One of the biggest challenges facing Free-Space Optics deployment is proper understanding of optical signal propagation in different atmospheric conditions. In an earlier study by the author (30), attenuation by rain was analyzed and successfully modeled for infrared signal transmission. In this paper, we focus on attenuation due to scattering by haze, fog and low clouds droplets using the original Mie Scattering theory. Relying on published experimental results on infrared propagation, electromagnetic waves scattering by spherical droplet, atmospheric physics and thermodynamics, UlmTech developed a computer-based platform, Simulight, which simulates infrared signal (750 nm-12 μm) propagation in haze, fog, low clouds, rain and clear weather. Optical signals are scattered by fog droplets during transmission in the forward direction preventing the receiver from detecting the minimum required power. Weather databases describe foggy conditions by measuring the visibility parameter, which is, in general, defined as the maximum distance that the visible 550 nm signal can travel while distinguishing between the target object and its background at 2% contrast. Extrapolating optical signal attenuations beyond 550 nm using only visibility is not as straightforward as stated by the Kruse equation which is unfortunately widely used. We conclude that it is essential to understand atmospheric droplet sizes and their distributions based on measured attenuations to effectively estimate infrared attenuation. We focus on three types of popular fogs: Evolving, Stable and Selective.

This paper describes the development and reports measured performance of integrated CMOS receiver and transmitter circuits for use in an optical wireless link operating at bit rates up to 310 Mb/s. The receiver presented is an angle-diversity design and consists of multiple sectors each driving an individual pre-amplifier channel. The speed limitation for the receiver circuit is determined substantially by the parasitic capacitance introduced by the photodetector. With current PIN devices this capacitance may be comparatively high, of order several picofarads as a relatively large field of view is required for optical wireless applications. The design incorporates an on-chip selector with external controls determined by the signal level. Signals from detectors that receive optical power above a certain threshold level are passed to a combiner circuit. In the transmitter, in order to avoid limiting the optical performance of the emitter, the electrical response of the LED driver is enhanced by current-peaking and charge-extraction circuitry. A novel timing generator is used to achieve fast rise and fall times. Experimental results confirm that the true performance evaluation of high-speed circuits can be severely hindered by parasitics associated with wire bonding and packaging of chips. Flip-chip packaging, advantageous for its small form factor and low capacitance leading to high speed has been investigated. This has led to the development of fully integrated receiver and transmitter systems where the photodetector and photoemitter devices are directly bonded to supporting CMOS substrates which furnish the necessary support electronics.

The widespread use of Optical LANs is dependent on the ability to fabricate low cost transceiver components. These are usually complex, and fabrication involves the integration of optoelectronic and electronic devices, as well as optical components. A consortium of four UK universities are currently involved in a project to demonstrate integrated optical wireless transceiver subsystems that can provide eye-safe line of sight in-building communication at 155Mbit/s and above. In this paper we discuss the flip-chip integration of two-dimensional arrays of novel microcavity LEDs with custom CMOS integrated circuits in order to produce solid state tracking emitters. Design, fabrication and integration of these structures are detailed. The scaleability and future capability available given further optimisation and development of these systems is also discussed.

In this work we present laser-based novel devices that maximize the emitted power for constant eye safety level and beam divergence angle, i.e., without affecting the eye safety classification or the necessary tracking accuracy. This is achieved by breaking the spatial coherence of the beam, which allows the system to be considered as an extended light source. The system comprises a laser, a diffuser, a collimator and, sometimes, other optical elements. As an example, one of the devices is composed of a laser, a Lambertian reflective-type diffuser, and a single-piece reflective-refractive collimator of 20 mm aperture and ultra-high numerical aperture (NA = 1.43), which re-collimates the radiation into 3.5 deg. (full angle). According to the IEC 60825-1:1993 (amendment 2, 2001-01), the Accessible Emission Limit (AEL) (Class 1, wavelength λ = 780 nm, exposure T = 30000 s) for this device is 35.9 times greater than that of a laser with the same divergence angle (15.6 dB), i.e., this device is allowed to emit 35.9 times more power than that of the laser alone with the same divergence angle. The switching time, the beam divergence and the eye safety classification remain the same. This power gain varies with the design conditions. In the cases analyzed it goes from = 8.4 (9.24dB) to 551.3 (27.4 dB).

Multiple-subcarrier modulation (MSM) offers several potential advantages for optical wireless systems. MSM makes it possible to perform frequency-division multiplexing, while maintaining the simplicity of intensity modulation and direct detection. Indoor optical wireless systems may be subject to multipath distortion; in such systems, MSM enables transmission at high bit rates with minimal intersymbol interference. The average-power efficiency of MSM is lower than that of baseband modulation techniques, such as on-off-keying or pulse-position modulation, however. We present a technique for improving the average-power efficiency of multiple-subcarrier systems using binary or quadrature phase-shift keying (BPSK, QPSK) with bandlimited pulses, which are, necessarily, not time-limited. In order to insure the non-negativity of the transmitted intensity signal, we use a time-varying bias signal, which is a baseband pulse-amplitude modulation signal also using these pulses. We focus on the problem of designing this time-varying bias signal, which is complicated by the fact that the pulses are not time-limited. We show that the proposed scheme can significantly enhance power efficiency, with only a minimal increase in implementation complexity.

For optical wireless networking, it is desirable to employ nondirected links whose performance depends on the reflection characteristics of the indoor surfaces. The non-Lambert reflection pattern, Phong model, is considered in the calculation of the multipath impulse response function as well as Lambert reflection pattern. The bit error rate (BER) of the optical wireless direct-sequence spread spectrum (DSSS) system using biorthogonal Walsh codes is investigated on a non-Lambert reflection channel. Simulation results demonstrate that biorthogonal DSSS systems can combat multipath dispersion with small power penalties very well. And by comparison, we show how much the error of the performance prediction for the biorthogonal DSSS system will occur when using Lamertian approximation on the non-Lambert channel.

In this paper, we present an infrared wireless indoor communication system that bases on Ethernet network. The bit rate of Ethernet is 10Mbps, but after Manchester coding, in the physical layer the actual bit rate is 20Mbps. In our designs, the transmitter uses laser diodes (LDs). The transmitter consists of differential input circuit, LD driver circuit. The receiver consists of a coated truncated spherical concentrator whose field of view (FOV) is 40 degree, a large area Si PIN photo-detector followed by transimpedance amplifier, second-stage amplifier, low-pass filter (LPF), high-pass filter (HPF), limiting amplifier and differential output circuit. The network is constructed as a base-terminals configuration and two transit wavelengths are used for base and terminals respectively to avoid collision. Experimental testing was conducted in a room with size 5m × 5m × 3m and the network could work well.

In the wireless infrared communications it is necessary to place a non-imaging optic concentrator in front of the planar PIN photo-detector in order to achieve more optic energy that can result in greater electrical SNR ratio. Incidence rays, which have different directions relative to the axis of concentrator, will produce different optic gain, and then it changes the channel impulse response compared with the one which is made in a simple planar PIN detector system. In addition, the field of view (FOV) will dramatically affect the system performance. Decreasing the FOV of concentrator will reduce the multipath-induced intersymbol interference (ISI), but at the same time it also reduces the total optic energy that is received. This paper demonstrates the effect of concentrator on impulse response of IR wireless indoor channel and compares the differences of the impulse responses of diffuse channel, basing on whether a hemisphere concentrator or a truncated spherical concentrator has been used. The paper also presents the relationships between FOV and system bandwidth and between FOV and average optical energy. A conclusion of optimal result is drawn.

Under favorable visibility conditions, scintillation becomes the limiting factor in estimating free space optical (FSO) link availability. In the summer of 2001, Terabeam performed several experiments to characterize the impact of scintillation on FSO products under development. In the experiment, a transmitter and a receiver were placed 1000 and 2600 meters apart and operated with the receiver at several different sized apertures and under diverse atmospheric conditions. The experimental probability of fade is then calculated and compared with two theoretical models, namely the lognormal and the gamma-gamma.

Reliability is the main challenge facing Free Space Optical Communications today. Fog plays a key role here, but so does range and optical power throughput. Presently, the FSOC beam has a divergence in the milliradian range in order to compensate for beam wander caused by platform drift and vibrations and atmospheric index of refraction fluctuations. This large beam divergence limits the power throughput and the maximum range for communication. A beam that is collimated to the microradian level would greatly improve the power throughput (by three to four orders of magnitude) and thus, greatly improve the range, but this would also require a fast (> 1 kHz), accurate (microradiam) beam pointing system. Stress-Optics technology can provide that system. In Stress-Optics a stress field can be imposed on almost any optical material to modulate the index of refraction within that material in a predetermined manner. In general, stress can bring about a greater change in the index than can other fields that might be imposed. Stress-Optics accomplishes uniform beam steering in two dimensions to moderate angles, as well as beam shaping, including spherical and cylindrical lensing. The method is simple, solid-state, fast (< 5 mseconds), precise (< 1 microradian), inexpensive and durable. This patented Stress-Optic technique, when operated in conjunction with CMOS imaging feedback from a reference beam, can compensate for beam wander and defocusing from both platform and atmospheric sources. The stabilization is to microradian accuracy with kHz response times and milliradian 2-D collimation control.

Free-space optical (FSO) links for high-speed communications between buildings must consider detrimental environmental effects including interference from sunlight in the receiver's instantaneous field of view (IFOV). Sunlight can degrade receive sensitivity resulting in link disruptions, even with significant optical filtering. Thus it is important to characterize this environmental effect for designing and testing optical transceivers. Background light levels are highly dependent on the geometry and environmental conditions of a specific link making general statements difficult. However, we have characterized the likelihood and frequency of direct or reflected sunlight passing into or near a terminal's IFOV. We have also measured detector solar power levels under sunny and partly cloudy conditions, and measured detector sensitivity degradation as a function of background light levels. This paper presents a summary of our results.

Significant product testing is required to develop a truly "carrier-grade" free-space optical transceiver. In addition to engineering a system with high availability and an appropriate feature set, it is necessary to ensure reliable operation across a wide range of operating conditions, including temperature extremes, inclement weather (e.g. rain, snow, wind), and vibration. It is also important to guarantee consistent performance between units, in order to enable efficient manufacturing processes and improve the customer experience. This paper describes the engineering confidence and design acceptance testing process and results for Terabeam's next-generation access transceiver. These tests include bench performance testing, thermal and other environmental tests, and performance testing at range.

We consider the acquisition process in short-range (1~10 km) free-space optical communication between moving parties when covertness is the overriding system performance requirement. In order to maximize covertness, it is critical to minimize the time required for the acquisition phase, during which the party initiating contact must conduct a broad-field scan, and risks revealing his position. Assuming an elliptical Gaussian beam profile, we show how to optimize the beam divergence angles, scan speed and design of the raster scan pattern so as to minimize acquisition time. In this optimization, several constraints are considered, including: SNR required for accurate bearing detection and reliable decoding, limited receiver bandwidth, limited scanner speed, and beam divergence as limited by the scanner mirror dimensions. Design examples are given to illustrate the design procedure and by these examples, we show that the optimum beam profile is often elliptical with high eccentricity, though the search field is circularly symmetric.

In free-space optical (FSO) communications, conditions may be met when laser links suffer from solar background radiation (SBR). There are four types of such conditions Direct sunlight hitting a photodetector Reflected sunlight (glints) Sunlight scattered by hydrometeors Sunlight scattered by surrounding objects (walls, etc.) Direct sunlight may cause total break of communications (link outage), and thus affect the link availability. However, experiments prove that the sunlight does not cause irreversible degradation of semiconductor photodetectors used in FSO systems. Estimations are made of the link outage periods duration for various types of SBR conditions, also other effects caused by SBR have been considered. Recommendations are presented for the link directivity optimization to avoid (or to minimize the probability of) communication interrupts caused by SBR. A nomographic chart has been developed to forecast periods of time when direct or scattered solar radiation may cause link outage. With this chart, a user in any point of the globe, knowing the link orientation (azimuth and elevation angles), can see when and for how long (if at all) may the link operation be affected by unfavorable SBR conditions, also in many cases it is possible to recommend insignificant modifications in the link orientation causing material improvement in FSO system performance.

AirFiber's Hybrid Free Space Optics Radio (HFR) provides greater than 1 km range at 99.999% availability. The HFR does this by hitlessly combining the data streams from two data paths, one a free space optic (FSO), the other a 60 GHz millimeter wave system. FSO systems alone are limited in range to less than 500 meters in most cities in the world due to fog outages yet have modest attenuation in rain. Fog is transparent to 60 GHz systems yet they suffer from rain and clear air oxygen absorption. The HFR system uses the strengths of each of these very high bandwidth (up to 1.25 Gb/sec), unlicensed technologies, to provide 99.999% wireless availability at ranges greater than 1 km in all weather conditions. We present some range curves and data from the output of an AirFiber HFR system during a severe (300 dB/km) fog event of many hours duration.

Free-Space Optics (FSO) is a proven, reliable technology for last mile telecommunications applications, used worldwide for both enterprise network building-to-building connections and for wireless access to more traditional land line communications networks. In most mid-latitude coastal cities, link availability at distances above a few hundred meters is primarily affected by fog and low clouds. At longer distances, heavy rain and snow can also affect the link. The most mature technology used in FSO equipment relies on low cost semiconductor lasers or LED’s operating in the near infrared at wavelengths of 785 nm or 850 nm. In the past few years, systems operating at 1550 nm have also been developed. At first the vendors of these systems claimed that the 1550 nm wavelength had better propagation characteristics in severe weather than the 785 nm wavelength. With further analysis and research, those claims were withdrawn. Now there are claims that even longer wavelengths near 10 microns will solve the FSO link availability issues associated with severe weather. Hype about such magic wavelengths for FSO is both a disservice to the investors who will lose the money they are investing based on exaggerated claims, and to the rest of the FSO industry which should be creating realistic expectations for the capability of its equipment. In the weather conditions which normally cause the highest attenuation for FSO systems, namely coastal fog and low clouds, 10 microns offers no propagation advantage over shorter wavelengths.