OPNET Models Research Articles

Simulation Research on EDF Scheduling Algorithm of CAN Network Based on OPNET

In order to analyze and evaluate real-time performance of CAN network, the OPNET models of CAN network with Fixed Priority Scheduling (FPS) and Earliest Deadline First (EDF) have been set up in this paper. Based on simulation experiments, the performances of average delay and transmission success ratio in the two models are analyzed and the experiments also indicate that under normal load condition the two scheduling algorithm both can send data frame successfully before the deadline. In a slight overload condition FPS algorithm makes the transmission delay of low priority data frame significantly increased, and even occurs the phenomenon that low priority data frame can not be sent at all times; while the EDF algorithm can make all the nodes to complete the data frames in less than 3ms transmission delay, it meets the delay requirements that are delay should less than 10ms in most control systems, it indicates that EDF algorithm has the better scheduling and availability in soft real-time system.

Development of an Interface Between the IEEE 802.11b WLAN Technology and the New Mobile ad hoc Subnetwork Relay SNR) Technology

In this paper, we investigate the development of an interface between the two following ad hoc networking technologies: IEEE 802.11b and the subnetwork relay (SNR) technology, which is the only fully distributed multi hop wireless ad hoc networking technology available on the market today. The newly formed hybrid protocol could be adapted to both the civilian and military domains and could offer new wireless ad hoc applications in domains not explored yet. Three integration approaches are discussed and technically compared. The preferable approach, which consists of the SNR protocol above the physical layer of 802.11b, seems the most promising. The following work, that has already started, consists in building Opnet models of each approach in order to further explore and analyse their behaviour.