We propose practical sets of rules for adapting Clear Channel Assessment (CCA) and Transmit Power (TP) parameters by simulating ensembles of randomly generated wireless networks and collecting throughput statistics. The rules' performances depend strongly on network topology, with increases in throughput in many cases. However, networks with a high clustering coefficient are often adversely effected by the adaptations. But simulations of small-scale networks show that apparently adverse adaptations may still yield benefits for uneven demand combinations. Finally, we have found that throughput is not usually correlated with the number of hidden or exposed nodes in any non-trivial network set-up.

We consider multi-hop wireless mesh networks and examine whether capacity may be improved by distributing the data flows for each origin-destination (OD) pair across multiple routes. The network geometry and the application of technologies, such as beamforming or full-duplex, are described by a matrix that describes which pairs of transmissions (or 'links') are compatible in that they can occur simultaneously without collisions or other conflicts. Under a binary interference framework, the conflict matrix is used to derive the maximal sets of compatible links, and secondly, these are used to derive a system of linear inequalities that bounds links' data flows. These steps are computationally expensive, but we show how the system usually collapses when re-expressed in terms of flows on routes. The theory is illustrated in terms of a simple 'Braess' network example with a single OD pair. We then consider networks with two OD pairs whose data flows have some nodes in common and thus contend with each other. We show how to design networks that exploit multiple relay nodes and routes so as to increase capacity. We then examine the same problem on ensembles of random networks. We find that in many cases, capacity can be improved if the OD pairs distribute their traffic over several routes. We pose and solve a set of linear programs that model (i) cooperative behavior; (ii) the optimization of one OD pair when presented with a fixed route assignment by the other; and (iii) variants of these games when both OD pairs are in contention with background (single-hop) traffic.

This paper is concerned with smartphone based inertial sensors for locating pedestrians. Two set of experiments were conducted over seven subjects. One set of the subjects walked along a straight path at a perceived normal and speedy pace, and the other set of subjects traversed a simple map consisting of two L shaped paths. In the experiments, a smartphone - HTC 2710e was handheld, with screen facing upwards was used to record the acceleration, magnetometer and gyroscope samples at 20Hz. The accelerometer samples are then analyzed to detect and count the footsteps, while gyroscope and magnetometer samples are used to estimate the heading. The results of our experiment show that the average error in foot step detection rate is less than 3% in normal walking and 5% in speedy walking using Fast Fourier Transform (FFT) based approach. While the most nearly correct estimated displacement is via Weiner approach with average percentage error of 5.74% in normal walking and 8.07% in speedy walking. The average percentage error in heading is less than 110in turnings in both the L shaped paths.

Lane changes are important in quantifying traffic for both operational and planning purposes. Traditional in-lane loop detectors do not count lane changes; hence, historically, traffic engineers have estimated them using other data sources. This paper provides a method for estimating the number of lane changes based on observations of “straddling” vehicles that are simultaneously detected by the loops in adjacent lanes. In the data considered here, such “straddles” almost always correspond to vehicles that are in the process of changing lanes. However, many lane changes take place between detector sites and, hence, do not result in straddles. The methods developed here estimate probability distribution for the number of lane changes given an observed number of straddles. The efficacy of this approach depends on calibration issues and on the size of the aggregation period. In the evaluation study presented here, the results are good: The proposed method gives the number of lane changes with approximately 10% error, even though the number of lane changes per aggregation period varies by a factor of 10 over time.

Freeway traffic state estimation and prediction are central components in real-time traffic management and information applications. Model-based traffic state estimators consist of a dynamic model for the state variables (e.g., a first- or second-order macroscopic traffic flow model), a set of observation equations relating sensor observations to the system state (e.g., the fundamental diagrams), and a data-assimilation technique to combine the model predictions with the sensor observations [e.g., the extended Kalman filter (EKF)]. Commonly, both process and observation models are formulated in Eulerian (space-time) coordinates. Recent studies have shown that this model can be formulated and solved more efficiently and accurately in Lagrangian (vehicle number-time) coordinates. In this paper, we propose a new model-based state estimator based on the EKF technique, in which the discretized Lagrangian Lighthill-Whitham and Richards (LWR) model is used as the process equation, and in which observation models for both Eulerian and Lagrangian sensor data (from loop detectors and vehicle trajectories, respectively) are incorporated. This Lagrangian state estimator is validated and compared with a Eulerian state estimator based on the same LWR model using an empirical microscopic traffic data set from the U.K. The results indicate that the Lagrangian estimator is significantly more accurate and offers computational and theoretical benefits over the Eulerian approach.