Hopfield Neural Network - Based Approach for Joint Dynamic Resource Allocation in Heterogeneous Wireless Networks
This paper presents a comprehensive approach to solve the problem of joint dynamic resource allocation (JDRA) in heterogeneous wireless networks using a Hopfield neural network (HNN). A generic formulation for packet services with delay constraints is proposed to decide the optimal bit rate and radio access technology (RAT) allocation. Some illustrative simulations results in a basic scenario are presented to evaluate performance of the proposed algorithm.
Policy-based channel access mechanism selection for QoS provision in IEEE 802.11e
The appearance of the 802.11e standard made the 802.11 wireless local area networks (WLANs) capable of providing different levels of quality of service (QoS). This standard defines two new channel access mechanisms (CAMs), namely the enhanced distributed channel access (EDCA) and the HCF controlled channel access (HCCA). The introduction of two CAMs increases the complexity of traffic management, since now not only the available resources have to be distributed, but also the optimal CAM for each data flow has to be selected. However, the best CAM selection algorithm is still not clear. This paper presents and analyses different CAM selection policies in order to derive the best-suited strat…
User bandwidth usage-driven HNN neuron excitation method for maximum resource utilization within packet-switched communication networks
Mobile and wireless systems beyond 3G are being designed under the user-centric paradigm. Dynamic resource allocation (DRA) is a topic on intensive research to address efficiently such paradigm. Hopfield neural networks (HNN) have proved useful in the past to solve this kind of complex optimization problems. Recently, various approaches have been proposed to realize HNN-based user-centric DRA. However, the initial algorithms suffer from severe instability problems impacting the overall performance. This letter analyses the source of the existing limitations and proposes an enhanced formulation, ensuring maximum resource utilization while optimizing the convergence of the neural network. The…
HSDPA Link Adaptation Improvement Based on Node-B CQI Processing
In this paper HSDPA link adaptation (LA) based on Channel Quality Indicator (CQI) reports is optimised. A pre-processing of the last received CQI reports is done before the execution of the LA algorithm in the Node-B in order to obtain more profitable channel quality estimations and hence improve the LA performance. Different types of processing techniques are presented and assessed, considering from the simplest sample averaging to some more elaborated predictive algorithms. Results demonstrate that a non negligible enhancement in the LA performance can be obtained if medium and high speed users are considered.
A Delay-Centric Dynamic Resource Allocation Algorithm for Wireless Communication Systems Based on HNN
This paper proposes a dynamic resource-allocation (DRA) algorithm for packet data services in wireless communication systems based on Hopfleld neural networks (HNNs). The resource-allocation algorithm assumes a delay-centric approach in that it maximizes resource utilization of the overall system while minimizing the packet delay. The real-time (RT) working capability of HNN hardware implementation means that a very powerful scheduling DRA algorithm can be obtained. A generic formulation of the algorithm is presented to establish the optimal bit rate allocation. In addition, some illustrative examples of this formulation are given, considering specific wireless communication systems, such a…
Joint Dynamic Resource Allocation for Coupled Heterogeneous Wireless Networks Based on Hopfield Neural Networks
This paper proposes an algorithm to solve the problem of Joint Dynamic Resource Allocation in heterogeneous wireless networks. The algorithm is based on Hopfield Neural Networks to achieve fast and suboptimal solutions. The generic formulation is particularized and evaluated in an HSDPA and 802.11e WLAN coupled networks. Some illustrative simulations results are presented to evaluate the performance of the new algorithm as compared with other strategies. The obtained results confirm the validity of the proposal.
Complete Shadowing Modeling and its Effect on System Level Performance Evaluation
Computer simulations are a common procedure for assessing the performance of new algorithms. To conduct a valid and accurate study, the models employed in such simulators need to be carefully selected. Regarding shadowing modeling, one-dimensional models are fairly commonplace in the literature. While simple and with low computational costs, these models can not produce correlated fading values for mobiles that are in nearby positions and, besides, do not include the cross-correlation effect. To overcome these limitations, this paper presents a bi-dimensional shadowing model which introduces both the spatial correlation and the cross-correlation present in real systems. Finally, the impact …