. Thus, in order to obtain a BER of 10-6, the optical coverage is around 75 m (3 carriages) for a throughput of 10 Mbps, 42 m (2 carriages) for a throughput of 100 Mbps, and 21 m (one carriage) for a throughput of 1 Gbps.
Finally, in , a model of the system specially designed for railway environment is presented. Laser are here used instead of LED, in order to obtain larger coverage area and more power. The system is then modeled by a Gaussian source, instead of a Lambertian source. A link budget analysis is performed showing a link margin of 17.75 dB for the worst atmospheric conditions. Simulation results with the Matlab tool are given in terms of BER performance of the system. It is showed that it is possible to have beam coverage up to 75m for throughputs up to 50 Mbps.
Summary on Optical-Based Solutions
Works presented in this section highlighted different aspects. First of all, it appears that works performed in UK are largely dominated by simulations and no measurement in real sites, with railway constraints were performed yet. Conversely, works in Japan are quite advanced and promising for a new option for providing Internet on board trains. Very high throughputs can be obtained at very high speeds. Throughputs up to several hundred of Mbps were measured in a real site. However, the major drawback remains the cost of CAPEX and OPEX of optical-based solutions. Optical terminals have to be deployed at least every 400 m along the track, which leads to a very high investment. Furthermore, this rolling out leads to high cost of maintenance. Finally, optical solutions performance are very dependent on atmospheric conditions.
|< Prev||CONTENTS||Next >|