1. INTRODUCTION

FSO may sound new and experimental but in fact it predates optical fibre and has its roots in wartime efforts to develop secure communication systems that did not require cable and could withstand radio jamming.

 As a commercial communications technology, FSO has been around for more than a decade, but it is only recently that interest in this technology has started to grow. It was only recently developed for use in metropolitan area networks. The technology has its roots in military applications that reach back as far as the 1940s.

It was not until the 1960s, however, that the first significant FSO technology advancements began to occur in the United States, Europe and Middle East, where military researchers, engineers and technicians applied the use of infrared lasers in communications devices with the aim of providing secure data and voice transmission that would not be susceptible to “jamming” of radio frequency-based communications systems.

These early FSO systems were capable of transmitting merely a handful of kilobits over the air, but theadvent of the Internet and its impact on telecommunications was decades away. In fact, European researchers of FSO systems in the 1960s experimented with ways to send FSO signals through both underground and underwater pipes, seeking to bend the invisible light beams with mirrors where a straight line-of-site could not be established.

2. FUNDAMENTALS OF FREE SPACE OPTICS

FSO is an optical wireless, point-to-point, line-of-sight broadband solution.

1. Lasers Through Free Space

FSO is an optical technology and simple concept involving the transmission of voice, video and data through the air using lasers. It is not a disruptive technology; it is more of an enabling technology that promises to deliver that ever-eluding high-speed optical bandwidth to the ultimate end users. FSO offers many advantages when compared to fiber. It is a zero sunk-costs solution. The principle advantages of free space optics (FSO) are:

1. Significantly lower cost on average than the build out of a new fiber optical solution, or leased lines

2. FSO can be deployed in days to weeks vs. months to years

3. Bandwidth can easily be scaled (10 Mbs to 1.25 Gbps) per link
As opposed to fiber, FSO can be redeployed if the customer moves or cancels service. It is also a fraction of the cost and time, allowing carriers to generate revenue, while also taking advantage of the high capacity of optical transmissions. FSO allows service providers to accelerate their deployment of metro optical networks as well as extend the reach of such optical capacity to anyone who needs it.

2. FSO and Fiber Optics

FSO systems share several characteristics with fiber optics. FSO can use the same optical transmission wavelengths as fiber optics, namely 850nm and 1550nm and they use the same components such as lasers, receivers and amplifiers. Some systems already include fiber connections inside the transmission link heads, to separate electronics and optics. Similar to fiber optics, FSO systems also target the high-bandwidth market. However, while fiber optics can be used over longer distances, FSO targets shorter distances due to the variability of the terrestrial atmosphere as a transmission medium.

One common feature of FSO equipment commercially available today is that most of these systems perform optical to electrical back to optical (O-E-O) conversion steps in the process of sending and receiving information through the air and connecting back to the attached networking interface fiber. This feature does not automatically constitute a performance limitation, but O-E-O conversion can impact the ability to scale an FSO system easily to ultra-high bandwidth capabilities. The fiber optic communications industry realized from the start the importance of an all-optical system approach, as higher backbone capacity — along with wavelength division multiplex technology. An important breakthrough to reach this goal occurred when fiber systems with erbium doped fiber amplifier (EDFA) became commercially available. It was then, that the concept carrying multiple wavelengths over a single piece of optical fiber achieved commercial attention. The invention of EDFA amplifier technology paved the way for optical transmission at multiple wavelengths over longer distances without the need to perform expensive O-E-O conversion and separate electrical amplification of each specific wavelength at every repeater station.

3. Bandwidth Drivers/Trends

The push to build more high-speed networks was spurred by unprecedented growth in bandwidth usage. Telecommunications carriers will implement multiple technologies in their networks and will use the best access technology for the particular situation. The chart below shows how these technologies address different market segments based on technology, technical capabilities (reach, bandwidth), and economic realities.

Fig.1.A number of compelling factors are influencing this bandwidth surge:

1. Metro Optical Network Deployments

Substantial investments by carriers to augment the capacity of their core fiber backbones have driven dramatic improvements in both price and performance and substantially increased the capacity of these large backbone networks. However, to generate the communications traffic and revenue needed to fully utilize and pay for these backbone upgrades, higher bandwidth connections must reach the end customers. This requires substantial bandwidth upgrades at the network edge. Essentially, to fully leverage their backbone investments, service providers will also need to expand and extend the reach of their metropolitan optical network to the edge. FSO presents an opportunity that allows carriers to achieve that goal for one-fifth the

cost when compared to fiber (if it is even available) and at a fraction of the time.

2. Increased Competition

Regulation changes and significant investments by various funds have increased the competitive climate in these metro networks. Each of the existing or new entrants is “racing” to gain an advantage over their competition. FSO is one of the evolutionary technologies that allows a carrier to acquire and retain new customers quickly and cost-effectively, thereby gaining an entry point over competition. The lack of infrastructure and rising bandwidth demands offer a unique opportunity for FSO.

3. Changing Traffic Patterns and Protocol Standards

Multiple traffic types characterize metro networks. Where voice was once the dominant traffic type, data has emerged as the winner. Moreover, these networks are also a mixture of multiple protocols ranging from SONET, IP, ESCON, FICON, etc.

4. Wireless World

With the rapid adoption and slow deployment of wireless technologies such as LMDS and MMDS in response to high bandwidth communication needs in the metro area, many service providers still find themselves short of bandwidth to satisfy all needs. Most RF-based technologies require spectrum licenses and provide only limited bandwidth.

1.  FSO SYSTEM DESIGN ISSUES

4. Free-Space Optics Subsystems

Fig.3.1FSO Major subsystems

Figure  illustrates the major subsystems in a complete carrier-grade free-space optics communications system. The optical apertures on a free-space system can have an almost infinite variety of forms and some variety of function. They can be refractive, reflective, diffractive, or combinations of these. In figure2 , the transmit, receive, and tracking telescopes are illustrated as separate optical apertures; there are several other configurations possible where, for example, a single optic performs all three functions thereby saving cost, weight, and size. On the transmit side, the important aspects of the optical system are size and quality of the system. Size determines the maximum eye-safe laser flux permitted out of the aperture and may also prevent blockages due to birds. Quality, along with the f-number and wavelength, determine the minimum divergence obtainable with the system. On the receive side, the most important aspects are the aperture size and the f-number. The aperture size determines the amount of light collected on the receiver and the f-number determines the detector's field of view. The tracking system optics’ field of view must be wide enough to acquire and maintain link integrity for a given detector and tracking control system. 

Several means are available for coupling the laser to the output aperture. If a discrete

diode is used, the diode is usually micro-lensed to clean up the astigmatism of the output beam and then is free-space coupled to the output aperture by placing the laser at the focus of the output aperture optical system. The distance from the laser aperture to the output aperture must be maintained such that the system divergence remains in specification over the temperature ranges encountered in an outdoor rooftop environment. This can be accomplished with special materials and/or thermal control.

Diode lasers are driven with a DC bias current to put the devices above threshold, and

then, on top of that, are modulated with an AC current to provide, for example, on/off

keying (OOK) for data transmission. For lasers with output powers below approximately 50 mW, off-the-shelf current bias and drive chips are available; for higher power lasers, custom circuits or RF amplifiers are generally used.

The receive detector is coupled to the receive aperture through either free-space or fiber. Depending on the data rate and optical design alignment, tolerances can be extremely restrictive. For example, for data rates to 1.25 Gbit/s, detectors with relatively large active areas (500-micron diameter) can be used, making alignment to the receive aperture fairly straightforward. For fiber-optic coupling into multimode fibers, the correct size is about 63 microns in diameter, which makes alignment much tougher.

Detectors are generally either PIN diodes or avalanche photodiodes (APD). For carrier class free-space optics systems, an APD is always advantageous since atmospheric induced losses can reduce received signals to very low levels where electronics noise dominates the signal-to-noise (SNR) ratio. Of course the APD must be capable of meeting the system bandwidth requirements. Usually a trans-impedance amplifier is used after the detector because in most cases they provide the highest gain at the fastest speed.

5. Bit Error Rate, Data Rates, and Range

In figure, which depicts a set of buildings in Denver, Colorado, the effects of fog on visibility range is illustrated. The tall building in the foreground is about 300 m from the photographer. The left photo shows clear air, at 6.5 dB/km (2000 m visibility range), as measured with a nephelometer mounted at the photographer's site.The distant mountain ranges are easily visible at many miles distance. During a fog which measured about 150 dB/km (visibility range of about 113 m), as shown in the middle photo, the building is still visible at 300 m, but the scenery is washed out beyond this range. As shown in the right photo, at 225 dB/km (visibility range of about 75 m) the building is completely obscured.

Fig.3.2 Denver, Colorado Fog/Snowstorm Conditions

Table: 3.1 Environmental Attenuation

2. FSO SYSTEMS AND NETWORK SECURITY

FSO systems operating in the near infrared wavelength range do not require licenses worldwide for operation. FSO system installations are very simple, and the equipment

requires little maintenance. Because FSO systems send and receive data through the air—or the “free space” between remote networking locations—network operators and administrators are concerned about the security aspect of this technology.

Such concerns are not valid for FSO systems. FSO systems operate in the near infrared wavelength range slightly above the visible spectrum. Therefore, the human eye cannot visibly see the transmission beam. The wavelength range around 1 micrometer that is used in FSO transmission systems is actually the same wavelength range used in fiber-optic transmission systems. The wavelength range around 1 micrometer translates into frequencies of several hundred terahertz (THz) which is higher than that used in commercially available microwave communications systems operating around 40 GHz.. FSO systems use very narrow beams that are typically much less than 0.5 degrees. E.g., a radial beam pattern of 10 degrees roughly corresponds to a beam diameter of 175 meters at a distance of 1 kilometer from the originating source, whereas a beam of 0.3 degrees divergence angle typically used in FSO systems corresponds to a beam diameter of 5 meters at the same distance1. This wide spreading of the beam in microwave systems, combined with the fact that microwave antennas launch very high power level is the main reason for security concern. To overcome security concerns, the microwave industry uses wireless encryption protocols (WEP) to protect the transmission path from being intercepted.

The interception of FSO systems operating with narrow beams in the infrared spectral wavelength range is by far more difficult. The small diameter of the beam of typically only a few meters in diameter at the target location is one of the reasons why it is extremely difficult to intercept the communication path of an FSO system:

The intruder must know the exact origination or target location of the (invisible) infrared beam and can only intercept the beam within the very narrow angle of beam propagation. Fig.4. shows an actual example of a 4 m rad beam projected onto the target location where the opposite terminal is located. At a distance of 300 meters the beam diameter is about 1.3 meters, while at a distance of 1 kilometer the beam expands to 4 meters.

Fig.4.1  Example of beam spot diameters at various distances for a beam divergence angle of 4 mrad.

The direct interception of an FSO beam between the two remote networking locations is basically impossible because the beam typically passes through the air at an elevation well above ground level. Due to the fact that the transmission beam is invisible and that any attempts to block the beam would occur near the FSO equipment terminus points, the transmission process imposes another obstacle. Picking up the signal from a location that is not directly located within the light path by using light photons scattered from aerosol, fog, or rain particles that might be present in the atmosphere is virtually impossible because of the extremely low infrared power levels used during the FSO transmission process. The main reason for excluding this possibility of intrusion is the fact that light is scattered isotropic ally and statistically in different directions from the original propagation path. This specific scattering mechanism keeps the total number of photons or the amount of radiation that can potentially be collected onto a detector that is not directly placed into the beam path well beyond the detector noise level. Fig 5. illustrates the physics of this scattering mechanism.

 Fig .4 . 2 Illustration of the physics of the light scattering mechanism while the light beam travels from the originating laser sources (left) to the receiver at the opposite communication location.

3. FSO DRIVERS

The key drivers for FSO: market, economic, service, business and environment are as shown

Fig 5.1 FSO Drivers

6. MARKET DRIVERS

Increasing Number of Internet Users/Subscribers

Increasing E-Commerce Activities

MMDS/LMDS

Deployment of 3G and 4G

7. ECONOMIC DRIVERS

Faster Service Activation

Ultra-scalability and Bandwidth Allows for Lower Inventory Costs

Multiple Applications/Services Support

8. SERVICE DRIVERS

Increasing Demand for High-Speed Access Interfaces

Need to Eliminate the Metro Gap

Need for Real Time Provisioning

4. FSO CORE APPLICATIONS

Common applications of FSO include:

6.1 Metro Network Extensions

FSO can be deployed to extend an existing metro ring or to connect new networks. These links generally do not reach the ultimate end user, but are more an application for the core of the network.

6.2 Enterprise

The flexibility of FSO allows it to be deployed in many enterprise applications, including LAN-to-LAN connectivity, storage area networking and intra-campus connections. FSO can be deployed in point-to-point, point-to-multipoint, ring or mesh connections.

6.3 Fiber Complement

FSO may also be deployed as a redundant link to back-up fiber. Most operators deploying fiber for business applications connect two fibers to secure a reliable service plus backup in the event of outage. Instead of deploying two fiber links, operators can deploy an FSO system as the redundant link.

6.4 DWDM Services

With the integration of WDM and FSO systems, independent players aiming to build their own fiber rings may use FSO to complete part of the ring. Such a solution could save rental payment to Incumbent Local Exchange Carriers (ILECs), which are likely to take advantage of this situation.

5. FSO CHALLENGES

FSO performance can be affected by some conditions:

Weather severity at which FSO signal attenuation can be impacted Rain at 6 inches per hour, Wet snow rate of 4 inches per hour, Dry snow rate of 2 inches per hour, Fog with visibility of < 6% of the transmission distance

7.1 Physical Obstructions

Birds can temporarily block the beam, but this tends to cause only short interruptions and transmissions are easily resumed.

7.2 Scintillation

Heated air rising from the ground creates temperature variations among different air pockets. This can cause fluctuations in signal amplitude which lead to “image dancing” at the receiver end.

7.3 Safety

The safety of FSO is often a concern, since it uses lasers for transmission. This challenge has more to do with perception than reality. The two major concerns typically expressed involve questions about human exposure to laser beams and high voltages within the laser systems and their power supplies. Several standards have been developed covering the performance of laser equipment and the safe use of

lasers. Safety of the lasers does not depend on its frequency, but rather on the classification of the laser. There are two primary classification bodies, the CDRH and the IEC. Commercial systems on the market today are compliant with both standards.

8. Application of FSO in Mobile Wireless Infrastructure

As the number of mobile communications subscribers is expected to reach 500 million worldwide by 2003, mobile operators in most countries are working to offer next-generation wireless services with greater voice capacity and enhanced multimedia applications such as streaming audio and video. These services will be delivered to users on a platform commonly known as “3G,” or Third Generation, technologies that offer the capacity and bandwidth to offer an always on, ubiquitous wireless experience. Network planners of interim and next-generation (2.5G/3G) wireless networks are faced with a number of significant challenges when modeling and executing infrastructure deployment plans. For incumbent service providers, these challenges include prioritizing overlay coverage areas for next-generation data services that accelerate customer adoption—utilizing as much of the current access infrastructure as possible—as well as cost-effective coverage augmentation and gap-filling strategies. For “Greenfield” (new entrants to a market) operators, the challenge is more significant: determining the most expedient path to a launch-ready coverage area in key markets through the use of shared or self-constructed infrastructure, followed by aggressive build-out plans to expand coverage. Bandwidth requirements for interconnection and backhaul of 2.5G/3G base stations will be significantly greater than that of 2G networks. In fact, bandwidth demands in most cases will increase from 2 megabits (Mbps) to between 8 Mbps and 34 Mbps initially, with future requirements estimated to be 155 Mbps and beyond. Although licensed microwave and leased line facilities have served the 2G markets well, technical limitations and cost-effectiveness will not be enough to satisfy both the bandwidth and business requirements for next-generation networks. In addition, because of the higher transmission frequencies used in 3G, a significantly greater number of new base stations will be required for 3G networks, which will tax financial and planning resources for both established operators and new entrants to markets. Operators will be driven by time and profitability pressures to be creative with network topologies and architectures, utilizing new technologies and tools available to reach the goal of rapid cost-effective deployment. Indeed, to achieve aggressive build-out and cash-flow schedules, 3G license holders will turn to a variety of technologies for the interconnection and backhaul of cell-site traffic. These technologies and transmission

methods include copper, fiber-optic cable, licensed microwave—and increasingly—Free-Space Optics (FSO).

Free-Space Optics as an effective access and backhaul technology for mobile network infrastructure has been largely overlooked until recently. FSO is the process of transmitting optical bandwidth through the atmosphere via infrared beams of light, rather than through copper, fiber-optic cable, or radio frequencies. As with licensed microwave, the technology is line-of-sight and requires a stable platform for deployment. FSO, though has many features that have created its emergence as an effective tool to relieve and accommodate the deployment pressures facing mobile operators today and well into the future.

The Free-Space Optics technologies fit within the rapid deployment plans, infrastructure cost constraints and considerations for the longer-term bandwidth and OPEX requirements essential for a successful 3G business.

Additionally, the paper aims to illustrate similarities of deploying FSO to other technologies, as well offering the benefits and limitations of the technology. This white paper focuses mainly on the European and Asian wireless markets, as these regions of the world are further along in the process of deploying 2.5 and 3G systems. A white paper with a focus on the North American and South American wireless markets will be a future topic for 2002.

8.1 Mobile Wireless Infrastructure

Constructing the access section of a 2G or 3G network is a significant percentage of the total infrastructure investment of any incumbent or Greenfield mobile operator. Aside from the capital expenditures of constructing copper or fiber facilities or acquiring spectrum licenses for PDH/SDH radios, there will be increased operational expenses from the use of leased copper or fiber facilities through a PTT when a mobile operator’s own facilities cannot meet the needs of the network. More importantly, infrastructure investment is relative to the increased number of sites required for a 3G network as compared to a 2G network as well as the increased bandwidth required to each base station or point of aggregation. Incumbent operators face an additional—nonetheless separate—challenge. As many incumbents operating

existing 2G networks upgrade sites to 3G, both technologies will share, and therefore operate, on many of the same physical sites. This will require not only additional infrastructure at each site (NodeBs) but additional bandwidth as well. In general, the existing bandwidth to each site will not be sufficient to accommodate access for both services. Whereas a single E1 connection may have been sufficient to each cell site in a 2G network, 3G cell sites or shared 2G/3G sites will require a minimum of 4xE1, and will gradually require as much as STM-1 to each site.

As with any technology, Free-Space Optics has limitations, primarily related to distance and availability caused by atmospheric attenuation. Overlying the advancement that has occurred on speed, management and interfaces, much of the development of FSO over the last two years has focused on tools that help accommodate for its technical limitations—just as planning tools have done so for licensed and unlicensed microwave systems.

In the licensed RF world, the development and use of network planning and management tools have verified the technologies’ value and accommodated for its shortcomings. As a result, its use is rarely questioned, and mobile operators embrace use of the technology. Similarly, mobile operators now have similar tools at their command, offering the same deployment assurances as other wireless technologies, with the same or greater economic benefits.

Planning tools aside, FSO has many fundamental advantages that separate it from the access alternatives, including bandwidth scalability, speed of deployment and license-free operation that meet the demographic, geographic, and regulatory needs of the mobile operator and increasing customer demands.

8.2 Technology Choices for 3G Infrastructure Deployment

In this document, only Layer 1, physical media connections will be highlighted along with the associated network elements in the UMTS standard. In addition, the document will only highlight necessary physical connections in a mobile operators access and core network. The two fundamental components of the UMTS access network are:

NodeB - a physical unit for radio transmission/reception with cells. In 2G networks, the BTS (or Base Transceiver Station) is the equivalent of a NodeB. The primary task of the NodeB is the conversion of data from the handsets, as well as measurement of the quality and strength of the connection, transporting this data to the RNC (listed below). It is possible that a single NodeB serves one or more cells, but, depending on coverage and capacity requirements, not necessary.

In 3G networks that overlay existing 2G networks, a 3G NodeB may be co-located with a 2G BTS for cost savings.

RNC – a physical unit for the management and central control of radio resources in the network.

In 2G networks the BSC (or Base Site Controller) is the equivalent of an RNC. Mobile operators have an array of theoretical choices when planning to expand or upgrade networks. The necessary physical connections in the UMTS model are:

• NodeB to NodeB connection (Iub connection)

• Aggregation or extension of multiple cells or NodeBs to a single point (star configuration)

• Closing links between RNC sites (Iur connection).

• Extension of RF sectors for more efficient use of spectrum.

The following are descriptions of the types of access technologies used in a mobile operators network and the advantages and disadvantages associated with each.

8.3 Copper Leased-Line Facilities

Leased line facilities provided through a PTT have provided relatively reliable connections in 2G networks between BTSs and traffic aggregation from BTSs to the BSCs. Leased line facilities are generally available

when the BSC/BTS is located within a commercial building or large residential building. They are generally more expensive and less reliable when brought to a freestanding location. In fact, according to a Lucent Technologies estimate, 50 percent of cell site downtime can be attributed to T1/E1 outages.

The other significant issue with copper leased line facilities is bandwidth. In 2G networks, a single E1/T1 connection was sufficient to carry voice channels and overhead data for the network. This will not be sufficient for 3G or overlay 2G/3G networks. The current 3G standards allow for great flexibility in the amount of bandwidth that can be delivered to a NodeB, but it is generally accepted that each NodeB will require a minimum of 4xE1. As 3G networks evolve from the current ATM structure to Ethernet and IP, bandwidth requirements may be as much as STM-1 at the edge of the network.

Single T1/E1 connections are relatively inexpensive at $500 to $800 in North America, but higher in Europe and the Far East. However, NxE1 connections are not only significantly more expensive, they are more difficult to provision and take longer to install and test.

Key Advantages: Availability, ubiquity

Key Disadvantages: Bandwidth, reliability, high recurring operating expenses

8.4 Fiber

Fiber facilities provide both the bandwidth and the availability required for future-proof bandwidth to 3G sites, and therefore will prove to be one of the most effective enablers of next-generation wireless technologies. Because 3G services utilize frequencies as high as 2000Mhz, signal attenuation characteristics will require a significantly larger number of cell sites to cover the same geographic area as a 2G network. With this reality, there is a higher probability that optimal NodeB or BTS placement will be on or near a fiber route in densely populated urban areas. In general, mobile operators will be required to make use of existing fiber facilities from either the PTT or a competitive carrier, as building these connections will prove far to costly to justify the technological and economic returns.

Shared tower providers whose business model includes providing ample bandwidth to each site are aware of this economic reality and completing long-term tower licensing agreements with both incumbent and Greenfield wireless operators eager to deploy their networks quickly. Nonetheless, network planners must strike a balance between the number of fiber-fed, shared towers, adding to operating expenses, with private coverage extensions from these sites via either licensed RF or Free-Space Optics.

Key Advantages: Virtually unlimited bandwidth, availability

Key Disadvantages: High recurring operating expenses (from leasing fiber); long, costly deployment/ provisioning (both from leasing and from self-deploying)

8.5 PDH and SDH Microwave Systems

Point-to-point microwave is the dominant access and backhaul technology in mobile networks. Conversely, backhaul is the largest market for point-to-point microwave systems. Even in the recent telecom downturn in almost all segments, the point-to-point microwave market actually grew by 7 percent between 2000 and 2001.1 The technology has provided mobile operators the quickest means for network rollout and capacity expansion. The reasons for this are fairly simple – the operator saves on the costs of trenching cable or leased line connections from a PTT or competitive provider. As a result, estimates are that two-thirds of all base station connections are microwave.

The most utilized frequencies in mobile wireless networks are 18Ghz, 23Ghz and 38Ghz. Although 58Ghz radios are capable of accommodating shorter distance hops (less than 1km) in 3G networks, they are susceptible to weather degradation of signals (rain fade) and operate in the unlicensed band. To accommodate for greater attenuation, higher power may be used. However, this approach can also cause interference to other radio links or other nearby networks. If the links are used in a topological star configuration, adaptive power control can be difficult to utilize as varying power between closely placed links can cause interference. The market offers many tools to plan deployment of radios of varying frequencies and bandwidths that create a systematic and logical approach to link deployment.

However, obtaining licenses (in some frequency bands a license is required each individual link) and the coordination of frequencies to reduce interference and conserve spectrum can be a daunting task. In addition, licensed frequencies can be difficult to obtain in some countries. The trend is to seek higher frequencies, but they pose their own problems: shorter ranges, greater attenuation due to rain, etc.

Nonetheless, mobile operators may find it difficult to both acquire enough spectrum to accommodate the number of base stations in a dense urban or urban area and will also have challenges in planning and deploying the required number of links.

Key Advantages: Low operating expenses, fast provisioning

Key Disadvantages: Tedious deployment (related to spectrum planning), license fees, lower bandwidth

8.6 Free-Space Optics

Free-Space Optics as a technology has been in development for more than 30 years. In fact, prior to sending light through glass (as with fiber optics), light was sent through free space. At its early stages of development, however, severe attenuation limited the technology, especially in communications networks.

In the past decade, however, vast technological improvements have resulted in higher bandwidth, longer distances and higher availability that meets or exceeds those of other wireless technologies, such as SDH/PDH radios. In addition, the development of planning and atmospheric tools offers a look and feel to the behaviors of SDH/PDH planning, although the parameters are different and there are no licensing issues

with which to contend.

These technological improvements have attracted the attention of both landline and mobile wireless operators, both of which are constantly seeking methods to cost-effectively increase velocity of services, accelerating payback periods and leveraging capital budgets. For mobile operators, this means providing coverage expansion or coverage augmentation quickly and generating revenue. Because of the cost effective

availability and bandwidth characteristics of FSO, the technology is gaining a more important position in the toolset of wireless operators.

FSO can be planned with similar parameters and performance to PDH and SDH radio systems. As with microwave, the technology is line-of-sight and limited in distance capability by the atmosphere. In many cases, though, the technology is more effective than microwave, as there are no spectrum considerations to be made when deploying FSO. Free-Space Optics is a license-free, environmentally friendly technology that can be used anywhere in the world. As mobile operators currently using licensed PDH/SHD radios are continually searching for more effective use of spectrum to accommodate for the new bandwidth and topology requirements of next generation networks, FSO has the capability to ease pressure of network planners attempting to make the most efficient use of limited frequencies for RF systems. In many cases, use of the technology may completely negate the need for the use of SDH/PDH radios in many mobile wireless applications, therefore limiting the need for mobile operator to acquire additional spectrum.

FSO is high-performance optical bandwidth, and therefore generally operates at optical speeds, similar to fiber. Unlike analog RF bandwidth, which is requires manipulation and error correction to reach speeds above STM-0, FSO is effective at high speeds, and can therefore easily accommodate lower speeds.

Bandwidths can be reduced to as little as 4xE1, 8xE1, 16xE1, or can move well into the SDH hierarchy, easily attaining STM-0, STM-1, STM-4, and even STM-16, at distances beyond 1km, depending on bandwidth. In addition, FSO can migrate to higher speeds easier than microwave, as the same wavelength can be used for all transport distances below STM-4. Beyond STM-4, accommodating technology features such as beam tracking and higher wavelengths are used.

FSO fits well into dense urban and urban applications. Depending on the atmospheric conditions of a highly populated area, FSO can comfortably transmit bandwidth up to 1.5km. In regions where there is little fog, network planners can deploy the technology confidently well beyond 2km. As FSO attenuates in conditions such as dense fog, operating distances may be longer or shorter, depending on the market.

However, unlike RF, the FSO signals do not attenuate significantly in rain, making the technology especially useful in areas where heavy rain is prominent, such as Southeast Asia and many areas in Europe. Comparatively, deployment distance capabilities are greatly dependent on the frequency utilized. The lower the frequency, the longer the deployable distance with carrier-grade availability. Free-Space Optics becomes a logical alternative in applications where microwave at 38Ghz and above would normally be used, as the FSO achieves comparable availability.

With licensed microwave, BER performance can vary from 10-3 to 10-9, depending on the quality of the system. At most bandwidths, Forward Error Correction is necessary to accommodate for the inherent error properties of microwave systems. The use of FEC increases the latency in the link as well as the overall network, which can cause network performance and quality degradation. As bandwidth to the individual network components increases over time, so to will the latency at each point in the network utilizing microwave.

Because FSO behaves similar to fiber, the technology generally operates at a BER of 10-12, helping improve overall network performance with reduced latency. In fact, because light travels through the atmosphere faster than through glass, latency through the free space segment of the link actually decreases. For copper leased line services, the tradeoff is relatively simple. Similar to licensed RF, the advantage of

FSO is ownership of the connection. Utilizing leased line services, no matter what the bandwidth, requires a mobile operator to incur monthly operational expense that varies from $400 per month to as much as $4,000 per month per connection, depending on the connection speed and distance from the PTT central office.

In addition, FSO solutions have consistently proved approximately one-fifth the installation and provisioning costs of trenching physical fiber, taking less than one-fifth to one-tenth the time to install and provision.

Regulatory approval and construction of fiber installations can increase deployment times significantly, negatively impacting financial returns expected from coverage extension or augmentation of the mobile wireless network.

For mobile operators deploying infrastructure utilizing licensed PHD/SDH radios, operators are faced with myriad regulatory, licensing and planning fees not required for deployment of FSO systems. If a mobile operator invests heavily in the use of PHD/SHDH radios, maintenance costs associated with frequency licenses as well as frequency coordination activities will also be circumvented with the use of FSO. In addition, FSO has the flexibility to be deployed behind windows, on towers or on buildings. Therefore roof rights are not always necessary.

Because most FSO systems can operate behind windows, on towers, atop buildings, or any combination, network operators have a variety of deployment options. In many instances, NodeB equipment will be placed in a commercial building, either in a basement or on an upper floor. If roof rights are obtained for the BTS in the same building, the FSO unit can be used in the same area as the BTS on the building.

However, if the line-of-sight backhaul route is away from the BTS, a mobile operator and contract for space behind a window.

FSO is simple to deploy, requiring as little as two to three hours for setup and alignment. As a general rule, the time to provision power and interface facilities within a building takes longer than the actual installation, alignment and testing of the FSO link.

Most FSO systems utilize one or two separate or combined methods to accommodate for tower installations:

Beam divergence and multiple transmit beams. Generally, beam divergence is a factory pre-set feature in which the narrow beam of the laser is deliberately dispersed into a cone shape that increases in diameter over distance. This divergence helps accommodate for building or tower sway as the result of wind or earth movement.

Multiple beams take the incoming data from the fiber and split it into identical paths, which are then sent through a diverged beam and through free space. There are FSO systems that use two, three and four redundant paths. Redundant paths serve several purposes. First, redundant paths accommodate for heat scintillation, which is a result of temperature variations in the atmosphere as the beam travels. Secondly, multiple beams help reduce the effects of atmospheric turbulence, such as wind, vortices and other thermal sources that can cause minor distortion of the laser beams. Multiple beams also accommodate for temporary airborne physical obstructions, such as birds. In multiple beam FSO systems, only one beam has to reach one receiver for the transmission to be completed error free. Systems that utilize four transmit lasers and four receive lenses offer the most robust configuration are considered to be "carrier grade" and have been deployed in carrier networks.

FSO link heads can operate with as little as 20 watts, making the technology more power efficient than most microwave units of similar bandwidth. In fact, operating at OC-3, many FSO systems utilize 25 percent of the power of competing microwave systems, saving mobile operators operational expenses.

8.7 Other Mobile Wireless Applications

Several network equipment manufacturers are deploying NodeB/BTS site extension solutions utilizing Free-Space Optics for transport. These technologies give both incumbent and Greenfield mobile operators the ability to extend or move sectors for coverage expansion or sector optimization. In addition, these technologies can feed indoor coverage antenna arrays for buildings, airports, or train stations.

RF signals are normally transmitted to and from an antenna array as an analog signal to a cell site with coaxial cable. The distance between the BTS/NodeB cabinet and the antenna array is limited due to loss of signal strength as the connection length increases. However, there are many instances where BTS/NodeB location does not coincide with the optimal antenna array placement to cover a sector. In these cases, mobile operators are challenged with trading optimal configuration in an adjacent sector to accommodate for less than optimal configuration in surrounding sectors.

Using these site extension technologies, an analog signal from a BTS is transmitted through the vendor’s equipment and then sent, via Free-Space Optics, to another location, usually within 200 meters to 800 meters from the BTS cabinet. The antenna array covering one or more sector is installed at this remote location.

There are several unique advantages to using this technology in mobile networks:

• The placement requirements of the antenna array will no longer be limited to the location of the BTS/NodeB equipment. For example, using this technology, an operator can lease space in a building for the BTS/NodeB equipment and place the antenna array on a road sign or power line several hundred meters away. For sector optimization, an operator can co-locate BTS/NodeB equipment in a shared facility and place the antenna array at a location offering the best coverage for the sector.

• High capacity BTS/NodeB equipment can service multiple antenna arrays from one location, reducing leased space operating expenditures. An operator split BTS/NodeB equipment into sectors, and, utilizing FSO, can service multiple antenna arrays from one site. Historically, a mobile operator would need to equip each antenna array with either micro- or pico-cell equipment at each site, resulting in increased capital expenditures for equipment and operating expenditures for site leases.

8.8 Planning Network Deployments – Familiar Ground, Familiar Tools

Planning FSO deployments are somewhat similar to that of PDH/SDH radios. Bandwidth and distance requirements are checked against atmospheric conditions for the deployment area. Error performance and availability results are checked against the requirements for the overall network. FSO links are deployed when the planning variables meet the design parameters of the FSO system.

However, whereas microwave deployment tools generally require conducting interference analysis, scatter analysis, and frequency coordination analysis, these are not necessary with FSO systems. Because FSO utilizes unlicensed infrared light, there is no interference with similar frequencies or concern for signal degradation from interference of similar signals.

9. Comparison of WiFiber to Free Space Optics

GigaBeam's WiFiberTM digital millimeter-wave radio is the first FCC approved product to exploit the recently released 71 to 76 GHz and 81 to 86 GHz frequency bands. WiFiber offers true full duplex Gigabit data rates (1.25 Gbps or 1 GigE) in a cost effective radio architecture. Transmission distances of over 1 mile can be achieved with carrier-class 99.999% availability under all weather conditions throughout most of the USA.

Wi Fiber faces several competing technologies to bridge the last mile gap and to provide high speed broadband connectivity. This paper demonstrates how WiFiber competes effectively and offers significant advantages over free space optics systems.

Free space optic (FSO) technologies operate in the very highest regions of the frequency spectrum, near visible light. FSO technologies employ a laser transmitter to generate a focused optical light wave that carries data through the atmosphere to an optical receiver located at a fixed distance from the transmitter.

Very wide bandwidths are available in the optical part of the spectrum that enable high data rates up to 1 Gigabit per second and beyond. However transmissions in these optical frequency ranges are drastically shortened by fog. The result is that FSO communication links provide carrier-class 99.999% availability for distances of only a few hundred yards in fog prone coastal and metropolitan areas. Because of this limitation, some FSO equipment vendors have been bundling their equipment with microwave radio links to compensate for this weakness.

Several physical effects need to be accounted for and overcome in planning and installing any high resilience, single beam optical transmission path :

. Pointing effects: Birds flying through a narrow optical beam can block the path, causing momentary outages that would affect timing-sensitive data traffic. Similar effects occur with any airborne particles such as snow, sand, dust, flying debris, or even residue from agricultural burning.

. Precise alignment: Tower sway or movements of solid buildings as they naturally heat up and cool down during the day can misalign narrow beam systems.

. Scintillation: Longer east I west facing optical links can be affected by diurnal sunlight effects.

Properly designed FSO equipment compensates for these effects through multiple beam architectures and beam tracking technologies. These equipment enhancements result in complex and unwieldy equipment, with increased cost and reduced equipment reliability consequences.

Concerns over laser-based technologies include limited MTBFs as the useable life of any laser deteriorates through laser use and aging. Also any laser system raises concerns over eye safety. Reputable FSO vendors certify their equipment against recognized eye-safety standards. Although slight compliance is unregulated and voluntary. Finally, the optical part of the spectrum is unlicensed and unregulated, affording the user no federal rights or protection for the link.

FSO systems can be deployed for high data rate applications including' fiber extension, fiber backup. and for disaster recovery. However short practical transmission distances plus equipment reliability and usability considerations have slowed the deployment of such systems or resulted in the swap out of such 'equipment for more reliable and robust radio based technology. Most FSO equipment vendors have acknowledged these limitations and have started offering their equipment bundled with a parallel radio product to mitiqate these effects.

Wi Fiber offers transmission data rates equivalent to the highest speed FSO links, with robust performance over significantly longer distances. Being based on time-proven and well understood radio technology, Wi Fiber links are easy to architect and system performance can be accurately predicted and guaranteed over much longer transmission distances.

WiFiber's millimeter-wave radio architecture offers the only way to provide high availability multi gigabit wireless connectivity at distances of 1 mile and beyond.

Table 8.1

CONCLUSION

FSO, an emerging technology will grow from its current nascent state to a strong niche player within three years. FSO equipment currently is being deployed for a variety of applications, such as last-mile connections to buildings, which may provide the greatest opportunity since FSO provides the high-speed links that customers need without the costs of laying fiber to the end user. In 2005, last-mile access will represent over two-thirds of the total FSO equipment market.

Currently, the increase in high-bandwidth applications at the edge of the network, coupled with the lack of a high-speed infrastructure connecting the edge to the core, has turned the threat of a connectivity bottleneck into reality affecting the end-users, service providers who face delays in laying fiber and building optical infrastructure, resulting in incomplete networks, lack of revenue and increased competition. FSO allows them to provide this optical connectivity cost effectively, quickly and reliably. Such flexibility makes FSO systems an extremely attractive method for service providers to truly solve the connectivity bottleneck.

Free-Space Optics communication systems are among the most secure networking transmission technologies. Unlike microwave systems, it is extremely difficult to intercept the FSO light beam carrying networking data because the information is not spread out in space but rather kept in a very narrow cone of light. To intercept this invisible light beam, the intruder must be able to obtain direct access to the light beam. Despite its potential, FSO still has many hurdles to overcome before it will be deployed widely. Carriers, like Allied Riser and XO Communications, may use FSO in conjunction with other technologies to expand their current networks while others, such as Terabeam.

Because of high availability, bandwidth scalability, and deployment simplicity, Free-Space Optics has emerged as a solution and deployment option in the deployment of next-generation wireless networks. As a result of its worldwide license-free operation, combined with a multitude of applications, FSO is providing an attractive and cost-effective option for network planners concerned about fast and reliable network builds, and executives concerned about faster paths to profitability.

REFERENCES

1. Mendelson, James S. and Dorrier, Charles R. “Free Space Optics: Fixed Wireless  

     Broadband,”

 2. White, Chad. “How to Squeeze More Data Over What’s Already There”    

      Technology Investor.

 3. Smith, Brad. "Going the Last Mile" Wireless Week

 4. S. Bloom PhD, “The Physics of Free Space Optics,” available at www.freespace optics.com