在現今製程快速演進的背景之下，系統單晶片(system on chip, SoC)的技術也隨之快速進步，使得晶片在原有面積之下被整合了更多功能模組在裡面。目前廣為人知的系統單晶片整合平台為ARM (Advanced RISC Machines)公司所提出的AMBA (Advanced Microcontroller Bus Architecture)匯流排架構，此平台受到眾多使用者喜愛的主因是其在面積成本很低的前提下還能提供不錯的效能表現，但隨著製程的演進，其架構開始面臨擴充能力不足的問題。因此，有人開始陸陸續續提出改善的方案，本篇論文所探討的即為我們所提出的一種新型結構，用來改善原有匯流排(bus)的缺陷。
由於新型結構的擴充方式類似網路(network)結構，因此在設計上必須考慮許多網路結構上交換器(switch)會面臨的問題。在這邊主要探討的是拓撲結構與功能模組位置規劃對整體性能表現的影響，討論的項目有換手次數、輸出率、平均延遲與面積等等，透過這些討論，比較分析不同拓撲結構上的差異，在選擇擴充結構時才能有實驗數據作為參考。

With the advances of the semiconductor process, the technology of system on chip (SoC) progresses rapidly. Therefore, the original area of one chip was integrated with more functional modules. At present, the well-known SoC platform which is proposed by ARM (Advanced RISC Machines) Company is the AMBA (Advanced Microcontroller Bus Architecture) bus framework. The main reason why this platform is popular with many people is that it can provide good efficiency with lower cost of the size. However, as the process evolves, its structure encounters the problem of insufficient expansion capabilities. Consequently, some people started to propose plans to improve this situation one after another. In this paper, we proposed a new structure to improve the defects of the original bus.
The expansion of the new structure is similar to the network structure, so we must take many problems of the switches on the network structure into consideration. We mainly explore the topology and functional modules planning that impact on the system performance. The items discussed in the paper include average hops, throughput, latency and area. Through these discussions and comparative analysis between different topology, we can use the experimental data as a reference to expand the structure.

[1]S. Borkar, “Obeying Moore’s law beyond 0.18 micron,” in Proc. ASIC/SOC Conf., pp. 26-31, Sept. 2000.
[2]R. Ho et al., “The future of wires,” Proc. IEEE, vol. 89, no. 4, pp. 490-504, Apr. 2001.
[3]“Core-Connect Bus Architecture,” IBM, 1999.
[4]“Wishbone Bus,” Open-Cores, 2003.
[5]“Avalon Interface Specifications,” Altera, 2011.
[6]“AMBA Specification,” ARM, 1999.
[7]L. Benini and D. Bertozzi, “Network-on-chip architecture and design methods,” IEE Proc.-Comput. Digit. Tech., vol. 152, no. 2, pp. 261-272, Mar. 2005.
[8]C. Grecu et al., “Structured interconnect architecture: A solution for the non-scalability of bus based SoCs,” Proc. GLSVLSI, pp. 192-195, Apr. 2004.
[9]D. Sylvester and K. Keutzer, “Impact of small process geometries on microarchitectures in systems on a chip,” Proc. IEEE, vol. 89, no. 4, pp. 467-489, Apr. 2001.
[10]M. Drinic et al., “Latency-guided on-chip bus network design,” IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst., vol. 25, no. 12, pp. 2663-2673, Dec. 2006.
[11]K. Lahiri et al., “Efficient exploration of the SoC communication architecture design space,” in Proc. IEEE/ACM Int. Conf., pp. 424-430, Nov. 2000.
[12]K. Sekar et al., “Dynamically configurable bus topologies for high-performance on-chip communication,” IEEE Trans. Very Large Scale Integr.(VLSI) Syst., vol. 16, no. 10, pp. 1413-1426, Oct. 2008.
[13]R. Lu and C.K. Koh, “A high performance bus communication architecture through bus splitting,” in Proc. Design Automation Conf., pp. 751-755, Jan. 2004.
[14]C.T. Hsieh and M. Pedram, “Architectural energy optimization by bus splitting,” IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst., vol. 21, no.4, pp. 408-414, Apr. 2002.
[15]P. P. Pande et al., “Performance evaluation and design trade-offs for network-on-chip interconnect architectures,” IEEE Trans. Comput., vol. 54, no. 8, pp. 1025-1040, Aug. 2005.
[16]J.Y. Kim et al., “A 118.4 GB/s multi-casting network-on-chip with hierarchical star-ring combined topology for real-time object recognition,” IEEE J. Solid-State Circuits, vol. 45, no. 7, pp. 1399-1409, July 2010.
[17]J. Kim et al., “Performance analysis of NoC structure based on star-mesh topology,” in Int. SoC Design Conf., pp. 162-165, Nov. 2008.
[18]K.J. Chen et al., “Star-type architecture with low transmission latency for a 2D mesh NoC,” in IEEE Asia Pacific Conf., pp. 919-922, Dec. 2010.
[19]C. Jueping et al., “Integrated modeling generation and optimization for packet based NoC topology,” in IEEE Int. Conf., pp. 293-298, Apr. 2010.
[20]W. Chen et al., “Heterogeneous design methodology with configurable regular topology set for scalable network-on-chip designs,” in IEEE Int. Conf., pp.1293
-1296, Oct. 2007.
[21]F. Davik et al., “IEEE 802.17 resilient packet ring tutorial,” IEEE Commun. Mag., pp. 112-118, Mar. 2004.
[22]S.J. Lee et al., “An 800MHz star-connected on-chip network for application to systems on a chip,” in IEEE Int. Solid-State Circuits Conf., pp. 468-469, Feb. 2003.
[23]S. Kumar et al., “A network on chip architecture and design methodology,” in Proc. IEEE Computer Society Annual Symposium, pp. 117-124, Apr. 2002.
[24]W. J. Dally and B. Towles, “Route packets, not wires: On-chip interconnection networks,” in Proc. Design Automation Conf., pp. 683-689, June 2001.
[25]F. Karim et al., “An interconnect architecture for networking systems on chips,” IEEE Micro, vol. 22, no. 5, pp. 36-45, Sept.-Oct. 2002.
[26]S. Murali and G. D. Micheli, “Bandwidth-constrained mapping of cores onto NoC architectures,” in Proc. Design Automation Conf., pp. 896-901, Feb. 2004.