Analysis of color image features extraction using texture methods

In this paper, a binary-heap based fair scheduling algorithm (BFSA) for optical burst switching (OBS) networks is proposed. BFSA calculates the cost function (CF) for every generated burst at edge nodes. CF is calculated using four fair parameters associated with each burst and it is updated at every hop until reaching the destination node. Moreover, a minimum priority queue is implemented using a binary heap tree for all incoming bursts at each code node. In contention case, the minimum cost burst that occupies the root node of the heap is considered as a victim and it is replaced by the new arriving high priority burst flow. BFSA is a fast scheduling scheme that is easily implemented at core nodes with no extra hardware components. BFSA is evaluated in terms of burst loss ratio and throughput using NS-2 simulator. Results reveal that BFSA is able to address fairness in OBS networks as it achieves better burst loss ratio and better throughput. BFSA performs much better than other fairness schemes such as the last hop pre-emption (LHP) and the fair prioritized pre-emption (FPP) schemes. In addition, it shows better fairness performance for higher priority traffic flows. Copyright © 2016 Praise Worthy Prize - All rights reserved.