Open Access
Add to BibSonomyAbstract
We demonstrate an interference cancellation technique of optical AND gate receiver using optical thyristor for fiber-optic code division multiple access (FO-CDMA) systems. In particular, we fabricate the optical thyristor operating as optical hard-limiter and evaluate that the optical AND gate receiver using fabricated optical thyristor excludes the peaks of side-lobe and cross-correlation result in the system performance degradation. It found that the optical AND gate receiver using optical thyristor excludes the intensity of interference signal resulting in that the peaks of side-lobe and cross-correlation can be fully eliminated for any two users. Therefore, the optical AND gate receiver using optical thyristor is shown to be effective to accommodate more simultaneous users.
©2008 Optical Society of America
1. Introduction
The success of optical signal processing based on fiber optic delay elements in optical sensors and optical computing has prompted an intense interest in their use in all-optical communications networks [1]. Like other spread spectrum systems, the FO-CDMA receiver suffers from interferences of other simultaneous users, which is called multiple-access interference (MAI). Because the FO-CDMA is interference limited system, the number of simultaneous users is much less than the number of subscribers. Many schemes have been proposed to increase the number of simultaneous users [2, 3]. Among various optical CDMA techniques introduced to date, FO-CDMA using optical orthogonal codes, and their possible variations, has received much attention due to their simple structure and compatibility with today’s intensity modulation/direct-detection fiber-optic transmission system [4-6].
Furthermore, the auto-correlation of each sequence exhibits a narrow main-lobe and adequately small side-lobes, and the cross-correlation between any two sequences always remains small. Also, the peaks of side-lobe and cross-correlation are designed to be constant by adjusting the length of optical delay line (ODL). However, optical receivers with conventional structure present the peaks of side-lobe and cross-correlation according to time delay when optical decoder detects the desired signal from encoded optical signal. These peaks degrade the system performance and make it difficult to detect desired signal [7,11]. If the peaks of side-lobe and cross-correlation can be minimized or eliminated, the system performance of FO-CDMA can be much improved. Figure 1 shows the optical AND gate receiver using optical thyristor for FO-CDMA system. The optical AND gate receiver using optical thyristor reduces drastically the intensities of interference signals. Therefore, we analyze the correlation properties of the FO-CDMA system with the optical AND gate receiver using optical thyristor, and found that the superior performance with increased optical codes can be obtained.
2. Optical thyristor operating as optical hard-limiter
An optical thyristor with a very thin layer and PnpN hetrojunction is an optical device with fast switching speed as shown in Fig. 2. When the light is given to the optical thyristor in OFF state, current-voltage characteristic curve moves from C1 to C2 or C3 according to input light intensity. The C1 is the I-V curve without optical input, and other curves (C2, C3) are the I-V curves with optical inputs. At this time, the optical thyristor still remains in OFF state by the light moving the current-voltage (I-V) curve to C2, but the optical thyristor turns into ON state by the light moving the I-V curve to C3. Accordingly, the operating point changes from S1 (OFF-state) to S2 (ON-state), and the optical thyristor emits the light of constant intensity [9]. This emission results from the external voltage (VD) and resistor. Moreover, because the light intensity emitted is constant regardless of applied light intensity, it can serve as an optical hard-limiter which is applicable to FO-CDMA system. Among these optical thyristors, the PnpN optical thyristor has more increased optical sensitivity and faster operation than conventional optical thyristor. Therefore, PnpN optical thyristor can be used as the optical hard-limiter.
Fig. 2. Typical S-shape I-V characteristics of an optical thyristor operating as optical hard-limiter
The vertical-cavity laser (VCL)-depleted optical thyristor (DOT) shown in Fig. 1 was grown on n-GaAs substrates by metal organic chemical vapor deposition (MOCVD). These devices have a PnpN triple junction structure with two distributed Bragg reflector (DBR) mirrors consisting of Al0.9Ga0.1As/Al0.16Ga0.84As layers with linearly graded transition layers. The alloy composition grading allows reduction in the series resistance of the devices. The cavity space between the top and the bottom mirrors is 4λ, where λ is the emission wavelength in the semiconductor medium. This device structure allows either electrical or optical switching from the OFF state to the ON state. During the ON state the VCL-DOT has low impedance and emits laser light. To increase the efficiency of optical emission, an undoped multiple quantum well layer has been incorporated as the active layer of the VCL-DOT. The active region also acts as an absorption region for optically switching the VCL-DOT. Although the active region is thin, enhancement of absorption is expected because of multi-reflection by the two DBR mirrors. The designed parameters such as doping concentration and layer thickness of the each active layer were simulated using a finite difference method. The top ohmic contact was fabricated using a lift-off process of Ti-Au (20 nm/150 nm), and the bottom was contacted with AuGe-Ni-Au (40 nm/20 nm/150 nm) located on the substrate. To pattern the devices, reactive ion etching was used to etch square mesas ranging in size from 30X30 to 60X60 µm with a size variation of 0.5 um. The VCL-DOTs are then placed in a wet thermal oxidation furnace to laterally oxidize the current confinement layer and anneal the ohmic contacts.
3. Results and discussion
We have experimentally investigated the optical AND gate receiver using optical thyristor, which minimizes or eliminates the peaks of side-lobe and cross-correlation result in the system performance degradation. Also, we discuss a means of reducing the effective MAI signals by placing the VCL-DOT in the optical AND gate receiver. Previously, optical receivers with conventional structure present the overlapping signals according to time delay when optical decoder detects the desired signal from encoded optical signals. The overlapping signals in the channel and the peaks of side-lobe due to decoding processing degrade the system performance and make it difficult to detect desired signal [4, 7]. If the overlapping signals in the channel and the peaks of side-lobe due to decoding processing can be eliminated, the performance of FO-CDMA system can be much improved. In this work, we demonstrate the oxidized PnpN VCL-DOT operating as optical hard-limiter in the optical AND gate receiver. In order to exclude the peaks of side-lobe and cross-correlation result in the system performance degradation, VCL-DOTs are placed before and after the ODL, as shown in Fig. 1. The first VCL-DOT placed in the optical AND gate receiver operates the intensities of pulses which are overlapped on the channel by MAI signals, as shown in Fig. 3(a).
Fig. 3. Optical input and the optical output pulses of (a) first and (b) second VCL-DOT placed in the optical AND gate receiver.
If an optical intensity is bigger than or equal to one, the first VCL-DOT placed in the optical AND gate receiver retransmits the encoded signal to ODL after clipping the intensity back to one. However, if an optical intensity was smaller than one, the response of the first VCL-DOT would be zero as shown in Fig. 3(a). This ideal nonlinear process would enhance the system performance because it would exclude some combinations of interference patterns from causing errors as in all previous works [7]. On the other hand, the length of ODL in the optical AND gate receiver is constituted by using the length of ODL decided in encoding processing. The length of the ODL should be adjusted so that there present only one optical signal that is larger than the optimal threshold. As a result, the interference signals due to decoding process can be eliminated when the second VCL-DOTs placed in the optical AND gate receiver operates the ON state for the same or bigger optical intensity than specified level and OFF state for the smaller signal than optimal threshold, as shown in Fig. 3(b). Figure 4 shows the response of optical receivers when user 1 transmits the information bits 1. The output of optical receivers to its desired sequence is a convolution that has a maximum at the sequence frame time (correlation time). Figure 4(a) corresponds to the auto-correlation of code A for all shifts of the code. The peaks of side-lobe equal to 0 or 1 (λ=1), is the desired constraint imposed on the auto-correlation of all optical orthogonal codes. The peaks of side-lobe result in the system performance degradation, and make it difficult to detect desired signal. Moreover, because the FO-CDMA is interference limited system, the number of simultaneous users is much less than the number of subscribers. Figure 4(b) clearly shows that the optical AND gate receiver using optical thyristor is effective to reduce the effect of the peaks of side-lobe due to decoding processing.
Fig. 4. Auto-correlation properties of (a) optical receiver with conventional structure and (b) optical AND gate receiver using optical thyristor when λ=1.
Figure 5 shows the output of the optical receivers designed for user 1 when user 2 transmits the information bits 1. In all previous works, as shown in Fig. 5(a), an error occurs when the desired transmitter sends a zero and the interference signal, due to the other user, reproduces the desired pattern up to the threshold or higher at the front end of the desired receiver’s correlator. In order to exclude the peaks of cross-correlation result in the system performance degradation, we propose the optical AND gate receiver using optical thyristor. Figure 5(b) shows that the optical AND gate receiver using optical thyristor fully excludes the peaks of cross-correlation due to decoding processing. Figure 5(b) clearly demonstrates that the peaks of cross-correlation shown in Fig. 5(a) are fully eliminated for any two users. It means that the more active user can access to the FO-CDMA system, and code length can be decreased due to the elimination of the peaks of side-lobe and cross-correlation spread along the time scale within the same interval by using the optical AND gate receiver using optical thyristor.
Fig. 5. Cross-correlation properties of (a) optical receiver with conventional structure and (b) optical AND gate receiver using optical thyristor when λ=1.
4. Conclusion
We have experimentally investigated the optical AND gate receiver using optical thyristor, which excludes the peaks of side-lobe and cross-correlation result in the system performance degradation. In particular, we fabricate the optical thyristor operating as optical hard-limiter and discuss a means of reducing the effective MAI signals by the optical AND gate receiver using the fabricated optical thyristor. As a result, we clearly show that the peaks of side-lobe and cross-correlation presented in previous works can be eliminated perfectly by placing the VCL-DOT in the optical AND gate receiver. It is found that the optical AND gate receiver using optical thyristor excludes the intensity of interference signal resulting in that the peaks of side-lobe and cross-correlation can be fully eliminated for any two users.
Acknowledgment
This work was supported by the Brain Korea 21 project.
References and links
1. P. Azmi, M. Nasiri-Kenar, and J. A. Salehi, “Internally Channel-Coded Framed Time-Hopping Fiber-Optic CDMA Communications,” J. Lightwave Technol. 23, 3702–3707 (2005). [CrossRef]
2. E. S. Shivaleela, A. Selvarajan, and T. Srinivas, “Two-Dimensional Optical Orthogonal Codes for Fiber-Optic CDMA Networks,” J. Lightwave Technol. 23, 647–654 (2005). [CrossRef]
3. H. M. H. Shalaby, “Performance analysis of an optical CDMA random access protocol,” J. Lightwave Technol. 22, 1233–1241 (2004). [CrossRef]
4. A. Keshavarzian and J. A. Salehi, “Optical Orthogonal Code Acquisition in Fiber-Optic CDMA Systems via the Simple Serial-Search Method,” IEEE Trans. Commun. 50, 473–483 (2002). [CrossRef]
5. W. Huang, I. Andonovic, and M. Tur, “Code acquisition in coherent optical pulse CDMA systems utilizing coherent correlation demodulation,” IEEE Trans. Commun. 48, 611–621 (2000) [CrossRef]
6. F. R. K. Chung, J. A. Salehi, and V. K. Wei, “Optical orthogonal codes: Design, analysis, and applications,” IEEE Trans. Inform. Theory IT-35, 595–604 (1989). [CrossRef]
7. J. A. Salehi, “Optical CDMA via Temporal Code”, IEEE Trans. Commun. 40, 1162–1170 (1992). [CrossRef]
8. M. Kuijk, P. L. Heremans, G. Borghs, and R. Vounckx, “Depleted double-heterojunction optical thyristor”, Appl. Phys. Lett. 64, 2073–2075 (1994). [CrossRef]
9. K. D. Choquette, R. P. Schneider, K. L. Lear, and K. M. Geib, “Low threshold voltage vertical-cavity lasers fabricated by selective oxidation,” Electron. Lett. 30, 2043–2044 (1994). [CrossRef]
10. T. Numai, M. Sugimoto, I. Ogura, H. Kosaka, and K. Kasahara, “Surface-emitting laser operation in vertical-to-surface transmission electrophotonic devices with a vertical cavity,” Appl. Phys. Lett. 58, 1250–1252 (1991). [CrossRef]
11. C. C. Chang, H. P. Sardesai, and A. M. Weiner, “Code-Division Multiple-Access Encoding and Decoding of Femtosecond Optical Pulses over a 2.5-km Fiber Link,” IEEE Photon. Technol. Lett. 10, 171–173 (1998). [CrossRef]
12. D. Zaccarin and M. Kavehrad, “An optical CDMA system based on spectral encoding of LED,” IEEE Photon. Technol. Lett. 5, 479–482 (1993). [CrossRef]
