As the internet is expected to better support many applications such as multimedia with limit bandwidth, new mechanisms are needed to control the congestion in the network. Active Queue Management (AQM) algorithms play an important role to ensure the stability of the Internet. Random Early Detection (RED) is the first active queue management algorithm proposed for deployment in TCP/IP networks. RED has some parameters tuning issues that need to be carefully addressed for it to give good performance under different network scenarios.We propose a new algorithm called Enhanced Random Early Detection (ENRED). ENRED works to improve these parameters to provide better congestion control over the network while remaining the advantage of RED.This paper will introduce ENRED and some features about RED and its variants. We simulate the proposed algorithm (ENRED) using the well-known network simulator ns-2, by comparing it to the original RED. Simulation results show that the proposed algorithm achieves better queue size than RED and decreases the delay and losses.

No abstract

No abstract

Beamforming design is a crucial stage in millimeter-wave systems with massive antenna arrays. We propose a deep learning network for the design of the precoder and combiner in hybrid architectures. The proposed network employs a parametric rectified linear unit (PReLU) activation function which improves model accuracy with almost no complexity cost compared to other functions. The proposed network accepts practical channel estimation input and can be trained to enhance spectral efficiency considering the hardware limitation of the hybrid design. Simulation shows that the proposed network achieves small performance improvement when compared to the same network with the ReLU activation function.

No abstract