Vehicle lateral stability management using gain-scheduled robust control

This paper deals with the design of a yaw rate controller based on gain-scheduled H_∞ optimal control, which is intended to maintain the lateral stability of a vehicle. Uncertain factors such as vehicle mass and cornering stiffness in the vehicle yaw rate dynamics naturally call for the robustness of the feedback controller and thus H_∞, optimization technique is applied to synthesize a controller with guaranteed robust stability and performance against the model uncertainty. In the implementation stage, the feed-forward yaw moment by driver’s steer input is estimated by the disturbance observer in order to determine the accurate compensatory moment. Finally, HILS results indicate that the proposed yaw rate controller can satisfactorily improve the lateral stability of an automobile.     

Securing one-way hash chain based incentive mechanism for vehicular ad hoc networks

One of the most promising applications of yet to come vehicular ad hoc networks (VANETs) is commercial advertising. However, in order to realize commercial advertising over VANET, proper incentives and security mechanisms must be taken into consideration due to the existence of selfish and/or malicious users in the real-world scenario. In this paper, we propose a secure incentive scheme for VANET advertising systems based on one-way hash chains. We also analyze the advertisement propagation behavior in our VANET advertising system using a mathematical model.     

Time-critical underwater sensor diffusion with no proactive exchanges and negligible reactive floods

In this paper we study multi-hop ad hoc routing in a scalable underwater sensor network (UWSN), which is a novel network paradigm for ad hoc investigation of the world below the water surface. Unlike existing underwater acoustic networks (UAN), the new UWSN paradigm dispatches large number (in the thousands) of unmanned low-cost sensor nodes to locally monitor and report otherwise not easily accessible underwater events in a time-critical manner. Due to the large propagation latency and very low bandwidth of the acoustic channel, a new protocol stack and corresponding models are required as conventional approaches fail. In particular, we show that neither proactive routing message exchange nor reactive/on-demand flooding is adequate in the challenging new underwater environment. Unlike the terrestrial scenarios, on-demand flooding cannot be both reliable and efficient due to widespread collisions caused by the large propagation delay. On the other hand, as in terrestrial scenarios, proactive routing is more expensive and less efficient than on-demand routing in typical underwater environments. We propose a “conservative” communications architecture that minimizes the number of all packet transmissions to avoid possible acoustic collisions. This is implemented in the non-intrusive underwater diffusion (UWD), which is a multi-hop ad hoc routing and in-network processing protocol with no proactive routing message exchange and negligible amount of on-demand floods. To achieve its design goal, UWD does not rely on GPS or power hungry motors to control currents. Instead, UWD is designed in a minimalist’s framework, which assumes homogeneous GPS-free nodes and random node mobility. Our simulation study verifies the effectiveness and efficiency of our design.     

Optimizing random network coding for multimedia content distribution over smartphones

It is known that random network coding (RNC) technology helps enhance multimedia content distribution systems in various ways; however, the enhancement can vary widely depending on how the technology is realized in the systems. RNC technology entails an encoding process at the server-side and a decoding process at the clients. Typically, the decoding process is the bottleneck especially when resource-limited mobile clients such as smartphones are employed. Thus, to fully exploit the benefit of RNC technology, it is crucial to maximize throughput and minimize latency of the decoding process of RNC at the client-side. In this paper, we explore the implementation space of RNC on smartphone platforms and propose best practices that optimize RNC performance on smartphone in terms of decoding throughput (or delay) as well as energy consumption. Via experimental results, we show that our proposal for optimizing RNC achieves throughput enhancement along with energy conservation at the same time on smartphones.     

An analytical framework for neighbor discovery strategies in ad hoc networks with sectorized antennas

This letter analyzes a class of neighbor discovery algorithms in ad hoc networks using directional antennas for both reception and transmission. We consider synchronized and random sector - timeslot assignment strategies and compare their behaviors. Analytic results validated by simulation clearly show the latency benefits of the synchronized sector assignment over the random assignment.     

Seamless IP mobility support for flat architecture mobile WiMAX networks - [WiMAX update]

The growing demand for wireless Internet services is accelerating the evolution of wireless networks toward all-IP architecture, and the mobile WiMAX network is a prominent example. Although currently deployed mobile WiMAX networks use hierarchical architecture, flat architecture is feasible and specified as a design alternative in the mobile WiMAX standard. In flat architecture the functionalities of the ASN-GW and BS are consolidated into a single element. In this article we first discuss the benefits and challenges of flat architecture mobile WiMAX networks. We then present a scheme to deal with the seamless mobility issue, which is one of the key challenges of the flat architecture. The proposed scheme combines two standard IP-mobility protocols, Proxy Mobile IP and Fast Mobile IP, and customizes them for IEEE 802.16e-based mobile WiMAX networks. This provides interoperability with existing mobile WiMAX networks. We demonstrate the viability of the proposed scheme through simulations using NS-2.     

Tactics based approach for integrating non-functional requirements in object-oriented analysis and design

Non-Functional Requirements (NFRs) are rarely treated as “first-class” elements in software development as Functional Requirements (FRs) are. Often NFRs are stated informally and incorporated in the final software as an after-thought. We leverage existing research work for the treatment of NFRs to propose an approach that enables to systematically analyze and design NFRs in parallel with FRs. Our approach premises on the importance of focusing on tactics (the specific mechanisms used to fulfill NFRs) as opposed to focusing on NFRs themselves. The advantages of our approach include filling the gap between NFRs elicitation and NFRs implementation, systematically treating NFRs through grouping of tactics so that tactics in the same group can be addressed uniformly, remedying some shortcomings in existing work (by prioritizing NFRs and analyzing tradeoff among NFRs), and integration of FRs and NFRs by treating them as first-class entities.     

E-ODMRP: Enhanced ODMRP with motion adaptive refresh

On-Demand Multicast Routing Protocol (ODMRP) is a multicast routing protocol for mobile ad hoc networks. Its efficiency, simplicity, and robustness to mobility render it one of the most widely used MANET multicast protocols. At the heart of the ODMRP’s robustness is the periodic route refreshing. ODMRP rebuilds the data forwarding “mesh” on a fixed interval and thus the route refresh interval is a key parameter that has critical impact on the network performance. If the route refresh rate is too high, the network will undergo too much routing overhead, wasting valuable resources. If it is too low, ODMRP cannot keep up with network dynamics, resulting in packet losses due to route breakages. In this paper, we present an enhancement of ODMRP with the refresh rate dynamically adapted to the environment. Simulation results show that the Enhanced ODMRP (E-ODMRP) reduces the packet overhead by up to a half yet keeping a packet delivery ratio comparable to that of the original ODMRP. E-ODMRP compares favorably with other published multicast schemes.     

Cross-layer optimization for wireless multihop multicast networks

Cyber-Physical System (CPS) is envisioned to tightly integrate the cyber-world of computation, communication, and control with the physical world. CPS is typically designed as a networked system of interacting sensors, actuators, and embedded computing devices to monitor and control the physical world. Thus, one of the essential building blocks of such a system is a highly efficient networking infrastructure. In this paper, we aims to develop an efficient wireless networking technology which can be utilized in CPS. More specifically, we develop a cross-layer optimization model based on the Network Utility Maximization (NUM) framework and its distributed solution for wireless multihop multicast networks exploiting multi-user diversity. It is known that the capacity of a wireless network can be increased by exploiting different channel conditions at different users, i.e., multi-user diversity; however, it is yet to be determined how much performance gain can be achieved by exploiting multi-user diversity in wireless multihop multicast networks. To address this problem, we extend the NUM framework and derive a new optimization problem including the benefits of multi-user diversity for multicasting scenarios in wireless multihop networks under a probabilistic media access control (MAC). In our problem, multi-user diversity is achieved via opportunistic scheduling. Then, we propose a distributed approximation algorithm for the problem. Our numerical results confirm that the benefit of multi-user diversity is prominent in a wireless multihop network with multicast flows.