在這一篇論文中，我們評估一個使用波長維度 (Wavelength dimension) 來解決競爭問題的緩衝式波分多工 (Wavelength Division Multiplexing, WDM) 光數據封包交換機的架構。在這個架構的每個輸出端都接續至共同的緩衝器和裝備可調式波長轉換器 (Tunable Wavelength Converters, TWCs) 來讓輸入端之間共同使用。結合波長維度的使用來解決競爭問題，這個方法不但降低了共用緩衝器的數目而且也使用較少波長轉換器的數量；因此，它是有成本效益的。根據這個架構，一個較低的數據封包遺失機率 (Packet loss probability)和較佳的網路流量是能夠達成。我們也可看到共用緩衝與波長轉換器式光交換機不但比無緩衝式光交換機使用較少的波長轉換器數目而且也比共用緩衝式光交換機使用較少的光延遲線 ( Fiber Delay lines)。
此外，我們分析網路流量的擁擠解決問題與如何去考慮一個WDM交換機中的光延遲線與波長轉換器的數目。我們的研究指出這個架構可以容忍較高網路擁塞並且指出了波長轉換器的最佳使用方式是將它共同使用。從數據上來看，當每一條光纖通道上傳輸的波長數大於等於10時，波長轉換器的最佳使用數量是波長數減一。但是在一般光交換機對於數據封包遺失機率的要求都在10-10之下，使用光延遲線變成是一個可以達到要求的方法，不過光延遲線也是有相當大的缺點。因此如何的去考慮一個WDM交換機中的光延遲線與波長轉換器的數目變成是在設計光交換機中的一個重大的課題。

A shared buffered WDM optical packet switch scheme employing the wavelength converters for contention resolution is proposed and evaluated. Based on the architecture, lower packet loss probability and better network throughput can be achieved. The architecture is dedicated to share buffer for each output and equipped with tunable wavelength converters (TWCs) shared among the input lines. Combined with the wavelength dimension for contention resolution, this reduces the size of each shared buffer and employs a lesser number of wavelength converters; therefore, it is cost-effective. We observe that shared buffered optical switch allows reductions on the employed number of converters compared with the bufferless architecture and number of fiber delay lines compared with shared-memory scheme.

Besides, we will not only evaluate throughput performance occurred with contention by simulation result but also to find the problem of the numbers between buffers and TWCs in a WDM Switch. In our research, the results show not only higher tolerant traffic load which means achieving high throughput on the architecture but also that the best utility rate of the converters is shared. Too more or fewer are not good for the switch design. From the data, the best choice of number of converters is (M-1) as the useful wavelength in each fiber channel is larger then 10. However, the good packet loss probability in WDM packet switch must be lower than 10-10, if we only use the converters to obtain the target, it is hard to complete. Therefore, combined with Shared buffers are a good way to reduce the packet loss probability, but its drawback is a huge size box and not in random choice. The numbers of fiber delay lines must be under the control which is as fewer as possible.

[1]. S. Yao, S. J. B. Yoo, B. Mukherjee and S. Dixit, "All-optical packet
switching for metropolitan area networks: opportunities and challenges," IEEE
Communications., Vol. 39, p. 142-148, Mar., 2001.

[2]. Gerd Keiser, “Optical Fiber Communications,” New York: McGraw-Hill, 2000.

[3]. Shun Yao, B. Mukherjee, and S. Dixit, “Advances in photonic packet
switching: an overview,” IEEE Communications Magazine, vol. 38, no. 2, pp.
84–94, Feb. 2000.

[4]. S. L. Danielsen, P. B. Hansen, and K. E. Stubkjear, “Wavelength conversion
in optical packet switching,” IEEE/OSA Journal of Lightwave Technology, vol.
16, no. 12, pp. 2095–2108, Dec. 1998.

[5]. V. Eramo and M. Listanti, “Packet loss in a bufferless optical WDM switch
employing shared tunable wavelegth converters,” IEEE/OSA Journal of
Lightwave Technology, vol. 18, no. 12, pp. 1818–1833, Dec. 2000.

[6]. K.-C. Lee and V. O. K. Li, “Optimization of a WDM optical packet switch with
wavelength converters,” in IEEE Infocom ’1995, pp. 423–430.

[7]. D. K. Hunter, M. C. Chia, and I. Andonovic, “Buffering in optical packet
switches,” IEEE/OSA Journal of Lightwave Technology, vol. 16, no. 12, pp.
2081–2094, Dec. 1998.

[8]. C. Guillemot et al., “Transparent optical packet switching: the European
ACTS KEOPS project approach,” IEEE/OSA Journal of Lightwave Technology, vol.
16, pp. 2117–2133, Dec. 1998.

[9]. L. Tancevski, A. Ge, and G. Castanon, “Optical packet switch with partially
shared buffers: design principles,” Proc. OFC ’01, vol. 2, pp. TuK3–1
–TuK3–3, Mar. 2001.

[10]. G. Castanon, L. Tancevski, and L. Tamil, “Routing in all-optical packet
switched irregular mesh networks,” Proc. IEEE Globecom ’99, pp.
1017–1022, Dec. 1999.

[11]. F. Forghierri, A. Bononi, and P. R. Prucnal, “Analysis and comparison of
hot-potato and single-buffer deflection routing in very high bit rate
optical mesh networks,” IEEE Transactions on Communications, vol. 43, no.
1, pp. 88–98, Jan. 1995.

[12]. F. Brogonovo, L. Fratta, and J. Bannister, “Unslotted deflection routing in
all-optical networks,” Proc. IEEE Globecom ’93, vol. 1, pp. 119–125,
1993.

[13]. S. Yao, B. Mukherjee, S. J. B. Yoo, and S. Dixit, "A unified study of
contention-resolution schemes in optical packet-switched networks," IEEE
Journal of Lightwave Technology, Mar., 2003

[14]. S. J. B. Yoo, C. Caneau, R. Bhat, M. A. Koza, A. Rajhel, and N. Antoniades,
“Wavelength converion by difference-frequency generation in AlGaAs
waveguides with periodic domain inversion achieved by wafer bonding,”
Appl. Phys. Lett., vol. 68, pp. 2609–2611, 1996.

[15]. S. J. B. Yoo, “Reduced parametric wavelength-interchanging crossconnect
architectures with scalability and modularity,” Proc. OFC ’98, vol. 2, pp.
59–61, Mar. 1998.

[16]. M. J. Karol, “Shared-memory optical packet (ATM) switch,” in Proc.
SPIE: Multigigabit Fiber Commun. Syst., 1993, vol. 2024, pp. 212–222.

[17]. M. G. Hluchyj and M. Karol, “Queuing in high performance packet witching,”
IEEE J. Select. Areas Commun., vol. 6, no. 9, pp.1587-1597, Dec.1988.