LTE (Long Term Evolution) is a broadband wireless communication specified by the 3GPP (Third Generation Partnership Project) on the way towards fourth-generation mobile. User traffic can be classified into two main classes which are RT (Real Time) and NRT traffic. To respond to RT's QoS requirements, resource allocation steps must be as optimal as possible. In this paper, we propose a new concept of estimation of reception time to avoid scheduling packets that may eventually be dropped inside their destination. The main objective of this proposition is to reduce radio blocks' wasting. The behavior of the scheduler has been simulated to check its performance. Simulations prove that the proposed scheme overcome existed schedulers.
LTE (Long Term Evolution) is a broadband wireless communication specified by the 3GPP (Third Generation Partnership Project) on the way towards fourth-generation mobile. User traffic can be classified in two main classes which are RT (Real Time) and Non-RT applications. Many schedulers (in both Uplink and Downlink) have been proposed for both types of traffics. In this paper, we intend to present a new strategy of downlink scheduling. It unites both types of schedulers. Our main object is to make our scheduler effective for RT applications as well as for Non-RT submissions. Our main contribution is that we estimate the time needed for a real time packet to be successfully received by its destination. This was basically done by including a new parameter called Approximated Reception Time.
The importance of Long Term Evolution-Advanced (LTE-A) comes from its support to Real Time (RT) applications. A robust scheduler is crucial to guarantee advanced services. Despite variations, in terms of LTE-A downlink schedulers, time sensitive applications are not yet satisfied. In fact, the requests to access LTE-A services are enormous and it exceed the network capabilities. Real time flows must be satisfied in term of Quality of Service (QoS) requirements. Delay is a key feature that should be considered by a scheduler. The RT flows must be scheduled before their deadline. Otherwise, they will be dropped. In this paper, we propose a novel LTE-A downlink scheduler based on delay-PLR (Packet Loss Ratio) prediction. It aims at providing a balance between low delays and acceptable PLR for the RT traffic. We evaluate our proposal by launching several simulations. Further, we compare the captured results with schedulers from the literature. Results show that our proposal do better than the benchmarks we considered and it guarantees the QoS of the Non RT (NRT) flows.