Baoling Liu

Energy efficient cooperative relaying and cognitive radio technologies to deliver green communication

So far, achieving high data rate has been the primary focus of research in wireless communication systems, without much consideration of energy efficiency. However, many of communication technologies significantly increase system complexity and energy consumption. In addition to the environmental problems that the governments restrict the greenhouse gases and pollution, energy costs also contribute a significant part of network operators overall expenditures due to the rapidly growing demand for mobile communication. Therefore, energy saving approaches are urgently required by both governments and network venders. Escalating energy costs and environmental concerns have already created an urgent need for more energy-efficient wireless communication. To address the challenge of increasing power efficiency in wireless networks and thereby to maintain profitability, this paper will discuss some research issues and challenges and suggest some paradigm-shifting technologies to enable an energy efficient or green cellular network.

Discrete location-aware power control for D2D underlaid cellular networks

Device-to-device (D2D) communication is a promising method to reduce power consumption and improve the throughput of cellular networks. However, densely deployed D2D pairs could result in severe interference to cellular users without proper power control. Therefore, the discrete location-aware power control (DLPC) scheme is proposed in uplink D2D underlaid cellular networks. The entire cell area is divided into several regions, and a total power budget for each region is conducted with weighted allocation, to meet the constraint on the outage probability of cellular user. Then DLPC scheme requires only the locations of active D2D pairs rather than channel information or massive calculations, which is of low complexity in implementation. Simulation indicates that DLPC can improve the outage probability of D2D pairs and the network throughput compared with the traditional greedy power control scheme.

Utility-Oriented Resource Allocation Scheme in LTE-WLAN Heterogeneous Networks

In this paper, a utility-oriented resource allocation scheme (UORA) is proposed to improve network efficiency for LTE-WLAN heterogeneous networks. Isolated resources such as OFDMA subcarriers and WLAN time slots, which are always ignored in existing schemes, are considered in the proposed UORA scheme to give guidance on engineering practice. Aiming at improving user satisfaction, UORA scheme also considers various traffic characteristics and dynamic transmission rate. To guarantee high overall average level of user QoS, the resource allocation problem is formulated as an constrained optimization problem to maximize the total user utility but not total transmission rate. Based on optimization theory, a heuristic iterative algorithm is constructed to solve the problem. Numerical simulation is also adopted to prove the performance enhancement of our proposed UORA scheme compared with previous ones.

Parameters Optimization for KFKM Clustering Algorithm Based on WiFi Indoor Positioning

Kernel fuzzy K-means (KFKM) clustering algorithm is widely used to manage the fingerprint database for WiFi indoor positioning system to reduce the computational complexity of the position matching process. In this paper, we propose a novel WiFi positioning scheme based on KFKM algorithm, which can achieve a better precision by further optimizing the parameters employed in KFKM. Our proposed scheme consists of three steps. First, we choose an interval of reference points (RP) to build the fingerprint database. Then we decide an appropriate number of clusters based on the structure characteristics of fingerprint database using sample density method. During the process of clustering, we optimize the kernel parameter by approximating actual kernel matrix to a hypothetical ideal kernel matrix to improve the positioning precision. Through simulation results, we show that compared with the existing KFKM algorithm, our proposed scheme achieves 23.48% improvement in terms of positioning accuracy.

An Efficient Multi-service Resource Scheduling Algorithm in Cognitive Relay System

Relay is one of the key techniques in LTE-Advanced, which can expand the network coverage, increase the system capacity and improve the cell-edge performance and the fairness among users. Based on the current resource scheduling researches with relay technology, this paper investigates a resource scheduling algorithm called QoS Quality of Service-based Two Hop Proportional Fair QTHPF, which can efficiently guarantee the multi-service QoS performance in a cognitive LTE-Advanced relay system. Our basic idea is to establish the optimized frequency lists corresponding to different services and to introduce the multi-service utility function to calculate scheduling priority. Extensive simulation results confirm that the proposed algorithm improves the fairness, QoS performance and system throughput, and reduces the channel switching frequency.

A Dynamic Channel Allocation Algorithm in Random Access Process for LTE System

Random access technology is necessary in media access control of telecommunication system. Its function is directly related to the performance of the whole system. How to use time and frequency resource of LTE system effectively and reduce access delay as far as possible is the focus of this paper. This paper proposes a dynamic random access channel allocation algorithm for competitive conditions. The algorithm focuses on improving the utilization of time and frequency resource in LTE system and avoiding waste caused by idle resource.

A Distributed User Association Algorithm in Heterogeneous Cellular Networks

In this paper, we study the user association scheme in a heterogeneous network HetNets, attempt to resolve the problem of interference avoidance and load balancing in HetNets. We formulate a logarithmic objective function, to reduce the utility of heavy load base station, and present a distributed load-aware association algorithm, which encourage users to associate the more lightly loaded stations. User will connect to low-power station e.g., pico or femto cell, even if they offer a lower instantaneous SINR than the macro cell base station. We observe that users are stimulated by our scheme to associate with low-load, low-power base station. Simulation results support our expectations: in the proposed algorithm achieves a good load balancing between macro and femto cells, and increase the user average throughput. In addition, compare with Max-SINR algorithm, traditional algorithm, our distributed Load-Aware algorithm has a low complexity.

Social-Based Hybrid Dissemination Strategy in Heterogeneous Opportunistic Networks

Opportunistic network is a special class of Mobile Ad-hoc Networks (MANETs) in which connectivity between nodes is intermittent. Nodes in opportunistic network transmit data in store-carry-and-forward paradigm. Data dissemination usually processes in publish/subscribe model. A hybrid dissemination mechanism is proposed in this paper which depends on the phenomenon that nodes within community may always act as the nodes in MANETs. This solution maintains an Ad-Hoc sub-network in communities and thus provides high reliability, while other parts communicate in opportunistic approach. To achieve the efficient data transmission, a community heads solution is devised to integrate subscriptions and schedule data transmission direction. Meanwhile, relationship between nodes and communities is used to select relays for forwarding delegation. The performance of this approach is verified by experimental simulation.

A contact-history based routing for publish-subscribe scheme in hierarchical opportunistic networks

Opportunistic networks is aimed to provide a solution for communication in disrupt tolerate networks, which has attracted lots of researches. However, for reasons such as finite buffer size and energy of nodes in opportunistic networks, the further improvement of network performance is limited. Therefore, we set powerful cluster nodes in the network to form hierarchical opportunistic networks. In this paper, a publish/subscribe scheme in hierarchical opportunistic networks is put forward. The object of publish/subscribe scheme is to realize the sharing of resources between nodes. In our scheme, nodes disseminate their data requests with a TTL value to limit their flooding and after receiving the subscribed data items, an ACK will be flooded to delete the useless data packets in the network. To choose appropriate relays for data packets, a utility function to estimate the relay ability of nodes is designed, whats more, cluster nodes are brought in as relay nodes too. Simulations are given to demonstrate the performance improvement of our scheme.