Laser Satellite Communication Network— Vibration Effect and Possible Solutions SHLOMI ARNON AND N. S. KOPEIKA, SENIOR MEMBER, IEEE A number of serious consortiums develop satellite communication networks. The objective of these communication projects is to service personal communication users almost everywhere on earth. The intersatellite links in those projects use microwave radiation as the carrier. Free-space optical communication between satellites networked together can make possible high-speed communication between different places on earth. Some advantages of an optical communication system over a microwave communication system in free space are 1) smaller size and weight, 2) less transmitter power, 3) larger bandwidth, and 4) higher immunity to interference. The pointing from one satellite to another is a complicated problem due to the large distance between the satellite, the narrow beam divergence angle, and vibration of the pointing system. Such vibration of the transmitted beam in the receiver plane decreases the average received signal, which increases the bit error rate. In this paper, we review 1) the present status of satellite networks, 2) developing efforts of optical satellite communication around the world, 3) performance results of vibration effects on different kinds of optical communication satellite networks, and 4) seven approaches to overcome the problems caused by transmitter pointing vibration. Keywords— Laser communication, optical networks, satellite optical communication, vibrations.
I. INTRODUCTION Communication from any place to another on earth is an attractive goal. One method to achieve this aim is by networking satellites together to cover the globe (Fig. 1). In this method, the information is transferred from the ground to the nearest satellite above and then propagates between the satellites to the satellite above the destination. This last satellite then transmits the information down to the destination. The idea of a satellite communication network is no longer science fiction. Today, a number of serious consortiums develop satellite communication networks. The objective of these communication projects is to service personal communication users almost everywhere on earth. The intersatellite links (ISL’s) in those projects use microwave radiation as the carrier. The use of optical ISL’s has some advantages over the use of microwave ISL’s: 1) smaller size and weight of the terminal, 2) less transmitter power, 3) higher immunity to interference, 4) larger data Manuscript received May 28, 1997; revised July 23, 1997. The authors are with the Department of Electrical and Computer Engineering, Ben-Gurion University of the Negev, Beer-Sheva 84105 Israel (e-mail:
[email protected];
[email protected]). Publisher Item Identifier S 0018-9219(97)07572-5.
rate, and 5) smaller transmitter beam divergence angle. The main disadvantage of optical ISL’s is the complex pointing system. The complexity of the pointing system derives from the necessity to point from one satellite to another over a distance of tens of thousands of kilometers with a beam divergence angle of microradians while the satellites move and vibrate. The pointing system compensates the motion of the satellites using the known Ephemerides data. Coupling of satellite mechanical vibration and tracking noise to the pointing system gives rise to vibration of the satellite transmitter beam in the receiver plane. Such vibrations of the transmitted beam in the receiver plane decrease the received signal, which increases the bit error rate (BER). In optical satellite networks, the problem is more complicated because all the satellites continually vibrate randomly. The concept of communication satellite networks appears in [1]–[5]. The concepts of optical ISL’s in communication satellite networks are described in [5]–[8]. Results of onboard measurements of the vibration spectra of communication satellites OLYMPUS and LANDSAT are specified in [9]–[12]. Results of measurements from the ground of satellite vibration spectra by laser radar are described in [13] and [14]. Chen and Gardner analyze the impact of random pointing and tracking errors on the design of coherent and incoherent optical intersatellite communication links. In this research, they calculate the effects of vibration on pulseposition modulation (PPM) and noncoherent frequency shift keying (NCFSK) [15]. Barry and Mecherle found relationships between root mean square (rms) standard deviation of the pointing error distribution and the burst error of the communication system [16]. More analysis of the effects of vibrations on satellite optical communication is described in [17]–[19]. References [20] and [21] describe the performance limitations of free-space optical communication satellite networks due to vibrations. Skormin et al. [22], [23] describe adaptive feed-forward techniques to reduce the effects of vibrations on the pointing system of laser communication. Marshalek and Koepf compare optical technologies for ISL’s in a global telecommunications network [24]. Lambert and Casey [25] describe most of the concepts of laser communication in space, including the effects of vibrations on the communication system.
0018–9219/97$10.00 1997 IEEE 1646
PROCEEDINGS OF THE IEEE, VOL. 85, NO. 10, OCTOBER 1997
Table 1 Satellite Communication Projects
Fig. 1.
Satellite communication network.
In this paper, we review 1) the present status of satellite networks, 2) the development efforts of satellite optical communication carried out by the United States, Japan, and the European Space Agency (ESA) and theoretical work going on in Israel, 3) performance results of vibration effects on digital and analog direct detection communication satellite networks, and 4) seven approaches (bandwidth adaptation, beamwidth adaptation, power control, coding, channel diversity, vibration isolation, and self-tuning feed forward) to overcome the problems caused by transmitter pointing vibration. II. SATELLITE COMMUNICATION NETWORKS—PRESENT STATUS Today, there are a number of serious consortiums developing satellite communication networks [1]–[5], [26]. The objective of these communication projects is to service personal communication users almost everywhere on earth. The amount of money invested worldwide in those projects is assumed to be tens of billions of U.S. dollars. The incentive for this huge investment is the expectation of a growing need for personal communication services (PCS’s) unlimited by the coverage of cellular systems. There are two trends for network development: a) using ISL’s (space switch) and b) using a ground station for each hop (ground switch). Using ISL’s almost eliminates the dependence on ground stations. This has some advantages: a) less investment in ground infrastructure; b) independence with regard to local telephone companies; c) saving of royalties for terrestrial services supplied; d) prevention of political difficulties; e) channel diversity; f) possible decrease in delay and latency. ARNON AND KOPEIKA: LASER SATELLITE NETWORK
The disadvantage of ISL’s is the complexity of the system. Some of the satellite communication projects and their characteristics are described in Table 1. Initial service for those projects is expected to be within ten years, i.e., at the end of this millennium and at the beginning of the new one. Most projects (except Odyssey) use low-earth-orbit satellites (LEO’s) to achieve short propagation delay and low-power user transmitters. In all the projects, the populous areas on earth are covered by the satellites. It is important to emphasize that these communication systems deal mostly with low data rates. It is logical to expect that demand for higher data rates will increase, as in terrestrial networks. Currently, only the IRIDIUM and TELEDESIC systems are designed with ISL’s. IRIDIUM uses radio-frequency (RF) radiation (K-band) for ISL’s. Each satellite has 23-GHz links to the satellite immediately ahead of it in its plane and to the satellite immediately behind it. In addition, each satellite has four cross links to the two nearest satellites in each of the two adjacent corotation planes. The ISL for Block 1 (first production series) transfers information between the satellites at a rate of approximately 10 Mb/s. For Block 2, the rate increases. In TELEDESIC, each satellite will communicate with two satellites ahead and two behind in the same orbit plane and sideways to adjacent terminals on the neighboring and next neighboring planes in each direction. Laser and microwave (60 GHz) cross links are evaluated for TELEDESIC. The data rates planned are 1 Gb/s or 1647
more. Some programs on the drawing board may use ISL’s. These include Motorola’s M-Star and Millennium, Hughes’ Spaceway, AT&T’s VoiceSpan, and Lockheed Martin’s Astrolink. Four important conclusions are derived from this section: a) limited RF spectrum resources are utilized by all those systems, b) the use of ISL’s has some advantages compared to ground-station hops, c) the demand for larger ISL bandwidths will increase in order to increase network capacity and profit, and d) huge amounts of money are to be invested in satellite communication networks. III. SATELLITE OPTICAL COMMUNICATION SYSTEM SCHEME In this section, we review the basic schemes of the satellite optical communication segments: the transmitter, receiver, and tracking system. A. Transmitter Scheme The transmitter (Fig. 2) model for on–off keying (OOK) includes [21] a laser transmitter, a telescope, and random attenuation (vibration effects). The messages arrive at the input of the transmitter. The transmitter converts electrical signals to optical signals using the laser. The transmitter telescope collimates the laser radiation in the receiver satellite direction. B. Receiver Scheme The receiver (Fig. 3) model for OOK [21] includes a telescope, an optical bandpass filter, input insertion losses, an optical amplifier, output insertion losses, an optical bandpass filter, a p-i-n photodiode, an electrical filter, and a decision rule circuit. The receiver telescope focuses the received radiation onto the optical filter. The optical filter prevents most of the background radiation from entering the subsequent stages of the system. The radiation propagates through the optical filter to the optical amplifier. The optical amplifier has losses in its input and output due to reflection and optical mismatch. The radiation from the optical amplifier output is an amplified version of the input radiation but with optical amplifier noise. The optical filter at the output of the optical amplifier prevents part of the amplifier noise from passing. The radiation is converted to an electrical signal by a photodiode. The electrical signal is filtered by the electrical filter. According to the electrical signal amplitude and arrival time, the decision circuit decides the kind of information received. C. Tracking System Scheme To establish optical communication between two satellites, the line of sight of their optics must be aligned during the entire time of communication. To meet this requirement, the satellites use the Ephemerides data (the position of the satellite according to the orbit equation) for rough pointing and a tracking system for fine pointing to the other satellite. The basic and popular method of tracking between satellites includes use of a beacon signal on one satellite and a quadrant detector and tracking system at the other 1648
Fig. 2. Transmitter scheme.
satellite [25]. The fine elevation and azimuth angle of the pointing system evaluates pointing direction from the output signal of the quadrant detector. In Fig. 4, we see the main components of the tracking system. The radiation from the beacon on one satellite is received by the telescope on the other satellite. The telescope focuses the received radiation onto the quadrant detector. The pointing and control unit calculates the telescope pointing direction according to the quadrant detector signal. IV. SATELLITE OPTICAL COMMUNICATION DEVELOPMENT—PRESENT STATUS The development of satellite optical communication is carried out mostly by the United States, Japan, and the ESA [5], [25], [27]–[32]. In addition, much theoretical work on satellite optical communication has been going on in Israel [20], [21], [33]–[44]. The United States supports programs at a number of agencies, such as the Ballistic Missile Defense Organization (BMDO), the National AeroPROCEEDINGS OF THE IEEE, VOL. 85, NO. 10, OCTOBER 1997
Fig. 4. Tracking-system scheme.
Fig. 3.
Receiver scheme.
nautics and Space Administration/Jet Propulsion Laboratory (NASA/JPL), the Massachusetts Institute of Technology (MIT) Lincoln Laboratory, etc. The BMDO has been funding the development of laser communication technology leading up to a space-to-ground demonstration to be flown on the second space technology research vehicle (STRV-2), which is projected to be launched in 1998. The goal of STRV-2 is to send and receive data at rates between 155 Mb/s and 1.24 Gb/s using a ground-based laser communication terminal. NASA/JPL develops the optical communications demonstrator (OCD). This program includes the development of an engineering model of flight terminal capability to communicate kilobits/second to megabits/second from the planets or megabits/second to gigabits/second from a highearth-orbit satellite to the ground. The OCD currently is undergoing system-level testing and performance evaluation. Previous U.S.-supported laser communication projects include a laser cross-link subsystem, a follow-on early warning system, a laser intersatellite transmission experiARNON AND KOPEIKA: LASER SATELLITE NETWORK
ment, and a laser communication demonstration system. To the best knowledge of the authors, these programs did not include actual launch of communication satellites. Some of these projects included development of a space qualified communication segment with a data rate as high as 1 Gb/s with a BER less than 10 . The United States is also supporting programs for developing technologies such as a master oscillator power amplifier, nonmechanical beamsteering elements, and high-data-rate codec chips for space communication at JPL, MIT Lincoln Laboratory, and a number of leading companies. The European program is known as the semiconductor intersatellite laser experiment (SILEX). This program is developing laser terminals at a rate of 50 Mb/s for LEOLEO and LEO-geosynchronous earth orbit (GEO) ISL’s. This program includes transmitting data from the French LEO SPOT 4 to the GEO advanced relay and technology mission ARTEMIS. A GEO-to-ground communication also is under consideration. The first host spacecraft (SPOT 4) is planned to be launched in February 1998. The launch of ARTEMIS is scheduled for 2000. The demonstration is scheduled for the middle of 2000. The Japanese space agency is developing two programs, the laser communication experiment (LCE) and the optical interorbit communication test satellite (OICETS). The LCE program includes launch of ETS VI satellites in 1994 and communication experiments from ground to satellite and back from Japan and the United States. The OICETS includes intersatellite communication links cooperative with the ESA SILEX program. The forward-link data rate is 2 Mb/s and the return-link data rate is 50 Mb/s. The launch date has been postponed to sometime in the year 2000. V. COMPARISON BETWEEN MICROWAVE AND OPTICAL ISL’S Comparisons between the microwave and laser technologies involve many considerations [5]–[7], [25]. In Table 2, we present some of the possible criteria for satellite communication segments. The important factors for satellite communication are mass, size, data rate, and power consumption. The optical link is better than the microwave link according to such factors. The main disadvantage of optical links is the lack of knowledge, subsystems, and experience from previous projects (technologies without history). Some researchers made quantitative comparisons between RF (23 and 60 GHz) and optical links for various satellite communication scenarios at data rates of 25 and 360 Mb/s. The results of the comparison for present technology are 1649
Table 2 Qualitative Comparison Between Optical and Microwave Links
Table 3
External Vibration Sources
that the optical payload is found to be at least 50% lighter and the prime power and size are functions of the data rate and distance between the satellites. For data rates of 100 Mb/s and above, laser cross links are generally considered to be lighter, smaller, and of lower power consumption than RF links.
Table 4
Internal Vibration Sources
VI. SATELLITE MECHANICAL IMPACTS AND VIBRATIONS In this section, we review most of the vibration and impact sources in satellite technologies. Due to these mechanical impacts, the satellite transmitter beam to the receiver satellite vibrates and the communication system performance is degraded. Measurement of satellite vibrations has been performed onboard the OLYMPUS and LANDSAT satellites [9]–[12]. The review is divided into two parts. The first describes external sources and the second describes the internal source of vibrations and impact. Such vibrations and impact propagate to subsystems onboard satellites and may cause disturbances for satellite subsystems, such as laser communication segments. A. External Sources Major external sources are listed [22] in Table 3. We present here some additional details. Satellites can be destroyed by colliding with meteorites. If the satellite collides with micrometeorites, the collision can cause vibrations and impact in the satellite structure. When a satellite circles the earth, the radiation levels on it change according to its relative position with regard to celestial bodies such as the earth, moon, and sun. These changes of radiation level cause gradients of temperature on the satellite structure, which cause structure deformations. Due to cycling of the satellite movement, elastic forces of tension and bending are created. Another source of vibrations are inhomogeneities of gravitational force through the satellite orbit (such as 1650
solar and lunar gravity, earth oblateness effects, the earth’s central gravitational field, and ellipticity of orbit). B. Internal Sources Some internal sources cause vibrations in laser communication. Their range and duration are listed [9]–[11] in Table 4. These results were measured on the OLYMPUS communication satellite by the ESA and are presented here only as an example. It is expected that in future satellites, the characteristics of vibration sources may change due to changes in design requirements. From this table, it is easy to understand what vibrations cause for each operation of mechanical systems. The strength and duration of the vibration are determined by characteristics of the subsystem. In practical LEO satellites, the solar array drive mechanism and antenna-pointing mechanism adapt their directions in order to optimize their performance. Consequently, vibrations are expected over most of their operation times. In Fig. 5, we see a plot of a simulation of vibration power density function for a laser communication system as a function of the mechanical frequency [25]. The PROCEEDINGS OF THE IEEE, VOL. 85, NO. 10, OCTOBER 1997
Fig. 5.
Typical vibration amplitude as a function of mechanical frequency (similar to [25]).
important points in this figure are the vibration limits at very low frequencies and the number of peaks included in the spectra. These peaks can relate to the operation of subsystems onboard the satellite, such as gimbal rate control jitter and gimbal position control jitter. The peak jitter can be filtered using appropriate stop-band filters. Summarizing this section, the vibration sources are various and are caused by internal and external mechanisms. Some of these vibrations and impacts may propagate into satellite subsystems, such as the laser transceiver, and cause disturbances in their normal operation modes.
VII. TRACKING NOISE The basic and popular method of tracking between satellites includes the use of a beacon signal on one satellite and a quadrant detector and tracking system at the other satellite [15], [25]. The fine elevation and azimuth angle of the pointing system is based on the output signal of the quadrant detector. The tracking system is very similar to electrooptic communication systems. Therefore, the tracking system suffers from the same various noise sources, such as laser relative intensity noise, Johnson (thermal) noise, dark current shot noise, signal shot noise, and background shot noise. The signal from the tracking system enters the control system. The control system points the transceiver to the other satellite. Noises from the control system are added to the pointing signal. All of these noise sources cause vibration of the pointing direction. Due to vibrations of the transmitter beam to the receiver satellite, the communication system performance is degraded. The simplest expression for tracking noise (standard deviation) ARNON AND KOPEIKA: LASER SATELLITE NETWORK
is given as [25] SF SNR
(1)
where SF is the angular slope factor of the angle-to-voltage transfer function expressed in units per radian and SNR is the signal-to-noise ratio in the tracking system bandwidth. The value also has been referred to as the noise equivalent angle [25] (NEA) of the tracking system. VIII. THE VIBRATION STATISTICS MODEL Due to noise in the tracking system and mechanical vibrations, described in the previous two sections, the satellite transmitter beam to the receiver satellite vibrates. This degrades the communication system performance. The statistical model of the transmitter vibrations derives from the sources of the vibrations. The two simple models popularly used as statistical models are Rayleigh and normal probability density distribution functions (pdf’s). The Rayleigh model is used mostly when the tracking noise is the dominant noise. To use this model, we assume that the SNR of the tracking system is large and that the azimuth and elevation tracking processes are independent and identically distributed so that the radial pointing error angle model is Rayleigh distributed with pdf [15] (2) where is the radial pointing error angle and is the radial standard deviation or NEA. The normal model is used mostly when we analyze systems with mechanical impacts. Most of the mechanical impacts are caused by internal satellite subsystems 1651
such as waveguide switches, solar array drive mechanisms, thruster firings, or antenna-pointing mechanisms controlled by satellite computer. The transmitter beam-pointing error is modeled by a normal distribution with pdf [33] (3) IX. RANGE EQUATION MODEL In this section, we derive a model that relates the received to the transmitted optical power as a function of the system parameters. The distance between the transmitter and the receiver satellite is meters. The instantaneous received power as a function of radial pointing-direction error angle is [15] (4) where the pointing loss factor is [15] (5) and (6) where and are the optical efficiencies and the antenna gain of the receiver and the transmitter, respectively. The wavelength of the laser transmitter is , and is the optical laser transmitter power. X. EFFECT OF VIBRATIONS ON PERFORMANCE OF SATELLITE OPTICAL COMMUNICATION SYSTEMS A. Performance Calculation In this paragraph, we present the basic analysis of vibration effects on the performance of communication systems. SNR and BER are the most acceptable performance parameters of communication systems. The average SNR of a communication system is [20] SNR
(7)
where is the total noise variance for transmitting a “1” as a function of the received optical power ( ), the radial pointing error , and receiver parameters – . The received signal for transmitting a “1” is . The average BER of communication systems is [20] SNR BER
(8) where BER( ) is the function that calculates the bit error probability. Each modulation format has its own function. 1652
Fig. 6. Normalized optical signal as a function of vibration amplitude and number of satellites (after [20]).
Here, is the total noise variance for transmitting a “0” and is the received signal for transmitting a “0.” We consider first analog and then digital communication. B. Analog Communication Network This communication system transmits and receives analog signals [20]. Pulse detection, if it is needed, is performed in the last link of the system. The network includes a transmitter satellite above the information source, repeater satellites, and a receiver satellite above the destination. The information is received in a transmitter satellite in the form of an RF signal from the information source. This information is then up-converted to an optical signal and transmitted to the neighboring repeater satellite. The information is passed between the repeaters to the receiver satellite. At the receiver satellite, the information is downconverted to RF and transmitted to its destination. In the following figures, we see the effects of vibrations and network size on network performance. In Fig. 6, we see the received optical signal as a function of vibration amplitude normalized to the square root of the transmitter gain and the number of satellites in the network. From this figure, it is seen that an increase of vibration amplitude dramatically decreases the received signal as the network dimension increases. Fig. 7 describes the normalized electronic noise as a function of vibration amplitude and the number of satellites in the network [20]. In this figure, it is seen that for low vibration amplitude, the noise increases when the network increases, but for high vibration amplitude, the noise converges quickly to a constant value even if the network dimension increases. Fig. 8 describes BER as a function of the vibration amplitude and the number of satellites in the network [20]. BER without any vibration is 0.5*erfc(10). From this figure, it is seen that increasing vibration amplitude has a very detrimental effect on BER. This analysis points out that even low values of vibration amplitudes in each satellite can degrade dramatically the performance of the entire network. PROCEEDINGS OF THE IEEE, VOL. 85, NO. 10, OCTOBER 1997
Fig. 7. Normalized noise as a function of vibration amplitude and number of satellites (after [20]).
Fig. 8. BER for OOK as a function of vibration amplitude and number of satellites. The value of BER without vibrations is close to 100300 (after [20]).
C. Direct-Detection Digital Communication Networks The network includes an above-ground transmitter satellite or airborne transmitter, regenerative satellites, and an above-ground receiver satellite or airborne receiver [21]. The information is received in a transmitter satellite from an RF ground or air station. This information is then upconverted to an optical signal and transmitted to the neighboring regenerative satellite. The information is passed between the regenerative satellites to the receiver satellite. At the receiver satellite, the information is down-converted to RF and transmitted to the ground or air station. The regenerative system includes a receiver and a transmitter. The optical signal is received by a regenerative satellite and is then converted to an electronic signal. The information is detected and decoded from the electronic signal. The received information is encoded and modulates on laser radiation and then transmits. Here, we present results of calculations for LEO satellite communication networks. In this analysis, we compare three modulation methods: OOK, PPM, and pulse polarization binary modulation (PPBM). ARNON AND KOPEIKA: LASER SATELLITE NETWORK
Fig. 9. SNR as a function of vibration amplitude for practical satellite communication systems. The OOK system is marked by a dashed curve, the PPM system by a dash-dot curve, and the PPBM system by a solid curve (after [21]).
All of the parameters of the comparison are described in [21]. The SNR of OOK, PPM, and PPBM changes from 3040, 6460, and 1520 to about 1440, 3200, and 720, respectively for a change in normalized vibration variance from 0.05 to 0.5 (Fig. 9). From Fig. 9, it is seen that the PPM method yields the highest SNR, OOK yields medium SNR, and PPBM yields the poorest SNR. These results can be qualitatively expected for the three schemes. In the PPBM scheme, for every single bit we transmitted, the bit power is low in order to keep the average transmitted power constant. In OOK, a signal is transmitted for only half of the bits, so that the bit power is higher than for PPBM. In PPM, we transmit very-short-duration high-power pulses. As a result, this scheme should be expected to exhibit the best SNR. Fig. 10 describes the BER of OOK as a function of normalized vibration variance and the number of satellites in the network. In this figure, it is seen that the BER increases as the vibration variance and the number of satellites in the network increase. From this figure, we can also see that for all three modulation schemes, the vibration amplitude is the dominant performance parameter while the size of the network is a minor performance parameter. The conclusion from this figure is that the performance of the network is determined more by a satellite that vibrates than by other satellites in the network. The BER of OOK, PPM, and PPBM changes from an infinitesimal value to about 10 , 1.2 10 , and 4 10 , respectively for 3.9 a change in normalized vibration variance from 0.05 to 0.5 (Fig. 11). From Fig. 11, it is seen that the PPM method exhibits the highest performance, OOK exhibits medium performance, and PPBM exhibits the poorest performance. These results can be explained on the basis of power criteria. The highest transmitted signal power is for PPM and the medium transmitted power is for OOK. In the PPBM scheme, we are always transmitting in one of the two polarizations, so the bit transmitted power is the lowest one. 1653
Fig. 10. OOK BER as a function of vibration amplitude and number of satellites in the network (after [21]).
amplitude vibration and impact are filtered by the standard stabilization system. The guideline in this review is to minimize the complexity and price of satellite networks. For example, we can almost eliminate vibration effects by the use of a very-high-power laser and wide transmitter divergence angle. This simple solution requires a lot of energy and large volume and size and creates problems of heat transfer. All these increase mission price and complexity. In practical system design, some or all of the seven methods we review can be used simultaneously. It is important to remember that some of the vibration sources are discrete processes with low probability so that most of the time, the vibration amplitudes are low. This makes it possible to simplify design. When vibration events occur, the communication system adapts its parameters to the vibration level using one or more of these methods. Because the high amplitude vibrations are short with low probability, the overall performance may hardly be affected. A. Bandwidth Adaptation The basic parameter that affects the performance of any communication system is the SNR. As a result of vibrations, the signal decrease causes the SNR to decrease and the BER to increase. There are two solutions to the problem of a decrease of received signal: increase the transmitted power or decrease the receiver noise. The disadvantages of increasing the transmitted power are: a) the energy consumption is higher; b) the weight and size of the satellite increase due to the need for a larger laser transmitter; c) the price and complexity of the satellite are greater; d) the problem of heat dissipation is more complex.
Fig. 11. BER for practical satellite communication systems. The OOK system is marked by a dashed curve, the PPM system by a dash-dot curve, and the PPBM system by a solid curve. Note that the OOK and PPBM curves coincide (after [21]).
This analysis points out that even low values of vibration amplitudes in each satellite can dramatically decrease the performance of the network. The vibration amplitude is the dominant network performance parameter, and the size of the network is a minor network performance parameter. One important conclusion is that the performance of the network is determined by the satellite that vibrates more than other satellites in the network. Having demonstrated the severity of the problem, we now consider possible solutions. XI. POSSIBLE SOLUTIONS FOR DECREASING THE EFFECT OF VIBRATION ON THE PERFORMANCE OF OPTICAL COMMUNICATION NETWORKS In this section, we consider seven methods to decrease the effects of satellite vibrations on satellite optical communication performance. These methods are used to overcome the low amplitude vibration that exists in the system. High 1654
The basic idea of this solution is to adapt the bandwidth and the receiver parameters to the change of the received power due to transmitter vibrations so as to cause a decrease in noise power [33]. Two effects help to achieve such a solution: 1) most of the high amplitude vibrations are caused by satellite internal mechanisms controlled by the satellite computer, and the effects of the vibrations on communication are known from mathematical analysis, simulation, and measurement, and 2) the duration of the high-amplitude vibrations is short. The first effect is under our control. The second one ensures that the overall performance of the communication system due to the adaptation of the bandwidth does not dramatically decrease. The adaptation procedure for the communication system bandwidth is as follows. a) The transmitter satellite computer operates one of the satellite subsystems. b) The computer sends an interruption warning to the communication system; the warning includes the characteristics of the vibration and the start time. c) The transmitter satellite computer calculates the required bandwidth. PROCEEDINGS OF THE IEEE, VOL. 85, NO. 10, OCTOBER 1997
Fig. 13.
Fig. 12.
Adaptive bandwidth transmitter (after [33]).
d) The bandwidth and the begin and end times are sent to the receiver. e) The receiver and the transmitter synchronously adapt the bandwidth and the receiver parameters to the vibration characteristics. f) The computer satellite operates the subsystem. g) The receiver and the transmitter synchronously change the bandwidth and the receiver parameters to normal operation when the vibration is damped. This model is useful for communication systems with two or more priorities for real time, for example, telephone calls and electronic mail. This means that when the bandwidth shrinks, electronic mail messages are delayed but the phone calls can continue. The scheme of the adaptive communication system is shown with block diagrams. The communication system description includes two block diagrams: the transmitter (Fig. 12) and the receiver (Fig. 13). The transmitter includes the communication controller, satellite computer, and laser transmitter. The inputs of the communication controller are the amplitude of the vibrations, measured with a sensor such as an accelerometer, and the communication data. The outputs of the communication controller are the control bus that controls the transmitted bit rate and the data bus that includes the data and the receiver control signal. The input to the satellite computer is the measurement of the accelerometer that measures vibration amplitude and builds a vibration data bank for future adaptation. The outputs of the satellite computer are the control signal to the satellite subsystem and the values of the vibration amplitudes. The values of the vibration displacements are based on the measurements of the accelerometer and knowledge of vibration amplitude created by the satellite subsystem. The inputs to the laser transmitter are the control bus and the data bus of the communication control. The output of the laser transmitter is ARNON AND KOPEIKA: LASER SATELLITE NETWORK
Adaptive bandwidth receiver (after [33]).
data modulated in the optics signal. The receiver includes an optical detector, a tuned low-pass filter, a digital detection unit, and a satellite computer. The inputs to the optical detector are the optics signal and control signals, which adapt the preamplifier resistance to the amplitude of the vibration signal. The output of the detector is the data bus, which is the optical signal converted to an electrical signal. The inputs to the tuned low-pass filter are the outputs of the detector and the control signal. The control signal adapts the bandwidth of the filter to the transmitted bit rate. The output signal is the filtered electronic signal. The input to the digital detection unit is the filter output. The output of the digital detection unit is the data and the vibration amplitudes (control bus). The input to the satellite computer is the vibration amplitude from the digital detection unit. The outputs of the satellite computer are the control signal of the preamplifier resistance and the filter bandwidth. In the next paragraph, we present the results of a comparison of a practical communication system with (adaptive system) and without (standard system) adaptation of bandwidth. The BER of the standard system changes from less than 10 to about 3 10 for a change in vibration amplitude from 0.2 to 0.8 rd (Fig. 14). For the same vibration amplitudes, the adaptive system BER changes from less than 10 to 10 . At the value of BER equal to 10 , the bandwidth begins to decrease in order to keep the BER constant while the received signal decreases. In Fig. 15, we can see that the information bandwidth of the adaptive system stays constant at 1 GHz until the vibration amplitude reaches close to 0.5 rd. At this point, the bandwidth decreases slowly to 0.5 MHz in order to keep the BER constant for high-amplitude vibrations. This analysis points out that an adaptive system can be a very good solution for short-period vibrations of optical communication satellites. B. Beamwidth Adaptation It is important in satellite optical communication to dissipate minimum power and to obtain minimum BER. 1655
Fig. 14. Adaptive (dashed line) and standard (solid line) system BER’s as functions of vibration amplitude (after [33]).
Fig. 15.
Adaptive system bandwidth as a function of vibration amplitude (after [33]).
This aim can be achieved with very small transmitter divergence angles to assure maximum received power. The disadvantage of too narrow a divergence angle in a simplistic manner is that the transmitter beam may sometimes miss the receiver satellite due to pointing vibrations. Also, for small divergence angles, the transmitter optics aperture is large and expensive. The optimum value of 1656
the received power as a function of the pointing vibration displacement determines the optimum beam divergence angle, i.e., the transmitter gain. The implementation of this adaptive model in satellite optical communication includes two subsystems: a vibration amplitude measurement unit and an adaptive variable telescope gain. If the vibration amplitude measurement unit senses change in vibration PROCEEDINGS OF THE IEEE, VOL. 85, NO. 10, OCTOBER 1997
amplitude, it adapts the telescope gain to optimum values in order to optimize the performance of the communication system for the new vibration level [15], [25], [34]. One of the promising methods to change the gain of the transmitter is the use of phased array techniques [45]–[47]. A phased array telescope is composed of several radiating elements. By feeding the radiant elements proper phase and amplitude differences, the array radiation pattern can be shaped as needed for minimum BER. For the past several years, optical phased arrays have been investigated for implementation in free-space optical communication. The main application intended for the optical phased array is to replace the fast steering mirror in the transmitter satellite. Inagaki and Karasawa [45] derive “a fiber type optical phased array” suitable for two-dimensional beam steering and optical fiber amplifier utilization. In this section, we present the use of an optical phased array to adapt the transmitter satellite gain to vibrations in the pointing direction. The importance of this scheme for satellite design is that phased array structures are small and light. We present a simple model of an optical phased array transmitter with radiant elements that select either of two states: “on” or “off.” Based on this concept, we can change the transmitter gain discretely in a simple manner. This simple model considers a phased array aperture as being equivalent to the sum of the active element apertures. In Fig. 16, we see the results of analysis of change of transmitter telescope gain as a function of the vibration. In this analysis, the transmitter power is adapted to the vibrations simultaneously to the transmitter gain in order to optimize system performance. This technique was first proposed by Chan and Gardner [15]. These results are calculated for a practical satellite communication system. In Fig. 16, we see the transmitter telescope gain as a function of vibration amplitude. Optimum transmitter telescope gain changes from about 6 10 to 2.95 10 for a change of from 1.6 10 to 3.2 10 . We also see four kinds of telescopes: 1) continuous gain, 2) one discrete level, 3) three discrete levels, and 4) six discrete levels [34]. It is seen that as we increase the number of elements, the discrete telescope gain converges to the continuous one. C. Power Control Satellite communication engineering deals with the problem of dissipating minimum power while obtaining the desired BER. For time-domain stationarity, the required power for a given BER can be calculated from the range equation. When the transmitter satellite direction vibrates, however, the instantaneous received power scintillates. One solution to this problem is to supply enough link power margin so that even if the received power scintillates, the BER matches the requirement. This solution has a number of disadvantages: a) high average energy consumption, b) higher dynamic range of the detector, and c) heat transfer problems. The advantage of this solution is its simplicity. A different solution is based on the procedure to adapt the transmitter power to the vibration amplitude [15], [34]. The ARNON AND KOPEIKA: LASER SATELLITE NETWORK
Fig. 16. Transmitter telescope gain as a function of 2 . The discrete level phased arrays for one, three, and six levels are marked by a solid line. Continuous gain is marked by a dotted line (after [34]).
Fig. 17.
Transmitter with power control.
logic behind this solution is to save energy due to the fact that most of the time the vibration level is low, so that the increase of the power is neglected. In Fig. 17, we can see a transmitter with power control. The communication system includes the power controller, laser transmitter, and data from the satellite computer. The input to the satellite computer includes measurements with accelerometers of vibration amplitude. They build a vibration data bank for present and future adaptations. The outputs of the satellite computer are the control signal to the satellite subsystem and the value of the vibration amplitudes to the power controller. The values of vibration amplitudes are based on the measurements of the accelerometer and knowledge of vibration amplitude created by the satellite subsystem controlled by the computer. The output of the power controller is the power signal that controls the laser transmitter power. The inputs to the laser transmitter are the power control signal and the input data. The output of the laser transmitter is data modulated in the optics signal. This transmitter transmits low power in its normal mode. When some kind of vibration occurs, the accelerometer measures 1657
its value and transfers the information to the computer for processing. The computer transfers the information to the power controller that adapts the transmitter power to the value required in order to keep the BER low. Another situation is created when the computer intends to operate one of the subsystems that cause high values of vibration. Before the operation, the computer informs the power controller of the expected operation. The power controller adapts the laser transmitter to this situation.
D. Coding Technique The information theory developed by Shannon shows that by adding sufficient redundancy to a message, the error probability can be reduced to any desirable small value. This condition is achievable only for information rates smaller than the channel capacity. Different methods of coding are derived in order to add redundancy to the message [48]–[50]. Two fundamental families of coding are the well-known block code and the convolution code. Error detecting and correction is possible using the redundancy information. Automatic repeat requests (ARQ’s) occur when errors are detected by the decoder in the receiver. This means that the receiver requests the transmitter to retransmit the last message again. Forward error correction (FEC) is a mechanism that corrects the limit number of error bits in the message. ARQ and FEC improve the BER of communication systems. The basic assumption in the mathematical development of this code is a channel with additive white Gaussian noise. This assumption means that the errors caused by the channel are statistically independent. Vibrations of the laser transmitter in satellite optical communication cause some time-burst (successive and statistically dependent) errors. For a channel such as those used for satellite optical communication, we must improve the coding scheme. One way is to construct codes that are capable of correcting burst errors, like Reed–Solomon codes [50]. In satellite optical communication, the vibration may be of millisecond duration, so for information rates of gigabits/second, megaerror bits are created. To correct megaburst errors, the code length is enormous and impractical. Another method to deal with the effects of burst errors is to interleave the information. In this method, the information is interleaved in such a way that the burst error is scattered on a number of code words so that the error per code word becomes short and independent. Paul et al. [6], [7] derive that for some practical communication systems where the high data rate stream is formed by multiplexing several low data rate channels, the preferred approach is to apply FEC to the individual channels using convolution coding and Viterbi decoding algorithms and to implement interleaving for burst-error correction. It is important to remember that coding techniques may yield lower performance compared to uncoded ones. This can happen when the SNR decreases below the code threshold. Using coding techniques in such situations does not yield any advantage. 1658
Fig. 18.
Communication network diversity example.
E. Channel Diversity Communication diversity means using a number of independent propagation paths for transmitting the same information. This method is used when the quality of the channel changes drastically in a random manner and we can use a number of paths at the same time. Using a number of channels increases the probability that at least one channel will be of sufficient quality in each time slot. The important point is the requirement for independence of the channel quality random process. Satellite communication networks are like meshes around the globe, so that every two points (satellites) can connect via many different routes. Therefore, if one route is strikebound due to vibration of one satellite in the path, the information can propagate through other paths to the destination. In Fig. 18, we present a simple scheme of a communication system with three parallel channels. For our discussion, each channel includes one or more satellites. For a system with only one channel, the probability that messages pass through the channel is . The probability that messages do not pass is . It is assumed that messages do not pass through the channel due to vibration of one of the channel satellites. The probability that a message will pass through a system with three parallel channels, as in Fig. 18, is . The probability of the message’s passing improves dramatically, for example, if , in which case and . F. Vibration Isolator The vibration isolator is a system that reduces the transmission of vibrations from the spacecraft body to the communication system. Two basic kinds of isolators are PROCEEDINGS OF THE IEEE, VOL. 85, NO. 10, OCTOBER 1997
used: passive and active [51]. The passive isolator includes a mechanical low-pass filter of a spring-mass system. The active isolator includes a vibration-control system, force actuators, and displacement sensors. The passive isolator is designed to reduce the vibration disturbances in the high frequency region in which the ability of disturbance rejection of the fine pointing mechanism is not sufficient. Systems like this may create isolation of 40 dB/decade above the resonance frequency of the spring-mass system. The active isolator is used to dampen low-frequency highamplitude vibrations that cannot be suppressed by other systems.
G. Self-Tuning Feed Forward Skormin et al. [22], [23] present the self-tuning feedforward jitter-rejection technique. This method uses a miniature accelerometer to monitor the vibration characteristics. A vibration and disturbance propagating along the mechanical path are monitored and electrically compensated for before they affect the communication system. Implementation of this model in a practical system is not so simple because: a) the jitter should be monitored along not one but three orthogonal axes; b) the complex mechanical configuration of the optical system causes the jitter-compensation signal to be defined as a linear combination of all three orthogonal jitter components; c) the transfer function of the accelerometer must be stable for changes in environmental conditions. Each term in the linear combination has different weighting factors. Successful implementation of the proposed jitterrejection scheme requires that the weighting factor be selected to minimize the overall jitter effect on the beampositioning errors. This implies that the set of weighting factors cannot be selected once just to match the geometry of the optical system but is to be periodically adjusted to track time-dependent gains of accelerometers and special jitter characteristics. The method to adapt the weighting factors is based on the simplex optimization strategy deriving the optimal factor values from the real system. This method was thoroughly investigated using a pointing acquisition and tracking simulator. Simulation results clearly indicate that the suggested approach is capable of significant reduction of the jitter-related beam-positioning errors. XII. DISCUSSION AND SUMMARY This paper deals with the effects of transmitter pointing direction vibration and possible solutions to those effects for satellite optical communication networks. Real-time applications such as video conferencing, medical imaging, and multimedia from any point on earth to any other point can become a reality if optical communication networks are ARNON AND KOPEIKA: LASER SATELLITE NETWORK
implemented. This review points out that even low values of vibration amplitudes in each satellite can dramatically decrease the performance of the network. Consequently, even low vibration amplitude should not be neglected in any satellite design. It is important to understand that the performances of satellite attitude-control systems [52] determine the satellite’s vibration level [53]. This new attitude system includes sensors (such as gyros, a global positioning system, a star tracker unit, a horizon sensor, an angular displacement sensor, accelerometers, and a quartz rate sensor) and filters (such as Kalman and complementary). The analysis and synthesis of the effects of vibrations on optical receiver and heterodyne communication systems are not included in this review due to its limited scope and length. For more details on the effects of vibrations on NCFSK, the reader is referred to [15] and [54]. Some additional results about vibration effects on satellite receivers may be found in [44]. It is our hope that this review can be the basis for future analyses and syntheses of satellite optical communication networks. ACKNOWLEDGMENT The authors wish to thank Dr. D. Paul, formerly of COMSAT, and Prof. D. Wulich and Dr. V. Lyandress, Ben Gurion University, for their help. They also wish to thank the Ministry of Science and Technology, Jerusalem, Israel, for its support. REFERENCES [1] R. J. Leopold and A. Miller, “The IRIDIUM communications system,” IEEE Potentials, vol. 12, pp. 6–9, Apr. 1993. [2] , “The IRIDIUM communications system,” in IEEE Conf. MTT-S Int. Microwave Symp. Dig., vol. 2, 1993, pp. 575–578. [3] F. Ananasso, “System market and regulatory aspects for satellite personal communications,” in Mobile and Personal Satellite Communications: Proceeding of the First European Workshop on Mobile/Personal Satcoms (EMPS’94), F. Ananasso and F. Vatalaro, Eds. Berlin: Springer-Verlag, 1995, pp. 3–13. [4] P. P. Giusto and G. Qualione, “Technical alternative for satellite mobile networks,” in Mobile and Personal Satellite Communications: Proceeding of the First European Workshop on Mobile/Personal Satcoms (EMPS’94), F. Ananasso and F. Vatalaro, Eds. Berlin: Springer-Verlag, 1995, pp. 15–27. [5] B. I. Edelson and G. Hyde, “Laser satellite communications, program technology and applications,” IEEE-USA Aerospace Policy Committee Rep., Apr. 1996. [6] D. K. Paul, F. Faris, R. Garlow, T. Inukai, B. Pontano, R. Razdan, A. Ganz, and L. Caudill, “Optical intersatellite links: Application to commercial satellite communications,” in Proc. 14th AIAA Int. Communication Satellite Systems, Washington, D.C., Mar. 22–26, 1992, pp. 277–289. [7] D. K. Paul, “Optical cross links for advanced Satcom networks,” presented at the Asia Pacific Microwave Conf., New Delhi, India, Dec. 17–20, 1996. [8] M. Fujise, M. Nohara, K. Uehara, and W. Chujo, “Broadband mobile satellite communication system by LEO–SAT and optical ISL’s,” in Proc. IEEE GLOBECOM, London, vol. 1, 1992, pp. 437–442. [9] M. Wittig, L. van Holtz, D. E. L. Tunbridge, and H. C. Vermeulen, “In orbit measurements of microaccelerations of ESA’s communication satellite OLYMPUS,” in Selected Papers on Free-Space Laser Communication II—SPIE Ms 100, D. L. Begly and B. J. Thompson, Eds. Bellingham, WA: SPIE, 1994, pp. 389–398. 1659
[10] S. Dyne, P. P. Collins, and D. Tunbridge, “Satellite mechanical health monitoring,” in IEE Colloquium Advanced Vibration Measurements, Techniques and Instrumentation for the Early Predication of Failure, 1992, pp. 4/1–4/8. [11] S. J. C. Dyne, D. E. L. Tunbridge, and P. P. Collins, “The vibration environment on a satellite in orbit,” in IEE Colloquium High Accuracy Platform Control in Space, 1993, pp. 12/1–12/6. [12] K. J. Held and J. D. Barry, “Precision pointing and tracking between satellite-borne optical systems,” Opt. Eng., vol. 27, no. 4, pp. 325–333, Apr. 1988. [13] K. L. Schultz and S. Fisher, “Ground-based laser radar measurement of satellite vibrations,” Appl. Opt., vol. 31, no. 36, pp. 7690–7695, Dec. 1992. [14] K. L. Schultz, D. G. Kocher, J. A. Daley, J. R. Theriault, J. Spinks, and S. Fisher, “Satellite vibration measurements with an autodyne CO2 laser radar,” Appl. Opt., vol. 33, no. 12, pp. 2349–2355, Apr. 1994. [15] C. C. Chen and C. S. Gardner, “Impact of random pointing and tracking errors on the design of coherent and incoherent optical intersatellite communication links,” IEEE Trans. Commun., vol. 37, pp. 252–260, Mar. 1989. [16] J. D. Barry and G. S. Mecherle, “Beam pointing error as a significant parameter for satellite borne, free-space optical communication systems,” Opt. Eng., vol. 24, no. 6, pp. 1049–1054, Dec. 1985. [17] P. W. Scott and P. W. Young, “Impact of temporal fluctuations of signal-to-noise ratio (burst error) on free-space laser communication system design,” in Proc. SPIE Optical Technologies for Communication Satellite Applications, K. B. Bhasin, Ed. Bellingham, WA: SPIE, 1986, vol. 616, pp. 174–181. [18] G. A. Koepf, R. Peters, and R. P. Marshalek, “Analysis of burst error occurrence on optical intersatellitelink (ISL) design,” in Proc. SPIE Optical Technologies for Communication Satellite Applications, K. B. Bhasin, Ed. Bellingham, WA: SPIE, 1986, vol. 616, pp. 129–136. [19] R. M. Gagliardi and S. Karp, Optical Communication, 2nd ed. New York: Wiley, 1995, ch. 10, pp. 305–344. [20] S. Arnon and N. S. Kopeika, “The performance limitations of free space optical communication satellite networks due to vibrations—Analog case,” Opt. Eng., vol. 36, no. 1, pp. 175–182, Jan. 1997. [21] , “The performance limitations of free space optical communication satellite networks due to vibrations—Direct detection digital mode,” Opt. Eng., to be published. [22] V. A. Skormin, M. A. Tascillo, and D. J. Nicholson, “Jitter rejection technique in a satellite-based laser communication system,” Opt. Eng., vol. 32, no. 11, pp. 2764–2769, Nov. 1993. [23] V. A. Skormin, M. A. Tascillo, and T. E. Busch, “Adaptive jitter rejection technique applicable to airborne laser communication systems,” Opt. Eng., vol. 34, no. 5, pp. 1263–1268, May 1995. [24] R. P. Marshalek and G. A. Koepf, “Comparison of optical technologies for intersatellite links in a global telecommunication network,” Opt. Eng., vol. 27, no. 8, pp. 663–676, Aug. 1988. [25] S. G. Lambert and W. L. Casey, Laser Communication in Space. Boston, MA: Artech House, 1995. [26] P. P. Horkin and K. A. Olds, “Future optical ISL characteristics from perspective of large commercial constellations,” in CRL Int. Topical Workshop Space Laser Communication—Current Status and Future Perspectives, Tokyo, Japan, Mar. 10–11, 1997, pp. 129–135. [27] J. R. Lesh, “Overview of the NASA/JPL lasercom program,” in CRL Int. Topical Workshop on Space Laser Communication—Current Status and Future Perspectives, Tokyo, Japan, Mar. 10–11, 1997, pp. 7–13. [28] H. Lutz, “ESA’s activities toward a short-range laser communications terminal,” in CRL Int. Topical Workshop on Space Laser Communication—Current Status and Future Perspectives, Tokyo, Japan, Mar. 10–11, 1997, pp. 14–20. [29] K. Wu, “BMDO laser communications development and validation program,” in CRL Int. Topical Workshop on Space Laser Communication—Current Status and Future Perspectives, Tokyo, Japan, Mar. 10–11, 1997, pp. 21–28. [30] Y. Suzuki, “Recent R&D activities for optical data link in NASDA,” in CRL Int. Topical Workshop on Space Laser Communication—Current Status and Future Perspectives, Tokyo, Japan, Mar. 10–11, 1997, pp. 29–33. [31] K. Arki, Y. Arimoto, M. Toyoda, M. Toyoshima, M. Shikatani, T. Takahashi, T. Fukazawa, H. Okazawa, Y. Suzuki, and 1660
[32]
[33] [34]
[35]
[36]
[37]
[38] [39]
[40] [41] [42] [43] [44]
[45]
[46]
[47]
[48]
[49] [50] [51]
T. Aruga, “Laser communications experiment using ETS-VI satellite,” in CRL Int. Topical Workshop on Space Laser Communication—Current Status and Future Perspectives, Tokyo, Japan, Mar. 10–11, 1997, pp. 34–47. V. W. S. Chan, “Optical space communication,” in CRL Int. Topical Workshop on Space Laser Communication—Current Status and Future Perspectives, Tokyo, Japan, Mar. 10–11, 1997, pp. 99–106. S. Arnon and N. S. Kopeika, submitted for publication. S. Arnon, S. Rotman, and N. S. Kopeika, “Beamwidth and transmitter power adaptive to tracking system performance for free space optical communication,” in Proc. SPIE Space Sciencecraft Control and Tracking in the New Millennium, M. A. Vandor Does, Ed. Bellingham, WA: SPIE, 1996, vol. 2810, pp. 176–187. See also Appl. Opt., vol. 36, no. 24, pp. 6095–6101, Aug. 1997. , “Optimum transmitter optics aperture for free space satellite optical communication as a function of tracking system performance,” in Proc. SPIE Photonics for Space Environments IV, E. W. Taylor, Ed. Bellingham, WA: SPIE, 1996, vol. 2811, pp. 252–263. See also IEEE Trans. Aerosp. Electron. Syst., to be published. S. Arnon and N. S. Kopeika, “Effect of particulates on the performance of optical communication in space and an adaptive method to minimize such effects,” Appl. Opt., vol. 33, no. 21, pp. 4930–4937, July 1994. S. Arnon, D. Sadot, and N. S. Kopeika, “Simple mathematical models for temporal, spatial, angular, and attenuation characteristics of light propagating through the atmosphere for space optical communication: Monte-Carlo simulation,” J. Mod. Opt., vol. 41, no. 10, pp. 1955–1972, Oct. 1994. , “Analysis of optical pulse distortion through clouds for satellite to earth adaptive optical communication,” J. Mod. Opt., vol. 41, no. 8, pp. 1591–1605, Aug. 1994. S. Arnon and N. S. Kopeika, “Free space optical communication: Analysis of spatial widening of optical pulses for propagation through clouds,” Opt. Eng., vol. 34, no. 2, pp. 511–515, Feb. 1995. , “Free space optical communication: Detector array aperture for optical communication through thin clouds,” Opt. Eng., vol. 34, no. 2, pp. 516–521, Feb. 1995. , “Probing and monitoring aerosol and atmospheric cloud via an electro-optic oscillator,” Appl. Opt., vol. 35, no. 27, pp. 5427–5434, Sept. 1996. , “Adaptive optical transmitter and receiver for space communication through clouds,” Appl. Opt., vol. 36, no. 9, pp. 1987–1993, Mar. 1997. S. Arnon, S. Rotman, and N. S. Kopeika, submitted for publication. S. Arnon and N. S. Kopeika, “Adaptive suboptimum detection of optical PPM signal with detection matrix and centroid tracking,” in Proc. SPIE Photonics for Space Environments, E. W. Taylor, Ed. Bellingham, WA: SPIE, to be published. K. Inagaki and Y. Karasawa, “Ultra high speed optical beam steering by optical phased array antenna,” in , Proc. SPIE Free Space Laser Communication Technologies VIII, G. S. Mecherle, Ed. Bellingham, WA: SPIE, 1996, vol. 2699, pp. 210–217. W. M. Neubert, W. R. Leeb, and A. L. Scholts, “Experimental results on an optical array antenna for nonmechanical beam steering,” in Proc. SPIE Free Space Laser Communication Technologies IV. Bellingham, WA: SPIE, 1992, vol. 1635, pp. 82–89. D. R. Wight, J. M. Heaton, B. T. Hughes, J. C. H. Birbeck, K. P. Hilton, and D. J. Taylors, “Novel phased array optical scanning device implemented using GaAs/AlGaAs technology,” Appl. Phys. Lett., vol. 59, no. 8, pp. 899–901, Aug. 1991. G. S. Mecherle, “Detection alternative for pulse position modulation (PPM) optical communication,” in Proc. SPIE Optical Technologies for Communication Satellite Applications K. B. Bhasin, Ed. Bellingham, WA: SPIE, 1986, vol. 616, pp. 105–116. J. L. Massey, “Capacity cutoff rate, and coding for direct detection optical channel,” IEEE Trans. Commun., vol. 29, pp. 1615–1621, Nov. 1981. K. Feher, Wireless Digital Communications. Englewood Cliffs, NJ: Prentice-Hall, 1995, ch. 5, pp. 254–284. T. Kashiwase and K. Kodeki, “Design and evaluation of a vibration isolator for inter satellite laser communications,” in CRL PROCEEDINGS OF THE IEEE, VOL. 85, NO. 10, OCTOBER 1997
Int. Topical Workshop on Space Laser Communication—Current Status and Future Perspectives, Tokyo, Japan, Mar. 10–11, 1997, pp. 191–194. [52] D. Calcutt and L. Tetley, Satellite Communications: Principles and Applications. London: Edward Arnold, 1994, ch. 9, pp. 181–185. [53] M. C. Algrain, “High-bandwith attitude jitter determination for pointing and tracking systems,” Opt. Eng., vol. 36, no. 7, pp. 2092–2100, July 1997. [54] S. Arnon and N. S. Kopeika, “The performance limitations of free space optical communication satellite networks due to vibrations—Heterodyne detection,” in Proc. SPIE Photonics for Space Environments, to be published.
Shlomi Arnon received the B.Sc.E.E. and M.Sc.E.E. degrees from Ben-Gurion University of the Negev (BGU), Israel. He currently is working toward the Ph.D. degree at BGU. He has published more than 14 journal papers. His research interests include multiscattering simulation (Monte Carlo) and satellite optical communication. Mr. Arnon received the 1997 BGU Ph.D. Student Excellence Award.
ARNON AND KOPEIKA: LASER SATELLITE NETWORK
N. S. Kopeika (Senior Member, IEEE) was born in Baltimore, MD, on November 12, 1944. He received the B.S., M.S., and Ph.D. degrees in electrical engineering from the University of Pennsylvania, Philadelphia, in 1966, 1968, and 1972, respectively. His Ph.D. dissertation, supported by a NASA Fellowship, dealt with detection of millimeter waves by glow discharge plasmas and the utilization of such devices for detection and recording of millimeter-wave holograms. In 1973, he joined the Department of Electrical Engineering, Ben-Gurion University of the Negev, Israel, where he currently is a Professor and incumbent of the Reuben and Frances Feinberg Chair in Electrooptics. During 1989–1993, he served two terms as Department Chair. During 1978–1979, he was a Visiting Associate Professor in the Department of Electrical Engineering, University of Delaware, Newark. His current research interests include atmospheric optics, effects of surface phenomena on optoelectric device properties, optical communication, electronic properties of plasmas, laser breakdown of gases, the optogalvanic effect, electromagnetic (EM) waveplasma interaction in various portions of the EM spectrum, and utilization of such phenomena in EM wave detectors and photopreionization lasers. He has published more than 120 reviewed journal papers in the above areas. He has been particularly active in research of time response and impedance properties of plasmas and authored a general unified theory to explain EM wave-plasma interactions all across the electromagnetic spectrum. His earlier published work on the optogalvanic effect preceded the naming of the effect. He also has done extensive work on the wavelength dependence of image resolution through the open atmosphere, particularly with regard to spatial coherence degradation and spatial frequency dependence resulting from forward light scattering by relatively large airborne particulates. He and his students have developed methods to predict atmospheric modulation transfer function, including both turbulence and aerosol modulation transfer function components, according to weather and to use that information in image restoration so as to deblur such effects. Also, methods to calculate numerically in real time optical transfer function for any type of image motion and vibration have been developed, and these too have also been used in image restoration. He contributed actively toward the development of postfabrication techniques for wavelength tuning of semiconductor light sources by external means. He is the author of the textbook A System Engineering Approach to Imaging (SPIE Optical Engineering Press, to be published). Recently, he has been involved in adaptive techniques for satellite optical communication. Dr. Kopeika is a senior member of the International Society for Optical Engineering, the Optical Society of America, and the Laser and Electrooptics Society of Israel.
1661
