Narrow Results

The IEEE 802.11e Enhanced Distributed Channel Access (EDCA) is designed to provide differentiated services without real hard performance guarantees, such as bandwidth and delay bounds in wireless LANs. In this paper we design a measurement-assisted, model-based call admission control (MM-CAC) scheme to support guaranteed QoS in the EDCA. For this purpose, we first develop an accurate analytical model of the legacy IEEE 802.11 DCF under non-saturation conditions; then using this model, we define the concept of equivalent number of competing entities, to convert a heterogeneous (multi-class) EDCA network into an equivalent homogeneous network and study traffic classes in isolation. Our analytical model is then invoked to decide to admit a new flow or reject it with much little computational complexity as compared to a fully fledged EDCA model. Numerical results validate our model and analysis, and simulations demonstrate the effectiveness of our MM-CAC scheme.

Distributed coordination function (DCF) is a primary random access mechanism of IEEE 802.11, which is the basic protocol of wireless LAN based on the CSMA/CA protocol. It enables fast installation with minimal management and maintenance costs and is a very robust protocol for the best effort service in wireless medium. The current DCF, however, is known to be unsuitable for real-time applications such as voice message transmission. In this paper, we focus on the performance issues of IEEE 802.11 which accommodates the prioritized messages. Existing results use the initial window size and backoff window-increasing factor as tools to handle the priority of the messages. Instead, we introduce a novel scheme which chooses the backoff timer with arbitrary probabilities. By this, one can greatly reduce the backoff delay of the lower priority messages without degrading the performance of higher priority. Additionally, we provide a step-by-step procedure which determines the system parameters such as the initial window size, backoff window-increasing factor, and the backoff timer choosing probabilities, guaranteeing the quality of service (QoS).

This paper presents a distributed and concurrent multiple access-based collision avoidance MAC scheme to increase spatial reuse and hence the overall throughput of wireless networks. We make use of location information and power capture techniques to realize this protocol. While other wireless MAC protocols have used power capture technique in a general manner to reject the interfering signals during the entire reception of a packet, we exploit the additional capability of power capture to "lock-on" to the intended packet during a period known as "capture time" to admit other parallel transmissions. Simulation results show that our protocol can outperform the IEEE 802.11 throughput performance by a factor of 2 or more when there are short and medium range transmissions within a given area.

Recently, using a wireless technology to transmit physiological signals, such as electrocardiogram (ECG), for a home tele-care system has received great attention. Although wireless transmission can provide the mobility advantage, it has to cope with the potential problem in limited bandwidth and induced interference. In this study, we propose an integration design in which a state-of-the-art compression algorithm called SPIHT (set partitioning in hierarchical trees) is combined with an unequal error protection scheme, to solve the problem for the ECG signals transmitted in Bluetooth packets. In this design, part of the SPIHT bit stream behaves like a fragile variable length code and needs a stronger protection with the forward error correction (FEC) code; the rest of the bit stream is lightly protected by the code. The simulation results show that the 2/3-rate FEC code in DM packets works effectively without the proposed scheme when the interference is fairly small. However, with the proposed scheme, the quality of received ECG signals is usually much better than that without the scheme when stronger interferences from fading channels and wireless LAN are encountered in an indoor environment. Consequently, the important features of an ECG waveform, such as P wave, QRS complex, and T wave, could be well preserved at a receiver site with clinically acceptable reconstruction quality. In addition, the data compression method in the proposed scheme can save total transmission power and time, and therefore reduce its potential interferences to other wireless devices.

This paper addresses the optimization of handovers in a Wireless LAN/Mobile IPv4/v6 environment by controlling the link layer handover process directly at the mobile node's side. One successful way to improve system performance of mobile clients in Wireless LAN multimedia networks is to use efficient and fast handover mechanisms when users move between different access points. We implemented some quality parameters of the wireless link provided by Network interface Card (NIC) firmware into a network-management software running on the mobile node. This concept allows a mobile user to automatically decide about when to start a handover and to which access point to connect to. The experience shows that our multimedia-enabled solution provides the necessary guidance to monitor and control the handover process without adding new extensions to the already existing Mobile IP standards.

In this paper, we describe a novel architecture to enable a secure communication among mobile devices using different wirelesstechnologies like wireless LAN, Bluetooth, cellular systems or even infrared. Making use of the combination of these technologies for the data transmission and for the signaling of the communication, we analyze several scenarios with increasing complexity. The complete picture appears in the last scenario where all technologies are involved and the network is composed of heterogeneous mobile nodes. The paper also presents a solution for the setup of a secured communication channel (i.e. a Virtual Private Network connection) between several heterogeneous mobile nodes controlled by the cellular network operator. The mobile nodes can be either cellular aware or non-cellular aware in this framework. We propose to setup the heterogeneous network communications via the cellular network using the cellular aware nodes.

In this paper, a new adaptive minimum contention window binary exponential backoff algorithm (referred to as AWBEB algorithm) is proposed to improve the performance of IEEE 802.11 Distributed Coordination Function (DCF) scheme, which is widely used for Ad Hoc networks and wireless LAN. We model the AWBEB algorithm scheme via establishing a bidimensional discrete-time Markov Chain, and the stationary distribution probabilities of AWBEB algorithm are derived. The performance of AWBEB algorithm is analyzed, and then AWBEB algorithm is simulated by computer programming. The analytical and numerical results show that the system saturation throughput and medium access delay of a packet for a given number of nodes N are better than the DCF algorithm.