Yazar "Sokullu, Radosveta Ivanova" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim Yok
    Öğe
    Deep Q-Learning based resource allocation and load balancing in a mobile edge system serving different types of user requests
    (Slovak Univ Technology, 2023) Yildiz, Onem; Sokullu, Radosveta Ivanova
    With the expansion of the communicative and perceptual capabilities of mobile devices in recent years, the number of complex and high computational applications has also increased rendering traditional methods of traffic management and resource allocation quite insufficient. Recently, mobile edge computing (MEC) has emerged as a new viable solution to these problems. It can provide additional computing features at the edge of the network and allow alleviation of the resource limit of mobile devices while increasing the performance for critical applications especially in terms of latency. In this work, we addressed the issue of reducing the service delay by choosing the optimal path in the MEC network, which consists of multiple MEC servers that has different capabilities, applying network load balancing where multiple requests need to be handled simultaneously and routing selection based on a deep- Q network (DQN) algorithm. A novel traffic control and resource allocation method is proposed based on deep Q-learning (DQL) which allows reducing the end-to-end delay in cellular networks and in the mobile edge network. Real life traffic scenarios with various types of user requests are considered and a novel DQL resource allocation scheme which adaptively assigns computing and network resources is proposed. The algorithm optimizes traffic distribution between servers reducing the total service time and balancing the use of available resources under varying environmental conditions.
  • Küçük Resim Yok
    Öğe
    DRONE BASED HOTSPOT NETWORK SYSTEM DESIGN
    (Ieee, 2019) Caliskan, Zafer; Sokullu, Radosveta Ivanova
    In this paper, a drone based system model and prototype is presented which allows mobile phone users to make Voice over Wi-Fi calls, a.k.a. VoWiFi, while the cellular system is not available or the cellular network signal level is under a pre-defined threshold to meet the quality of service for an IP call. the drone based system consists of a drone equipped with a smart phone main board which establishes a Wi-Fi hotspot network and the user equipment, smartphone capable of making a voice over Wi-Fi calls. the system performance is evaluated based on two major parameters: the Perceptual Evaluation of Speech Quality (PESQ) scoring, and the Received Signal Strength Indicator (RSSI). the relationship between these two important parameters is investigated in real life cellular networks under various conditions.
  • Küçük Resim Yok
    Öğe
    Mobility and traffic-aware resource scheduling for downlink transmissions in LTE-A systems
    (Tubitak Scientific & Technical Research Council Turkey, 2019) Yildiz, Onem; Sokullu, Radosveta Ivanova
    As new cellular networks support not only voice services but also many multimedia applications, the requirements for reliable data transmission at high speeds create heavy load on the system. Even though LTE/LTEA technology takes action towards alleviating this load, it is still necessary to manage resources effectively because of the inadequacy of the available radio resources. Thus, the scheduler at the MAC layer of the base station plays a very important role in resource allocation to the user. In this study a novel algorithm for resource allocation in mobile environments is presented, with two variations addressing different input traffic. The idealized case (I-MAS algorithm) relates to the full-buffer model, while the realistic case (R-MAS algorithm) takes into consideration the specific characteristics of the incoming user traffic. The paper includes performance evaluation of the suggested algorithms in terms of mean and edge throughput, system fairness, and BLER and comparison with well-known algorithms like the round robin (RR) and best CQI (B-CQI) (full-buffer model) and their extensions for real-life traffic models, RR Traffic and B-CQI Traffic, respectively. When the simulation results are examined, it can be seen that the I-MAS and R-MAS algorithms maximize the throughput while at the same time distributing the resources fairly among the users. They also prove to be quite robust in mobile environments even at higher user speeds.