Search

This paper investigates the Contention Window (CW) optimization problem in multi-agent scenarios, where the fully cooperative among mobile stations is considered. A partially observable environment is employed to model and analyze the CW optimization problem, and Smart Exponential-Threshold-Linear with Deep Q-learning Network (SETL-DQN) Multi-Agent (MA) algorithm is proposed to obtain the optimal system throughput through the CW Threshold optimization. In the determined scenarios, SETL-DQN(MA) can effectively cope with the mutual interaction among mobile stations. The simulation results show that our proposed method is superior from both static and dynamic scenarios and has the highest optimum packet transmission efficiency.

In this paper, a routing method based on reinforcement learning (RL) under software-defined networks (SDN), namely the Q-learning widest-path routing algorithm (Q-WPRA), is proposed. This algorithm processes the reward function according to the link bandwidth in the execution environment to find the optimal (i.e., widest) transmission path with the maximum bandwidth between the source and the destination through RL. The experimental results reveal that the Q-WPRA is outperformance than Dijkstra’s algorithm and Dijkstra’s widest-path algorithm to find the widest transmission path in SDN environment under different bandwidths, loss rates, and background traffic.

The increasing complexity of modern day networked applications and the massive demand on the Internet resources has reignited interest and concern in the underlying networking infrastructures and their ability to cope with such complexity and adapt to the demands of the business applications, particularly where such applications require a high degree of robustness and reliability. As a result, software-defined networking has emerged as a promising approach to the definition of network architectures that could carry a high degree of adaptability and robustness reminiscent of the future Internet. Fault tolerance and network updates are considered two of the current research challenges that hamper the growth of software-defined networking in this area. Therefore, this paper represents a step toward tackling these two issues in the context of single link failures. Our main contribution lies in the definition of new algorithms that aim to enhance the problem of finding alternative paths in large-scale networks with minimal cost and time-to-update factors. The new solution aims at increasing the efficiency of flow operation reduction during link failures. We evaluate our framework and show how its implementation results in improved efficiency.

According to the latest version of the IEEE 802.16 standard, the mandatory contention resolution method is the truncated binary exponential backoff, with the initial window size and the maximum window size controlled by the base station. However, the problem of choosing the right set of backoff parameters for the current network level remains unsolved and left as an open issue since this strategy might incur a high collision probability and the channel utilization could be degraded in congested scenario. In this paper, we propose two pragmatic adaptive algorithms, namely semi-dynamic and quasi-dynamic contention resolution schemes, that allow the base station to adjust its backoff window size based on current channel status. By controlling the size of backoff window according to varying network conditions, both schemes are able to achieve higher performance in comparison with the legacy IEEE 802.16 standard.