Despite the availability of high-speed telecommunication facilities in the neighbourhood of last-mile locations such as business centres, offices and homes, there are still some technological constraints that hinder high-bandwidth availability. Moreover, wide area network (WAN) between major cities is fibre based and extremely fast (>2.5Gbps). Local area network (LAN) in buildings is also fast (>100Mbps). Most of these businesses run high-speed data networks within their buildings, such as fast Ethernet (100Mbps) or gigabit Ethernet (1Gbps). But the connections in between access networks are typically a lot slower (64kbps-14Mbps), whether they are wired or wireless (Fig. 1).
Wireless access network has its constraints such as electromagnetic interference (EMI)/electromagnetic compatibility (EMC) problems, atmospheric attenuation, hazardous electromagnetic radiation and limited throughput, scarcity of frequency spectrum availability.
Wire-line connections using copper lines—such as xDSL (digital subscriber line) (8Mbps), T-1 (1.544Mbps), E1 (2Mbps) and cable modem (5Mbps shared)—have speed and/or reach limitations. Although fibre-optic solution—for example, synchronous transport module, say (STM1) (155.52Mbps), STM4 (622.08Mbps), STM16 (2588.32Mbps), STM64 (9953.28Mbps)—could be a very attractive option for providing high-speed connectivity, till date only five per cent buildings are lit up with fibre-optics due to huge cost and time involved in taking permission, digging, trenching and laying these cables under existing streets, sidewalks, lawns, buildings, etc.
Security is for the most part non-existent on these connections and is dependent upon preventing physical access to the cabling. But if such high-speed connectivity could be served through some other viable telecommunication technology, there would be no need to build a new infrastructure for last-mile access. This is possible through free-space optics (FSO) technology.
What is free-space optics
Free-space optics is a high-bandwidth cost-effective solution to the last-mile problem. It is a wireless, point-to-point (PTP), fibreless, laser-driven and line-of-sight optical communication technology that uses invisible light propagating in free space to transmit information. FSO can transmit data, voice and video simultaneously through the air at speeds capable of reaching 1.25Gbps at a distance of 6.4 km (four miles) in full-duplex mode—enabling fibre-optic connectivity without requiring a physical fibre-optic cable.
FSO was originally developed by US military and NASA, and is currently being used for more than three decades in various forms to provide fast communication links. With the technological advancement in optical technology, tracking mechanism, implementation of dense wavelength division multiplexing (DWDM) kind of technologies, fourth-generation FSO systems capable of offering speeds up to 10Gbps are expected to hit the markets.
How it works
FSO is a line-of-sight technology that uses a pair of FSO units consisting of an optical transceiver with a laser (transmitter) and a photo detector (receiver) to provide full-duplex (bi-directional) capability between two points—without the fibre.
The transmitters can be either light emitting diodes (LEDs) (single or multiple), typically 1mW, or lasers (single or multiple), typically 10-100mW. The detectors can be positive-intrinsic-negative (PIN) diodes with –43dBm or advanced photodiode (APD) with –53dBm as minimum received power. These transceivers are mounted on rooftops, walls or even windows of buildings pointing at each other (Fig. 2). FSO systems use low-powered invisible infrared laser light of wavelengths in the 750nm to 1550nm range.
FSO system deployment is surprisingly simple, and its link can be installed and aligned in less than an hour. Full installation takes no more than half a day. So, FSO can be used in disaster management and emergency conditions. These systems employ powerful lasers that can transmit through glass and windows, thus further increasing installation options and equipment security. FSO technology solves last-mile bottlenecks in hours, not weeks—and without digging or infrastructure upgrades required by other PTP connections because it does not require expensive fibre-optic cable or regulatory spectrum licences for radio-frequency (RF) solutions. Thus, it offers a fast and high return on investment.
Free-space optical spectrum is nearly unlimited, thus very dense frequency re-use is possible that makes very convenient deployment of FSO systems worldwide. Also, FSO communication is immune to radiofrequency interference or saturation. FSO technology provides fibre-like availability with data rates up to 1.25Gbps (expandable up to 10Gbps) that does so with the low latency, low jitter, low bit error-rate connections.
This technology offers better interoperability with fibre networks and delivers scalable bandwidth supporting all protocols. Moreover, FSO technology’s narrow beam transmission is typically two metres versus twenty metres and more for traditional, even newer radio-based technologies such as millimetre-wave radio. The lasers emit very sharp, focussed and narrow (small angle/small divergence) invisible light that results in very small spot size at a long distance (spot diameter (m) = angle (milliradians)×range (km)). This makes FSO hard to intercept, making it ideal for high-security applications such as financial, military and medical activities.
Fourth-generation FSO systems offer exclusive automatic power-level control, eliminating short-distance optical saturation found in first-generation systems. The latest fourth-generation FSO lasers are 100 per cent class 1M eye-safe technology. Also, they employ active tracking multiple-beam solution. Such systems eliminate mistargeting and provide parallel beam redundancy, solving problems such as a bird flying through the path of a first-generation FSO system.
FSO network topologies
Several network topologies are possible for FSO networks. PTP, star, ring and mesh are all feasible network architectures for laser communications. PTP topology is the simplest of the physical layouts of network devices. PTP connections mean that two devices (nodes) have a single path for data to travel between them with nothing breaking up that path.
On the other hand, in star connections, all devices are connected to a central hub utilising independent links. The central node is usually a hub or a multiplexer that utilises repeaters to forward data.
In a ring topology, all nodes are connected to one another in the shape of a closed loop, so that each node is connected directly to two other nodes, one on either side of it.
In a mesh topology, devices are connected with many redundant interconnections between network nodes. Each node is connected with two to four nodes and is able to receive, transmit and forward data packets. All topologies comprise a combination of PTP links.
Additionally, hybrid topologies including ring, mesh, PTP and/or star interconnections have been proposed for FSO systems (Fig. 3). In ring interconnections, customer nodes can be ring nodes of more than one ring or one hop away from another ring. At the same time, PTP connections with nodes that do not belong to the ring can exist. On the other hand, in star interconnections, central nodes (hubs) are connected with other central nodes utilising PTP links. The primary reason of extending the standard network topologies is the increase of reliability and coverage area whilst keeping the associated cost low.
The choice of network topology significantly affects the performance, reliability, scalability, design complexity and overall cost of a free-space optics network.
There are several signal propagation impediments for FSO that affect light propagation through free space. The turbulences in the signal may be due to absorption, diffraction, scattering, refraction, beam wander, scintillation, beam dispersion, etc that may be caused by physical composition of atmosphere, changes in refractive indices, building sway, fog, rain, dust, smog, snow, interference from background light sources (including the sun), shadowing, etc (Fig. 4).
1. Environmental factors. Fog. Fog is composed of water droplets between a few and a hundred microns in diameter, which allow absorption, scattering and reflection to occur when laser beams pass through it. Fog is a major challenge to FSO communications and can result in atmospheric attenuation of signal (10-100 dB/km). Heavy fog can result in a complete outage of the service.
Low clouds, rain, snow and dust. Low clouds are very similar to fog and may accompany rain and snow that may degrade optical signal significantly.
Rain. Rain can also degrade the performance of FSO transmission. Raindrop sizes are larger than fog so may cause scattering of optical signal. Extremely heavy rain can entirely disrupt the link. Water sheeting on windows can also deteriorate signal.
Heavy snow. Heavy snow may cause whiteout conditions and may result in ice build-up on windows. Laser beams are affected by heavy snow and rain that can cause attenuation to the laser beams of 60-1000 dB/km.
Sandstorms. Sand/dust particles can alter signal path to a finite extent. In desert areas, sandstorms occur very often but they occur rarely in urban areas.
The phenomenon that causes signal degradation due to atmospheric conditions is explained in the following lines:
Absorption. Absorption occurs when suspended water molecules in the terrestrial atmosphere extinguish photons. This causes a decrease in the power density of the FSO beam and directly affects the availability of telecom services. However, typical wavelength values of FSO systems are 785nm, 850nm and 1500nm. These wavelengths correspond to atmospheric windows where the attenuation is very small; thus the absorption attenuation also is very small.
Scattering. Scattering is caused when a light wave collides with the scatterer. The physical size of the scatterer determines the type of scattering. When the scatterer is smaller than the wavelength of light, it is known as Rayleigh scattering, which happens due to atmospheric gas molecules. When the scatterer is of comparable size to the signal wavelength, Mie scattering occurs, which is particularly caused by aerosol particles.
Scintillation. Heated air rising from the earth or man-made devices such as heating ducts creates temperature variations amongst different air pockets. Propagation through air pockets of varying temperature, density and refraction index causes beam spreading and wandering. Scintillation is temporal and can cause spatial variation in light intensity, which leads to increased error rate but not complete service outage.
2. Physical obstructions. Flying birds or some temporary obstruction in the path of FSO signal can temporarily block a single optical beam, but this tends to cause only short interruptions, and transmissions are easily and automatically resumed.
3. Building sway/seismic activity. The movement/swaying of tall structures/buildings due to winds, coal mine blasts and seismic activity can upset receiver and transmitter alignment. Usually building sway is not more than 4-10 milliradians, but such a small misalignment can also increase bit error rate. In such situations, beam divergence is kept larger than expected building motion. Automatic pointing and tracking is the best option for exploiting optimum performance of an FSO system.
4. Window attenuation. In some cases, where high-altitude rooftop weather losses are more, FSO systems are installed in windows. But, it has been observed that uncoated glass attenuates almost four per cent of signal per surface due to reflection. Tinted or insulated windows can have much greater attenuation. So, a trade-off between these two options (window or rooftop installation) should be achieved.
To sum up
During the last few years, FSO technology has become one of the hottest topics in the telecommunication industry because it has the most promising capabilities to the last-mile bottleneck problems. Technology and economics both favour the optical wireless technology. Although there are several factors that degrade FSO signal performance as well. But, with the advancement in the FSO technology, fourth-generation state-of-the-art FSO systems employing multibeam, multipath architecture, laser auto tracking, devices with large fade margin (extra power reserved for fog, rain, smog, etc), network with shorter link distances, use of eye-safe lasers with limited laser power density, etc have addressed issues encountered by first-generation FSO systems.
The author is a junior telecom officer with Bharat Sanchar Nigam Limited. He holds a Ph.D. degree in electronics engineering from Indian Institute of Technology (BHU), Varanasi. His current research interests include wired and wireless technologies for high-speed internet access