International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 65 - Number 17 |
Year of Publication: 2013 |
Authors: Sarkout N. Abdulla, Zainab T. Alisa, Arwaa Hameed |
10.5120/11015-6372 |
Sarkout N. Abdulla, Zainab T. Alisa, Arwaa Hameed . Parallel Computing for Detecting Processes for Distorted Signal. International Journal of Computer Applications. 65, 17 ( March 2013), 20-27. DOI=10.5120/11015-6372
Wireless devices such as hand phones and broadband modems rely heavily on forward error correction techniques for their proper functioning, thus sending and receiving information with minimal or no error, while utilizing the available bandwidth. Major requirements for modern digital wireless communication systems include high throughput, low power consumption and physical size. This research focused on the speed. The design of a four state convolutional encoder and Viterbi decoder has been studied and implemented. In order to solve the Viterbi decoding of lower speed problem, a Viterbi decode method has been parallelized. Message passing interface (MPI) with dual core personal computer (PC) and with cluster is selected as the environments to parallelize VA by distributing the states on the cluster and then by using block based technique. It is found that parallel Viterbi code when the states have been distributed on a number of computers provide poor performance; due to the communication overhead. So, in order to obtain better performance; it is suggested that to replace the multicomputer system (LAN with 100Mbps) with multiprocessor system (its speed tenth of Gbps). Block based parallelism can reach a maximum efficiency of 94. 25 % and speed-up of 1. 88497 when 2 PCs are used on a cluster with message length of 3000 bits. On 3 PCs, maximum efficiency of 95% and speedup of 2. 85 have been obtained with message length of 9000 bits. However; on 4 PCs; the system reaches maximum efficiency of 58. 333, maximum speed-up of 2. 33 with message length of 3000 bits.