Hung-Chin Jang

Applications of Geometric Algorithms to Reduce Interference in Wireless Mesh Network

In wireless mesh networks such as WLAN (IEEE 802.11s) or WMAN (IEEE 802.11), each node should help to relay packets of neighboring nodes toward gateway using multi-hop routing mechanisms. Wireless mesh networks usually intensively deploy mesh nodes to deal with the problem of dead spot communication. However, the higher density of nodes deployed, the higher radio interference occurred. This causes significant degradation of system performance. In this paper, we first convert network problems into geometry problems in graph theory, and then solve the interference problem by geometric algorithms. We first define line intersection in a graph to reflect radio interference problem in a wireless mesh network. We then use plan sweep algorithm to find intersection lines, if any; employ Voronoi diagram algorithm to delimit the regions among nodes; use Delaunay Triangulation algorithm to reconstruct the graph in order to minimize the interference among nodes. Finally, we use standard deviation to prune off those longer links (higher interference links) to have a further enhancement. The proposed hybrid solution is proved to be able to significantly reduce interference in a wireless mesh network in O(n log n) time complexity.

A hybrid knowledge representation as a basis of requirement specification and specification analysis

A formal requirement specification language, the frame-and-rule oriented requirement specification language FRORL, developed to facilitate the specification, analysis, and development of a software system is presented. The surface syntax of FRORL is based on the concepts of frames and production rules that may bear hierarchical relationships to each other, relying on multiple inheritance. To provide thorough semantic foundations, FRORL is based on a nonmonotonic variant of Horn-clause logic. Using the machinery of Horn-clause logic, various properties of a FRORL specification can be analyzed. Among the external properties of FRORL are formality, object-orientedness, and a wide spectrum of life cycle phases. Intrinsic properties are modularity, provision for incremental development, inheritance, refinement, reusability, prototyping, and executability. A software development environment based on FRORL has been implemented using the C language on a Sun workstation. >

Knowledge-Based Software Development for Real-Time Distributed Systems

The interplay of artificial intelligence and software engineering has been an interesting and an active area in research institution and industry. This book covers the state of the art in the use of knowledge-based approaches for software specification, design, implementation, testing and debugging. Starting with an introduction to various software engineering paradigms and knowledge-based software systems, the book continues with the discussion of using hybrid knowledge representation as a basis to specify software requirements, to facilitate specification analysis and transformation of real-time distributed software systems. A formal requirements specification language using non-monotonic logic, temporal logic, frames and production systems for new software engineering paradigms (such as rapid prototyping, operational specification and transformational implementation) is also discussed in detail. Examples from switching and other applications are used to illustrate the requirements language. Finally, the development, specification and verification of knowledge-based systems are investigated.

Network Status Detection-Based Dynamic Adaptation of Contention Window in IEEE 802.11p

In recent years the research on Vehicular Ad Hoc Network (VANET) has been paid much attention due to the development of wireless network becoming mature, people think highly of traffic safety issue, the popularity of Global Positioning System (GPS), promising applications of VANET being exploited. However, VANET has the essential property of rapid change of large scale network topology. Besides, the vehicles distributed over the road usually causes serious signal collisions and interferences. Thus, it calls for special concern to deal with the performances of throughput, collision and delay. In this paper we propose a Detection-Based MAC mechanism by modifying RTS/CTS to detect network congestion through message exchange and predict number of competing nodes. Both these information is used to dynamically adjust contention window size. Simulations focus on the MAC layer of IEEE 802.11p and the results show that the proposed method is able to effectively increase overall throughput, reduce packet collision rate and delay time.

QoS-constrained resource allocation scheduling for LTE network

Long Term Evolution (LTE) is the most promising technology for 4G mobile communication networks. The goal of LTE is to provide high data transmission rate, scalable bandwidth, low latency, and high-mobility. LTE employs Orthogonal Frequency Division Multiplexing (OFDM) and Single Carrier - Frequency Division Multiple Access (SC-FDMA) for downlink and uplink data transmission, respectively. As to SC-FDMA, there are two constraints in resource allocation. First, the allocated resource blocks (RB) should be contiguous. Second, these allocated resource blocks are forced to use the same modulation technique. This paper focuses on the resource allocation problem of LTE SC-FDMA system. The aim of this paper is to propose a QoS-constraint resource allocation scheduling to enhance data transmission on uplink SC-FDMA. The proposed scheduling is a three-stage approach. In the first stage, it uses a time domain scheduler to differentiate UE services according to different QoS service requirements. In the second stage, it uses a frequency domain scheduler to prioritize UE services based on channel quality. In the third stage, it limits the times of modulation downgrade in RBs allocation to enhance system throughput. In the simulations, the proposed method is compared to fixed sub-carrier dynamic resource allocation algorithm and adaptive dynamic sub-carrier resource allocation algorithm. Simulation results show that our method outperforms the other two methods in terms of throughput, transmission delay, packet loss ratio, and RB utilization.

Efficient energy management to prolong wireless sensor network lifetime

In this paper, we propose a Voronoi detection range adjustment method that utilizes distributed Voronoi diagram to delimit the area of responsibility for each sensor. We then use genetic algorithm to optimize the most suitable detection range for each sensor. Simulations show that VERA outperforms maximum detection range, K-covered [Huang and Tseng, 2003], and greedy [Cardei et al., 2006] methods in terms of reducing the overlaps among detection ranges, minimizing energy consumption, and prolonging network lifetime.

A Hybrid Design Framework for Video Streaming in IEEE 802.11e Wireless Network

Over the past decade, wireless network access and video streaming services have become more popular than ever. However, IEEE 802.11 DCF and IEEE 802.11e EDCA are not specifically designed for video streaming. This leads to the problem of transmitting overdue video packets and thus degrades both network performance and video quality. In this paper, we propose a hybrid design framework which consists of a MAC-centric cross-layer architecture to allow MAC- layer to retrieve video streaming packet information, a hybrid retransmission deadline and retry limit to save unnecessary packet waiting time, and a single-video multi-level queue to prioritize I/P/B slice (packet) delivery. Simulations show that the proposed methodology outperforms IEEE 802.11, IEEE 802.11 Timebase and IEEE 802.11e in packet loss rate, invalid packet ratio, delay time, jitter, and PSNR.

Fairness-based adaptive QoS scheduling for LTE

LTE employs Orthogonal Frequency-Division Multiple Access (OFDMA) to provide high data transmission rate in downlink. Users expose to different network environments, so their exploited services would experience different performance of throughput, delay time, and jitter. In the literature, many studies explored how to allocate resource blocks (RB) to different services efficiently. Famous examples are Proportional Fair (PF) and Modified Largest Weighted Delay First (MLWDF) scheduling algorithms. PF considers fairness-based service priority without Quality of Service (QoS) consideration. MLWDF favors those packets with timing constraints, resulting in uneven resource distribution. In this paper, we propose a fairness-based adaptive QoS scheduling algorithm for LTE downlink. We design an Adaptive Modified Largest Weighted Delay First (AMLWDF) formula, which considers delay time, instantaneous transmission rate, average transmission rate, delay constraints, QCI priority, and fairness at the same time. Simulation results show that AMLWDF outperforms Proportional Fair and Modified Largest Weighted Delay First in various performance indexes.

A Voronoi dEtection Range Adjustment (VERA) approach for energy saving of wireless sensor networks

Since the batteries in a wireless sensor networks cannot be replaced, efficient power management becomes an important research issue. If we can largely reduce the overlaps among detection ranges and decrease the amount of duplicate data then we can save energy more efficiently. Meguerdichian et al. exploit the coverage problems in wireless ad-hoc sensor networks in terms of Voronoi diagram and Delaunay triangulation. In this paper, we propose a Voronoi detection range adjustment (VERA) method that utilizes distributed Voronoi diagram to delimit the area of responsibility for each sensor. We then use genetic algorithm to optimize the most suitable detection range for each sensor. Simulations show that VERA outperforms maximum detection range, K-covered, and greedy methods in terms of reducing the overlaps among detection ranges, minimizing energy consumption, and prolonging the lifetime of the whole network.

Direction based routing strategy to reduce broadcast storm in MANET

In a Mobile Ad Hoc Network (MANET), mobile nodes may rapid change their positions and thus not only change the network topology but also may cause routing path disconnections frequently. Those routing protocols of on-demand would use flooding to broadcast RREQ (Route Request) messages to establish transmission route. However, flooding may cause broadcast storm and impair network performance severely. In this paper we propose a Directive Location-Aided Routing (DLAR) protocol that selects intermediate nodes based on the moving direction of the source node. The protocol design criteria are to keep the network disconnections as close to the destination node as possible and limit the scope of RREQ broadcast area to prevent broadcast storm. In addition, DLAR also employs ADOV as route discovery mechanism for LAR (Location-Aided Routing) and uses local repair mechanism to initiate route discovery at the local disconnected position in order to speed up route discovery process. We use NS-2 as network simulator to compare the performance of the proposed DLAR to that of the AODV, DSR and LAR in terms of packet delivery ratio, RREQ packet volume, end-to-end delay, average number of hops against node distribution density, node velocity, and number of node connections.