Dmitri Perkins

Factors affecting the performance of ad hoc networks

Mobile ad hoc networks (MANETs) are an emerging class of network architectures that are characterized by their highly dynamic topology, limited resources (i.e., bandwidth and power), and lack of fixed infrastructure. The primary motivation for such networks is increased flexibility and mobility. Random node mobility along with various other factors such as network size and traffic intensity may be very dynamic, resulting in unpredictable variations in the overall network performance. This study centers on investigating and quantifying the effects of various factors and their two-way interactions on the overall performance of ad hoc networks. This study contributes to the modeling and development of adaptive ad hoc protocols (routing, medium access control, scheduling and buffer management). Using 2/sup k/r factorial experimental design, we isolate and quantify the effects of five factors: node speed, pause-time, network size, number of traffic sources, and type of routing (source versus distributed), that affect the performance of ad hoc networks. Specifically, this paper evaluates the impact of these factors on the following performance metrics: throughput, average routing overhead, and power consumption. Our study was conducted using a library-based simulator for sequential and parallel simulation of wireless networks.

A survey on quality-of-service support for mobile ad hoc networks

The general field of mobile ad hoc networking is still in its infancy. Particularly, the challenge of providing Quality-of-Service (QoS) support for ad hoc networks is an open problem and remains relatively uncharted territory. Providing a complete QoS solution for the ad hoc networking environment requires the interaction and cooperation of several components. These components include: (1) a QoS routing protocol, (2) a resource reservation scheme and (3) a QoS capable medium access control (MAC) layer. In this paper, we present a survey of the current research that has addressed each of these components in the context of ad hoc networks. This work is intended to provide a broad and comprehensive view of the various components and protocols required to provide QoS support in computer networks, focusing primarily on ad hoc networks. First, we introduce the unique characteristics of mobile ad hoc networks, which distinguishing this new network architecture from traditional infrastructured wired and wireless networks (i.e. cellular-based networks). We also discuss the impact of these characteristics on QoS provisioning. Next, we describe the first QoS model proposed for mobile ad hoc networks and its relationship to QoS models proposed for the Internet. We then present a review of the proposed algorithms for each QoS component (e.g. QoS routing, resource reservation and the MAC layer). Copyright

Cross-Layer Hop-by-Hop Congestion Control in Mobile Ad Hoc Networks

The poor performance of TCP in wireless networks is further diminished in IEEE 802.11-based ad hoc networks networks due to the unnecessary actions of on-demand routing protocols when interference-induced packet drops happen at the medium access control (MAC) layer. In this paper, we present a cross-layer hop-by-hop congestion control scheme designed to improve TCP performance in multihop wireless networks. The proposed scheme, which we will refer to as CHCC, attempts to determine the actual cause of a MAC-layer packet loss and then coordinates the congestion-control response among the MAC, network, and transport protocols. The congestion control efforts are invoked at all intermediate and source nodes along the upstream paths directed from the wireless link experiencing the congestion-induced packet drop. Simulation results show that the proposed scheme achieves up to 70% higher throughput than TCP-Reno. The CHCC approach is interoperable with standard TCP and can be used transparently when using transport protocols (e.g., UDP) that do not implement congestion control.

MAC-SCC: medium access control with a separate control channel for multihop wireless networks

The recent research has shown that the basic RTS/CTS approach is not efficient in multihop wireless networks. In this paper, we propose a novel medium access control protocol with a separate control channel (MAC-SCC). A distinct feature of MAC-SCC (compared with other multichannel MAC protocols) is to use two Network Allocation Vectors (NAVs) for the data channel and the control channel, respectively. The transmission of the next data frame can be pre-determined during the current transmission via the separate control channel, and thus the frame collision probability and bandwidth wasted during backoff can be reduced. The system throughput is quantified via simulation. The results show that MAC-SCC yields a throughput gain up to 60% under high traffic load and has a significant lower link failure probability.

Using statistical design of experiments for analyzing mobile ad hoc networks

The performance of mobile ad hoc networks can be influenced by numerous factors, including protocol design at every layer; parameter settings such as retransmission limits and timers; system factors such as network size and traffic load; as well as environmental factors such as channel fading. In this work, we are concerned with understanding the functional relationship between these influential factors and performance of mobile ad hoc networking systems. We show how a systematic statistical design of experiments (DOE) strategy can be used to analyze network system and protocol performance, leading to more objective conclusions valid over a wide range of network conditions and environments. Using a DOE strategy and a 2k factorial design, we quantify the main and interactive effects of five factors (i.e., network density, node mobility, traffic load, network size, and medium access control scheme) on two response metrics (i.e., packet delivery ratio and end-to-end delay). Using these effects measures, we then develop two first-order linear regression models that define the functional relationship between the influential factors and two performance metrics.

Optimal Probabilistic Encryption for Secure Detection in Wireless Sensor Networks

We consider the problem of secure detection in wireless sensor networks operating over insecure links. It is assumed that an eavesdropping fusion center (EFC) attempts to intercept the transmissions of the sensors and to detect the state of nature. The sensor nodes quantize their observations using a multilevel quantizer. Before transmission to the ally fusion center (AFC), the senor nodes encrypt their data using a probabilistic encryption scheme, which randomly maps the sensors data to another quantizer output level using a stochastic cipher matrix (key). The communication between the sensors and each fusion center is assumed to be over a parallel access channel with identical and independent branches, and with each branch being a discrete memoryless channel. We employ J-divergence as the performance criterion for both the AFC and EFC. The optimal solution for the cipher matrices is obtained in order to maximize J-divergence for AFC, whereas ensuring that it is zero for the EFC. With the proposed method, as long as the EFC is not aware of the specific cipher matrix employed by each sensor, its detection performance will be very poor. The cost of this method is a small degradation in the detection performance of the AFC. The proposed scheme has no communication overhead and minimal processing requirements making it suitable for sensors with limited resources. Numerical results showing the detection performance of the AFC and EFC verify the efficacy of the proposed method.

Design and Implementation of CLASS: A Cross-Layer ASSociation Scheme for Wireless Mesh Networks

A widely-used association strategy in current 802.11 networks is to allow a Mobile Station (MS) to associate with the Access Point (AP) which has the best Received Signal Strength Indication (RSSI) value during its scanning. However, in 802.11-based wireless mesh networks, the conditions of the access link (e.g., traffic load of associated stations, and the frame error rate between the MS and the Mesh Router (MR)) and the conditions of the mesh backhaul (e.g., end-to-end latency, and asymmetric uplink and downlink transportation costs) have a significant impact on the network performance of the MS after its association. In this work, we propose a cross-layer association scheme for wireless mesh networks. The end-to-end airtime cost is used to determine the MR to which the MS should associate, comprising the access link airtime cost and the backhaul airtime cost. Our experimental results on a Linux-based testbed show that the proposed association scheme is capable of providing the mobile stations with the highest end-to-end network performance after their association.

BASH: A backhaul-aided seamless handoff scheme for Wireless Mesh Networks

Wireless mesh networks (WMNs) are comprised of mesh routers whose relatively small transmission range may result infrequent handoffs, leading to high packet delays and loss rates and limiting the performance of real-time applications over WMNs. In this work, we propose BASH - a backhaul-aided seamless handoff scheme. Our scheme takes advantage of the wireless backhaul feature of WMNs and allows a mobile station to probe the neighboring mesh routers by accessing the backhaul channel. After the probe request, the mobile station is able to switch back to its primary communication channel and resume its ongoing communication without waiting for the probe responses. The currently associated mesh router of the mobile station collects the probe responses and selects the new mesh route on behalf of the mobile station. Our works show that by utilizing the wireless backhaul, BASH (1) reduces the probing latency and, thus, the Layer-2 handoff latency; (2) allows partial overlap of the Layer-2 and Layer-3 handoffs, reducing the overall handoff latency; and (3) shortens the authentication latency by utilizing the transitivity of trust relationship. The experimental results show that BASH achieves an average Layer-2 handoff of 8.7 ms, which supports the real-time applications during the handoff.

Reducing localization errors in sensor ad hoc networks

In this paper, we present a localization system comprising two simple and efficient localization algorithms for sensor networks. We refer to the proposed localization system as the differential ad-hoc positioning system (DAPS). Differential GPS motivated the underlying idea on which the proposed localization system is based. Specifically, the algorithms use a differential error correction scheme that is designed to reduce the cumulative distance and positioning error accumulated over the multiple hops. Using simulation, we investigate DAPS and compare the proposed algorithms with non-differential based schemes. The key contribution of this work is the illustration of how differential error corrections can be calculated and incorporated into the localization process to effectively reduce the range measurement errors and, as such, significantly improve positioning accuracy.

Empirical model-based adaptive control of MANETs

The performance of mobile ad hoc networks (MANETs) depends upon a number of dynamic factors that ultimately influence protocol and overall system performance. Adaptive protocols have been proposed that adjust their operation based on the values of factors, such as traffic load, node mobility, and link quality. In this work, however, we are investigating the feasibility of an adaptive model-based self-controller that can manage the values of controllable factors in MANETs. In general, the proposed self-controller should determine a set of factor values that will maximize system performance or satisfy specific performance requirements. The model-based controller adapts or reconfigures system-wide parameters or protocol operation as a function of the dynamically changing network state. In this paper, we describe the proposed self-controller, its design issues, and provide a preliminary case study to demonstrate the effectiveness and tradeoffs of two potential empirical-modeling techniques: regression and artificial neural networks.