FlexRay為新一代的車用網路協定，車用網路是用來連接汽車內部電子控制單元並進行資料傳輸及控制，支援時間觸發(Time-Triggered)的架構來滿足新的X-by-wire應用需求，時間觸發的訊號具有週期性及可預測性的特性，需要確定性的演算法來控制訊號在預先定義的時間進度傳輸，以確保滿足訊號的即時要求。由於FlexRay原本限制所有static slots依序分配給各個節點在每個通訊週期重複使用，藉此來增加訊號傳送時的可靠度以及確保安全性，但此設計會造成過多的static slots未使用而形成浪費。針對此問題，我們提出一個即時排程的機制，將其應用於FlexRay的static segment，使得即時性及可靠性的要求仍能滿足的條件下，有效地降低static slots的使用需求量。我們使用real-time task model當作車用網路的即時系統模型，每個task ti包含兩個參數Ci,Ti，表示了該task的即時需求，亦即每Ti個通訊週期內必須分配至少Ci個static slots給ti。我們將通訊週期中的每一個static slots視為個別的channel，即時排程將task分解成多個subtasks以利排程但其總和仍能滿足原task的即時需求，爾後將這些subtasks分配至不同的channel中，使每個task會滿足distance constraint的特性，較原本的real-time task更加嚴謹，保證了在時間控制上的可靠性。每個channel中再配合rate-monotonic的排程，最後整個形成一套多channel固定週期循環且具確定性的排程，能夠事先決定每個訊號的傳輸。我們利用電腦模擬實驗來證實我們的即時排程在不同即時性的需求下，都能有效減少static slots的設置。

FlexRay is a new automotive network communication protocol for control and interconnection among ECUs (electronic control units) in the cluster. In the FlexRay protocol, a communication cycle consists of static segment and dynamic segment. The static segment is a TDMA scheme designed for transmitting time-triggered messages. Due to its determinism and reliability, it is particularly applicable to X-by-wire applications. Each static slot is allocated to a specified task and the task can transmit message during the exclusive slot. However, if the task has no message to transmit during its assigned slot, the slot cannot be used by other tasks. The overall utilization is low if the bandwidth requirement of each task is not high. To improve the system utilization, we apply the real-time scheduling techniques to devising a deterministic, static cyclic scheduling. The objective is to reduce the demand on the number of static slots needed for scheduling time-triggered tasks. Specifically, we treat the set of static slots that are in the same position in every communication cycle as an individual real-time channel. We model each task as a real-time task, specified by (Ci,Ti). It requires that for every Ti communication cycles, the system must allocate at least Ci time slots to satisfy the real-time constraint of the task. We decompose each such task into a set of subtasks, allocate them to the real-time channels and then apply the rate-monotonic scheduling algorithm to schedule the subtasks within each channel. Finally, we perform computer simulation to evaluate the effectiveness of our proposal. From the simulation results, we conclude that our proposal is able to effectively reduce the demand for the static slots under a wide range of real-time requirements.

第1章 導論 1
1.1 概論 1
1.2 研究動機 1
1.3 章節介紹 2
第2章 研究背景 3
2.1 車用電子網路協定 3
2.2 FLEXRAY網路協定 5
2.2.1 FlexRay的特性 5
2.2.2傳輸機制 6
2.3 REAL-TIME SCHEDULING 8
2.3.1 Rate-Monotonic及定理 8
2.3.2 Distance-Constrained 10
2.3.3 Specialization operation Sr 11
第3章 REAL-TIME SCHEDULING的實作與設計 14
3.1 REAL-TIME SCHEDULING在FLEXRAY的基本架構 14
3.2 REAL-TIME SCHEDULING機制 15
3.2.1 Specialization operation 17
3.2.2 Decomposition of Tasks 20
3.2.3 Allocation of subtasks 21
3.2.4 Slot Assignment scheme 22
3.3 即時排程演算法 24
第4章 模擬研究及實驗結果 28
4.1 實驗設定 28
4.2 實驗結果 29
4.2.1 Real-time scheduling實驗: utilization U和maximum Ti 29
4.2.3 Real-time scheduling實驗: parameter x of Sr 32
4.3 REAL-TIME SCHEDULING實驗探討 35
第5章 結論與未來工作 38
5.1 結論 38
5.2 未來工作方向 38
附錄 40
參考文獻 41

[1] FlexRay Consortium, FlexRay Communications System Protocol Specification, version 2.1 revision A, 2004-2005.
[2] Road Vehicles–Interchange of Digital Information–Controller Area Network (CAN) for High-Speed Communications, International Standards Organization (ISO). ISO Standard-11898, Nov. 1993.
[3] The Local Interconnect Network Consortium, http://www.lin-subbus.org/.
[4] MOST Specification v2.4, MOST Cooperation, http://www.mostnet.de, August 2005.
[5] H. Kopetz, A. Ademaj, P. Grillinger, and K. Steinhammer, “The Time-Triggered Ethernet (TTE) Design,” in ISORC’05: Proc. of the Eighth IEEE International Symposium on Object-Oriented Real-Time Distributed Computing(ISORC’05). Washington, DC, USA: IEEE Computer Society, pp. 22–33, 2005.
[6] S. Ding, N. Murakami, H. Tomiyama, and H. Takada, “A GA-Based Scheduling Method for FlexRay Systems,” in Proc. of EMSOFT, 2005.
[7] G. Cena and A. Valenzano, “Performance Analysis of Byteflight Networks,” in Proc. of the IEEE International Workshop on Factory Communication Systems, pp. 157–166, 2004.
[8] M. Schmid, “Automotive Bus Systems,” Atmel Applications Journal, Issue 6, pp. 29-32, Winter 2006.
[9] C. L. Liu and J. W. Layland, “Scheduling Algorithms for Multiprogramming in a Hard Real-Time Environment,” Journal of ACM, Vol. 20, No. 1, pp. 46–61, January 1973.
[10] A. Burchard, J. Liebeherr, Y. Oh, and S. H. Son, “New Strategies for Assigning Real-Time Tasks to Multiprocessor Systems,” IEEE Trans. on Computers, Vol. 44, Issue 12, pp. 1429–1442, Dec. 1995.
[11] M. Joseph and P. Pandya, “Finding Response Times in a Real-Time System,” The Computer Journal, British Computer Society, Vol. 29(5), pp. 390–395, October 1986.
[12] J. Lehoczky, L. Sha, and Y. Ding. “The Rate Monotonic Scheduling Algorithm: Exact Characterization and Average Case Behavior,” in Proc. of the Real-Time Systems Symposium, pp. 166–171, Santa Monica, CA, Dec. 1989.
[13] C.-C. Han and H. Y. Tyan, “A Better Polynomial-Time Schedulability Test for Real-Time Fixed-Priority Scheduling Algorithms,” in Proc. IEEE 18th Real-Time Systems Symposium, pp. 36–45, Dec. 1997.
[14] C.-C. Han, K.-J. Lin, and C.-J. Hou, “Distance-Constrained Scheduling and its Applications to Real-Time Systems,” IEEE Trans. on Computers, Vol. 45, Issue 7, pp. 814–826, July 1996.
[15] C.-C. Han, “Scheduling Real-Time Computations with Temporal Distance and Separation Constraints and with Extended Deadlines,” PhD thesis, University of Illinois at Urbana-Champaign, 1992.
[16] C.-C. Han and K.-J. Lin, “Scheduling Distance-Constrained Real-Time Tasks,” in Proc. IEEE 13th Real-Time Systems Symposium, pp. 300-308, Dec. 1992.
[17] H. Y. Tyan, C.-J. Hou, B. Wang, and C.-C. Han, “On Supporting Temporal Quality of Service in WDMA-Based Star-Coupled Optical Networks,” IEEE Trans. on Computers, Vol. 50, Issue 3, pp. 197-214, March 2001.