Simulation of error detection and correction for ATM technology

Abstract: The rapid progress that is achieved in many fields of digital services and applications, especially in the area of audio-visual services and related applications, and the urgent need to communicate and transfer data at high speeds; all this led to the development of Asynchronous Transfer Mode (ATM) protocol. This Protocol is the adopted protocol by the Broadband-Integrated Services Digital Networks (B-ISDN) technology in transferring information through the network. Hence, this protocol is used to transfer the information at rates reach to Gb/s and to provide the special requirements for these services and applications, but, in transferring information at high rates may produce some errors belong to lack of network resources that are often influential in one way or another on the performance of applications. Therefore, the techniques of error detection and correction are of significance in this area and particularly in applications that are demanding high requirements (high transfer rate, low Bit Error Rate (BER), high Quality of Service (QoS) requirements, low Cell Loss Rate (CLR)...). Where, these techniques are: Reed-Solomon, Selective Repeat Automatic Repeat reQuest (ARQ), and Go-Back-N ARQ. Each one of these techniques of error detection and correction has its own properties, so, this thesis shows the difference between these properties and suitability each of them for this Protocol. Where, these techniques are applied to ATM environment through the establishment of simulation software. The simulation software consists of the Graphical User Interface (GUI) and the Network core parts. The GUI part is used to access the parameters of both the network and the techniques that are resided in the Network core part. The simulation results showed that:For network resources (buffers, and queues); the error correction technique has a linear effects on network resources related to nk values, while, the error detection techniques have a random effects on these resources because of windows size variation and CLR. For network throughput; the throughput of error correction technique is linearly affected by the k  n ratio. The window size, and CLR are the parameters that error detection technique′s throughput depend on. On QoS; to compensate more cell losses, larger nk values are used on error correction technique and that have a linear effects on both Cell Transfer Delay (CTD) and Frame Transfer Delay (FTD). Fluctuations are appeared on FTD as the variation on window size of error detection techniques. The error correction technique (Reed-Solomon) has linear effects on resources, throughput, and QoS of the network, so, error correction technique is more suitable for services demand for reality.