OSA's Digital Library

Energy Express

Energy Express

  • Editor: Bernard Kippelen
  • Vol. 19, Iss. S4 — Jul. 4, 2011
  • pp: A639–A648

Optics InfoBase > Energy Express > Volume 19 > Issue S4 > Page A639

« Show journal navigation

Optimization of multi-element LED source for uniform illumination of plane surface

Deepa Ramane and Arvind Shaligram  »View Author Affiliations


Optics Express, Vol. 19, Issue S4, pp. A639-A648 (2011)
http://dx.doi.org/10.1364/OE.19.00A639


View Full Text Article

Acrobat PDF (1387 KB)


Advanced Search

Browse Journals / Lookup Meetings

Browse by Journal and Year


   


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools

Share
Citations

Abstract

In many lighting applications at the lux level, full coverage of the target surface, uniformity, and glare are important requirements. In the case of LEDs, luminaire optical output depends upon the source of luminous flux, radiation spatial distribution, geometry of the LED array, and source-to-target distance. One has to optimize these parameters to fulfill the needs of the application and to have an efficient LED luminaire. This paper discusses the interdependencies of these parameters. The optimization of a ring LED source configuration is demonstrated for an application of uniform illumination of a plane surface with minimum source flux and with no glare.

© 2011 OSA

1. Introduction

Recent developments in high-brightness power LEDs have changed the face of the lighting world. LEDs are replacing conventional light sources in almost all of the illumination applications. The output of LED luminaires depends upon the geometric and photometric parameters. Proper selection of these parameters can lead to a better and efficient illumination system. The majority of efforts claim to provide optimal solutions to have uniform illumination over a plane surface. According to them, uniformity can be achieved either by using wide-angle LEDs, or by increasing the LED density, or by increasing the source-to-target distance [1

I. Moreno, “Configuration of LED arrays for uniform illumination,” Proc. SPIE 5622, 713–718 (2004). [CrossRef]

–6

M. A. Gennert, N. Wittels, and G. L. Leatherman, “Uniform frontal illumination of planer surfaces: where to place the lamps,” Opt. Eng. 32(6), 1261–1271 (1993). [CrossRef]

]. The principle used in all of these is the merging of irradiance distribution of the LEDs to produce uniform illumination. For a given set of parameters for spatial distribution, the number of LEDs, and the source-target distance in a multi-element LED source, relative placement of LEDs should be properly optimized for better uniformity. Papers have reported the optimization of the distance between two LEDs using mathematical equations based on inverse square law and sparrow’s criteria to obtain uniform illumination [7

A. J.-W. Whang, Y.-Y. Chen, and Y.-T. Teng, “Designing uniform illumination systems by surface-tailored lens and configurations of LED arrays,” J. Disp. Technol 5(3), 94–103 (2009). [CrossRef]

,8

Z. Qin, K. Wang, F. Chen, X. Luo, and S. Liu, “Analysis of condition for uniform lighting generated by array of light emitting diodes with large view angle,” Opt. Express 18(16), 17460–17476 (2010). [CrossRef] [PubMed]

]. Different configurations of LED arrays were studied by Moreno to find the maximum LED density and the minimum LED-to-detector distance that can produce satisfactory uniformity [9

I. Moreno, J. Muñoz, and R. Ivanov, “Uniform illumination of distant targets using a spherical light-emitting diode array,” Opt. Eng. 46(3), 033001 (2007). [CrossRef]

]. However, greater LED density is limited by cost, available space, and by thermal problems [8

Z. Qin, K. Wang, F. Chen, X. Luo, and S. Liu, “Analysis of condition for uniform lighting generated by array of light emitting diodes with large view angle,” Opt. Express 18(16), 17460–17476 (2010). [CrossRef] [PubMed]

]. Guttsait has discussed an analysis of LED modules considering illumination level and uniformity for local illumination [10

E. M. Guttsait, “Analysis of LED modules for local illumination,” J. Commun. Technol. Electron. 52(12), 1377–1395 (2007). [CrossRef]

].

Many practical applications of illumination must meet the lighting requirements, of which uniformity is not the only criteria. Rather, full coverage of the surface with the recommended lux level and sufficient limitation on glare or hot spots is expected. These primary requirements must be fulfilled with a minimum flux source. The secondary issues like thermal effects, life, cost, etc., impose further restrictions on optimization of the configuration parameters. Here, optimization of geometric and photometric factors of the source is critical. To improve uniformity, if the source density is increased and the source-source distance is decreased for merging of irradiance, then the luminaire creates extra brightness, which gives rise to uncomfortable glare and causes loss in the efficiency of the eye. It also increases energy consumption and the cost of the luminaire. Having fewer sources means insufficient light, which leads to dullness in the working environment, strain on the eyes, and inefficient working conditions. Even an optimal source density with improper placement of LEDs can create hot spots in some areas, while some other areas can have low illuminance. Another way to fully illuminate a surface is to expand the Gaussian beam much larger than the area to be illuminated. One uses the central region of the beam where the luminous intensity varies slowly [11

J. A. Hoffnagle and C. M. Jefferson, “Design and performance of a refractive optical system that converts a Gaussian to a flattop beam,” Appl. Opt. 39(30), 5488–5499 (2000). [CrossRef]

]. Expansion is possible by increasing the angle aperture of the source or by increasing the source-to-target distance. But, both lead to a decrease in the illuminance level on the target plane. It decreases the efficacy of the source, as only a fraction of the flux source can be utilized for illumination of the surface under consideration.

Thus uniformity, optimal lux level, power efficiency, spatial distribution of the source, and source geometry are interrelated. The present work aims at optimization of all these parameters so as to achieve an energy-efficient illumination system fulfilling the requirements of the application. For optimization, the simulation tool ‘OPTSIMLED’ [12

R. Deepa and S. Arvind, “Modeling and simulation of multi-element LED source,” J.Light. Visual Environ. (to be published).

] is used, which analyzes the LED multi-element source module and computes the illuminance distribution, average lux on the plane surface, and the uniformity and diversity ratio of illumination. It also gives a colored illumination pattern and illuminance distribution pattern on the target plane. Using this tool, LED luminaires of variable geometric and photometric properties are simulated. The simulated results are analyzed further to optimize the source configuration so as to achieve the design goals of the application.

2. Photometric terms used

  • Illuminance: Illuminance is a measure of photometric flux per unit area or visible flux density. It is expressed in lux (lumens per square meter). When designing a lighting system, recommended illuminance needs to be achieved.
  • Luminous flux or source flux: It describes the total amount of light emitted by a source and is measured in lumens.
  • Radiation spatial distribution or angle aperture: The radiation pattern describes the relative light strength in any direction from the light source and is characterized by the spatial intensity distribution in the far field. Manufacturers provide θ1/2, view angle, or angle aperture when radiant intensity is half of the value at 0⁰ to characterize the spatial distribution of LEDs with a single LED peak.

3. Design goals and optimization methodology

3.1 Design goals

To illustrate the optimization process, an application of illumination of a plane surface with a minimum illuminance level, good uniformity, and minimum glare is considered. The task is to be accomplished with a minimum flux source. The optimization parameters are the number of LEDs, their relative placement, flux source, and spatial distribution. The design goals for this application are listed below:

  • 1) Full plane surface area is illuminated with minimum illuminance, ELT, lx.
  • 2) Uniform illumination is desired, which is characterized by the uniformity ratio. It is defined as,

    Uniformity  =  Maximum  illuminance   Average   illuminance  .

If a plane is illuminated with equal illuminance all over the surface, then the uniformity ratio becomes one. It is the ideal situation.

  • 3) Better diversity ratio gives difference between maximum and minimum illuminance levels.

    Diversity   Ratio =  Maximum  illuminance  Minimum  illuminance  .

For uniform illumination, maximum and minimum illuminance levels match, and ideally diversity ratio becomes one.

  • 4) Glare should be minimal. It imposes the upper limit on the acceptable illuminance level, EUT, lx. It is the minimum illuminance that produces an uncomfortable feeling for the eyes.
  • 5) Above four goals to be achieved with minimum source flux to guarantee maximum efficacy.
  • 6) Heat sink requirement should be optimum.
  • 7) Fabrication and maintenance cost of luminaire should be minimal.

3.2 Methodology

The optimization process is summarized in Flowchart 1. After finalizing the application, the design criteria are fixed and are fed to the tool. The configurations consisting of white LEDs are simulated using the simulation tool “OPTSIMLED” and are analyzed based on the design goals. Analysis results are used to optimize the source configuration. The geometry of the LED luminaire, the number of LEDs, their relative positions (x, y coordinates), and the spatial and spectral distribution of white LEDs is provided as input to the tool. For simplicity it is assumed that all LEDs are of the same specifications. The tool computes uniformity and diversity ratio and generates the illumination pattern on the plane. The illuminance distribution pattern specifies three illuminance regions:

  • i) Region I: having lux less than ELT lx,
  • ii) Region II: having lux level between ELT to EUT lx,
  • iii) Region III: having lux greater than EUT lx.

The first and third regions are non-acceptable regions because of the low and high illuminance levels, respectively. The third region is a hot-spot region, which produces glare.

The configuration that illuminates a full target plane with no glare is checked for source flux, and the one having the minimal flux requirement is further optimized based on the secondary issues described in Section 4.2.

Flowchart 1 Optimization methodology of LED luminaire.

4. Results and discussion

4.1 Primary aspects for optimization

This section discusses the optimization of an LED source luminaire to illuminate a work plane of dimensions 100 × 100 cm2 with ELT as 100 lx. The luminaire is placed parallel above the plane at a height of 50 cm. Glare is produced if the intensity goes above 300 lx (EUT), and the region is termed as a hot-spot zone. It means that the portion of the work plane having illuminance between 100 lx–300 lx is within acceptable zone.

Since illumination with full-area coverage is considered, LEDs with angle apertures less than 50⁰ are not considered as they create a spot of light. For optimization, LEDs with angle apertures varying from 60⁰ to 160⁰ are selected. The results are described for a single LED and for a multi-element LED source. In the multi-element source, many ring-shaped configurations with variable LED densities and radii of rings are considered.

A] Single LED source:

An LED is placed at the center of the illumination plane at a height of 50 cm. The system is simulated for LEDs with flux of 2000 lm, 2500 lm, 3000 lm, 3500 lm, and 4000 lm. For each configuration area with illuminance greater than 100lx, hot-spot area, and uniformity and diversity ratio, the illumination pattern and illuminance distribution are recorded. Graph 1 gives simulated results for the area of coverage (having illuminance greater than 100 lx) as a function of the angle aperture and luminous flux output for a single LED. It is found that full-area coverage with minimum illuminance of 100 lx is possible if the source flux is greater than or equal to 3500 lm. Below 3500 lm, the area coverage is not 100%; i.e., part of the target plane receives illuminance at less than 100 lx. For most of the angle aperture values the characteristics are almost horizontal for a source flux greater than 2500 lm; this means that there is no significant increase in coverage area. There is no point in using a source with a flux value greater than 3500 lm. However, illuminance distribution patterns show hot spots at the center for a source flux greater than 2500 lm.

Graph 1 Area coverage dependency on angle aperture and flux output for single LED.

The graph also gives the dependency of the coverage area on the radiation spatial distribution of the LED. The area of coverage increases with the increase in angle aperture, but again shows a decrease for higher values. As the angle of aperture increases, the total source flux spreads over a cone larger in diameter, which results in larger coverage of the area. At the same time, the illuminance level falling on the target plane decreases, which reflects the reduction of the diameter of the hot-spot area. [See Fig. 1 .] For an angle aperture of 140⁰–160⁰, the illuminance level becomes less to satisfy ELT, the minimum illuminance requirement, especially at the corners of the target plane. Therefore, the area of coverage reduces for higher values of the angle aperture. The reason for this behavior is also supported by the illuminance distribution curves given in Graph 2.

Fig. 1 Illuminance distribution patterns for Source flux = 3000 lm for angle aperture of 60⁰–160⁰ with Region I represented by violet color, Region II represented by green color, Region III represented by blue color.
Graph 2 Illuminance distribution curves for source flux = 3000 lm for angle aperture varying from 60° to 160°.

Graph 2 gives the illuminance distribution curves over the plane surface for angle apertures of 60⁰ to 160⁰. The curves become more flat with an increase in angle. The peak illuminance reduces with an increase in the area having a lux level greater than 100 lx. Even the hot-spot diameter reduces as the area with illuminance greater than 300 lx decreases. It is seen that the corner portion of the plane surface receives lux less than 100 lx. The contribution due to the corner portion increases with an aperture angle up to 130⁰. For 140⁰ onwards, the contribution starts decreasing due to the wide angular spreading of flux. This is the reason for a decrease in the area of illuminance greater than 100 lx for LEDs having wider spatial distribution.

The configurations for which illumination of the surface is greater than 90% are listed in Table 1. To select the optimized one, these configurations are further analyzed based on the uniformity ratio, diversion ratio, and hot spot. Though for the source flux = 3500 lm the area of coverage is larger, the hot-spot area is also large and even a higher flux source means that the source efficacy is less. Based on quantitative results, the optimized values for a single LED configuration placed at the center of the work plane at a height of 50 cm are a flux source of 3000 lm and an angle aperture of 130⁰. The reason for this is the configuration:

  • i) provides almost full coverage of the work plane (99%)
  • ii) provides better uniformity and diversity ratio (1.9 and 4.9 respectively)
  • iii) has a minimum hot spot area (1410 cm2)
  • iv) uses a minimum flux source fulfilling above criteria (3000 lm)
    Table 1  Quantitative Analysis of Configurations of Single LED
    Optical Specifications of LED Area (cm2) for Illuminance >100lxHot Spot Area (cm2) for Illuminance >300lxUniformity RatioDiversity Ratio
    Luminous FluxAngle Aperture (⁰)
    2500110911613002.055.96
    250012091168841.975.34
    250013090654841.94.9
    3000110971923082.055.96
    3000120985019561.975.34
    3000 130 9890 1410 1.9 4.9
    3500110991233162.055.96
    3500120995829961.975.34
    3500130995826281.94.9

B] Multi-element ring source with variable radial distance and LED density:

No single LED configuration provides 100% coverage with zero hot-spot areas. To overcome this limitation one must go for a multi-element source luminaire. Here more parameters, along with flux source and angle aperture, affect the illumination pattern. These are the number of LEDs and their relative placement in the luminaire. Ring-shaped multi-element LED configurations with variable radial distances and number of LEDs are studied for optimization. The LEDs are placed symmetrically around the center. If ‘r’ is the radius of a ring having ‘n’ numbers of LEDs, then the position of the ith LED on the source panel is given by x, y coordinates

x =  (r*cos ( i∗ θ step ),  y=  (r*sin ( i∗ θ step),

where θ step= 360/n.

Ring LED luminaire configurations are studied for LED densities varying from 2 to 7. Each configuration is simulated for center-to-LED ring radial distances of 10, 20, 30, and 40 cm. Flux output of individual LEDs in the configuration are adjusted such that the total flux of each configuration remains the same as 3000 lm. Graphs 3(a) to 3(f) show the dependency of the coverage area on spatial distribution and variable radial distance for variable LED densities. The results show that even for these configurations, wide-angle LEDs with angle apertures up to 130⁰ are preferable for larger coverage. For apertures greater than 130⁰, the area starts decreasing. As the radial distance increases, the characteristics are becoming more flat, indicating that the coverage area becomes independent of spatial distribution. Especially for four LED configurations, the choice of spatial distribution of LEDs has less variation in the area of coverage for a radial distance of 40 cm.

Moreover, it is seen that in all the configurations, the results are better for a radial distance of 40 cm. These configurations are further analyzed for illumination system parameters.

Graph 3 Dependency of coverage area on LED density and spatial distribution.

Quantitative results given in Table 2 reveal that a configuration with four LED is best as it gives full coverage, minimum uniformity ratio, minimum diversion ratio, and zero hot-spot areas. It is obvious since in the configuration step the angle is 90⁰, making the source geometry the same as square-array geometry, and the target plane is also square. To confirm the minimum source flux requirement, the optimized geometry is analyzed with a flux source at 2500 lm and 2800 lm. Here 97.7% and 98.9% coverage is possible with almost the same uniformity and diversity values. It confirms that a source with a minimum of 3000 lm is essential.

Table 2  Analysis of ring LED luminaire for a fixed 3000lm total flux
Number of LEDs23 4 567
Flux of each source(lm)15001000 750 600500428.5
Total flux of LED luminaire30003000 3000 300030003000
Angle aperture of source130130 130 130130130
Radial distance (cm)4040 40 404040
Uniformity ratio1.611.19 1.15 1.161.161.16
Diversity ratio3.762.33 1.88 2.042.012.01
Area (cm2) for illuminance>100lx92319812 10000 993599589958
Hot spot area (cm2)5160 0 000
Illuminance pattern on target plane with color representation as in Fig. 1 oe-19-S4-A639-i001oe-19-S4-A639-i002oe-19-S4-A639-i003oe-19-S4-A639-i004oe-19-S4-A639-i005oe-19-S4-A639-i006

4.2 Secondary aspects of optimization

It is revealed from the above simulated results that distributed LED sources are better in performance as far as illumination output is concerned. For the same source of flux output, the uniformity and diversity ratio approaches ideal with a reduction in hot spots. At this stage, one must consider secondary issues like thermal management, assembly, and maintenance costs for the final design. Out of this thermal management of high-power LEDs is a crucial area that affects power efficacy as well as light output. Some of the electricity in an LED becomes heat rather than light. If that heat is not removed, the LEDs run at a high temperature, which not only lowers the efficiency, but also makes them more dangerous and less reliable. Therefore, power LEDs require heat sinks for proper thermal management in a large surface area.

If an LED is of higher lumen, then a larger heat sink is required. If the distributed source consists of more LEDs, then the flux output of individual LEDs is less, i.e., the source power is less, the thermal runaway problem is less crucial, and the heat sink size is smaller. Another advantage is that all LEDs will not fail simultaneously, so zero illuminance will never occur. At the same time, there are some disadvantages with the distributed source. Assembly cost is high, as more luminaries must be fabricated. Maintenance costs are also high. Thus, these secondary issues affect the source luminaire choice.

5. Conclusion

This paper has focused on the optimization of LED luminaires in applications where lux level and glare are important along with uniformity and diversity. Here, the source luminaire is optimized based on primary and secondary objectives. From the above example, one can conclude the following:

  • • LEDs with angle apertures less than 60° tend to create hot spots and cannot be used to illuminate wide surface areas.
  • • At the same time, very wide angle LEDs are not advisable for applications where the target plane is to be illuminated with a minimum threshold of illuminance. Source flux gets spread over a large area and the lux level goes below the minimum threshold lux level, especially at the corners of the plane. Most of the light rays strike the plane outside the targeted area and source efficacy decreases.
  • • A single LED is not advisable to illuminate a full plane surface.
  • • In ring-shaped multi-element LED luminaires spatial distribution, ring radius, LED density, and distance between the LEDs decide the optimized geometry.
  • • High-power LEDs have adverse reactions to heat, and thermal management plays a vital role in finalizing the luminaire design.
  • • More than one LED luminaire distributed over an area is better for thermal management.

The optimization method demonstrated for one application can be applied to a decorative lighting application in which small zones are to be illuminated with specific illuminance and nearby areas require different lux values. Here, illuminance levels can be adjusted by switching the LED array ON and OFF according to the optimized configuration.

References and links

1.

I. Moreno, “Configuration of LED arrays for uniform illumination,” Proc. SPIE 5622, 713–718 (2004). [CrossRef]

2.

S. K. Kopparapu, “Lighting design for machine vision application,” Image Vis. Comput. 24(7), 720–726 (2006). [CrossRef]

3.

I. Moreno, “Design of LED spherical lamps for uniform far-field illumination,” Proc. SPIE 6046, 60462E (2006). [CrossRef]

4.

N. Wittels and M. A. Gennert, “Optimal lighting design to maximize illumination uniformity,” SPIE 2348, 46–56 (1994). [CrossRef]

5.

H. Yang, J. W. M. Bergmans, T. C. W. Schenk, J.-P. M. G. Linnartz, and R. Rietman, “Uniform illumination rendering using an array of LEDs: a signal processing perspective,” IEEE Trans. Signal Process. 57(3), 1044–1057 (2009). [CrossRef]

6.

M. A. Gennert, N. Wittels, and G. L. Leatherman, “Uniform frontal illumination of planer surfaces: where to place the lamps,” Opt. Eng. 32(6), 1261–1271 (1993). [CrossRef]

7.

A. J.-W. Whang, Y.-Y. Chen, and Y.-T. Teng, “Designing uniform illumination systems by surface-tailored lens and configurations of LED arrays,” J. Disp. Technol 5(3), 94–103 (2009). [CrossRef]

8.

Z. Qin, K. Wang, F. Chen, X. Luo, and S. Liu, “Analysis of condition for uniform lighting generated by array of light emitting diodes with large view angle,” Opt. Express 18(16), 17460–17476 (2010). [CrossRef] [PubMed]

9.

I. Moreno, J. Muñoz, and R. Ivanov, “Uniform illumination of distant targets using a spherical light-emitting diode array,” Opt. Eng. 46(3), 033001 (2007). [CrossRef]

10.

E. M. Guttsait, “Analysis of LED modules for local illumination,” J. Commun. Technol. Electron. 52(12), 1377–1395 (2007). [CrossRef]

11.

J. A. Hoffnagle and C. M. Jefferson, “Design and performance of a refractive optical system that converts a Gaussian to a flattop beam,” Appl. Opt. 39(30), 5488–5499 (2000). [CrossRef]

12.

R. Deepa and S. Arvind, “Modeling and simulation of multi-element LED source,” J.Light. Visual Environ. (to be published).

OCIS Codes
(230.3670) Optical devices : Light-emitting diodes
(220.2945) Optical design and fabrication : Illumination design

ToC Category:
Light-Emitting Diodes

History
Original Manuscript: April 1, 2011
Revised Manuscript: May 3, 2011
Manuscript Accepted: May 9, 2011
Published: May 12, 2011

Citation
Deepa Ramane and Arvind Shaligram, "Optimization of multi-element LED source for uniform illumination of plane surface," Opt. Express 19, A639-A648 (2011)
http://www.opticsinfobase.org/oe/abstract.cfm?URI=oe-19-S4-A639


Sort:  Author  |  Year  |  Journal  |  Reset  

References

  1. I. Moreno, “Configuration of LED arrays for uniform illumination,” Proc. SPIE 5622, 713–718 (2004). [CrossRef]
  2. S. K. Kopparapu, “Lighting design for machine vision application,” Image Vis. Comput. 24(7), 720–726 (2006). [CrossRef]
  3. I. Moreno, “Design of LED spherical lamps for uniform far-field illumination,” Proc. SPIE 6046, 60462E (2006). [CrossRef]
  4. N. Wittels and M. A. Gennert, “Optimal lighting design to maximize illumination uniformity,” SPIE 2348, 46–56 (1994). [CrossRef]
  5. H. Yang, J. W. M. Bergmans, T. C. W. Schenk, J.-P. M. G. Linnartz, and R. Rietman, “Uniform illumination rendering using an array of LEDs: a signal processing perspective,” IEEE Trans. Signal Process. 57(3), 1044–1057 (2009). [CrossRef]
  6. M. A. Gennert, N. Wittels, and G. L. Leatherman, “Uniform frontal illumination of planer surfaces: where to place the lamps,” Opt. Eng. 32(6), 1261–1271 (1993). [CrossRef]
  7. A. J.-W. Whang, Y.-Y. Chen, and Y.-T. Teng, “Designing uniform illumination systems by surface-tailored lens and configurations of LED arrays,” J. Disp. Technol 5(3), 94–103 (2009). [CrossRef]
  8. Z. Qin, K. Wang, F. Chen, X. Luo, and S. Liu, “Analysis of condition for uniform lighting generated by array of light emitting diodes with large view angle,” Opt. Express 18(16), 17460–17476 (2010). [CrossRef] [PubMed]
  9. I. Moreno, J. Muñoz, and R. Ivanov, “Uniform illumination of distant targets using a spherical light-emitting diode array,” Opt. Eng. 46(3), 033001 (2007). [CrossRef]
  10. E. M. Guttsait, “Analysis of LED modules for local illumination,” J. Commun. Technol. Electron. 52(12), 1377–1395 (2007). [CrossRef]
  11. J. A. Hoffnagle and C. M. Jefferson, “Design and performance of a refractive optical system that converts a Gaussian to a flattop beam,” Appl. Opt. 39(30), 5488–5499 (2000). [CrossRef]
  12. R. Deepa and S. Arvind, “Modeling and simulation of multi-element LED source,” J.Light. Visual Environ. (to be published).

Cited By

 Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.


Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited