Search

Researchers use peak and mean characteristics to quantify changes in kinematics and kinetic loading patterns. The variability of human motion warrants a more complex method of analysis. This paper will demonstrate that functional data analysis (FDA) can quantify differences throughout a lifting waveform that would not be evident when applying standard statistical methods to discrete variables. A functional ANOVA was used to identify variation in kinematic and kinetic waveforms resulting from fatigue experienced over a 45-min lifting task. Only trunk angular motion and hip angular velocity were significantly different across the work task using discrete peak values. FDA was able to demonstrate that additional areas outside of the peaks were significantly different in the late phase for several waveforms. FDA was able to demonstrate significant variation across the work period for kinetic and kinematic variables when statistics performed on discrete variables reported no variation. More robust methods for f...

This paper derives the multi-hop capacity of OFDM-based MIMO-multiplexing relaying systems. MIMO-multiplexing relaying presents a spectrally efficient means of realizing mesh supports in wireless networks operating over licensed bands by providing separate links for access and mesh relaying services on the same broadband radio channel. We show that for an NtimesN MIMO-multiplexing relaying system with amplification factor alpha at relay nodes, R-hops relaying degrade the capacity by at most -Nlog2(alpha2R/ (1 + Sigma r=1 R alpha2rNr)) + RNlog2(N) bits/sec/Hz. Therefore, greater capacity loss is experienced in MIMO-multiplexing relaying involving high order MIMO systems. We also illustrate that the capacity loss is independent of the OFDM configurations employed; thus network operators could employ higher OFDM configurations to compensate for data rate loss in access services when some of the MIMO-multiplexing links are dedicated to mesh relay. This pioneering analysis provides useful guidelines for network operators planning to employ MIMO-multiplexing option for mesh relay supports.

One of the main challenges for metro-scale WiMAX mesh network deployments is related to capacity scaling. In a full mesh mode, a WiMAX node acts as a mesh router as well as a client access node. To improve latency and speed performance, typical dual- and multi-radio mesh solutions use different radio channels to create separate links for access and mesh relaying services. The available spectrum is therefore split between mesh and client access services. Network operators operating the WiMAX system over the licensed spectrum are not keen to provide separate radio channels for access and mesh relay services, as this reduces the total number of users serviced per spectrum allocation. MIMO-multiplexing relaying approach however provides separate links for access and mesh relaying services on the same radio channel. In this paper, we discuss the multi-hop capacity of OFDM-based MIMO-multiplexing relaying in WiMAX networks. For an N×N MIMO-multiplexing relaying with amplification factor α at relay nodes, R-hops relaying degrade the capacity by at most -Nlog2(α2R/(1+ΣRr=1 α2r Nr)) +RNlog2(N) bits/sec/Hz. Therefore, greater capacity loss is experienced in networks employing high-order MIMO-multiplexing relaying. We also show that the capacity loss is independent of the OFDM configurations employed; thus network operators could employ higher OFDM configurations to compensate data rate loss in access services when some of the MIMO-multiplexing links are dedicated to mesh relay. This analysis provides useful guidelines for operators planning MIMO-multiplexing option for mesh support in WiMAX network.

In this paper, we propose a novel single hop-clustering scheme called step wise adaptive clustering hierarchy (SWATCH). SWATCH relieves the CH number variability problem by employing stepwise CH selection in two stages. It is a dynamic and straightforward scheme as LEACH. However, instead of selecting all CHs in one step, SWATCH splits the selection phase into an initial selection stage and an add-on selection stage. The initial selection is similar to LEACH. However, if the number of CHs in the initial selection is below a pre-determined target, the add-on selection will be invoked and will continue until an acceptable number of CHs have been selected. As a result, the number of CHs selected in each round tends to congregate in a narrow range around the optimal value. In order to evaluate the performance of SWATCH, we develop a hierarchical Markov chain model to track the behavior of the system. Numerical results verify our design objective in that the number of selected CHs highly conforms to the optimal value. Based on the results, an optimal number of CHs are greatly reduced the communication energy.

Node placement strategy is an intrinsic issue when provisioning of a wireless sensor network (WSN). In this paper, we address the placement problem for a class of heterogeneous WSNs, wherein nodes have different energy supplies and functionalities. We formulate a generalized node placement optimization problem aiming at minimizing the network cost with constraints on lifetime and connectivity. We propose a placement scheme with two phases. We model the placement of the first phase relaying nodes (FPRNs) as a minimum set cover problem and a dynamic programming algorithm is developed to solve it. For the placement of the second phase relaying nodes (SPRNs), we derive two fundamental design principles-the far-near strategy and max-min strategy. The implementation of the placement schemes is illustrated by examples. Our proposed mechanism is a first attempt towards facilitating realistic relay node placement in WSNs.

We propose a new framework for ant-based routing algorithms which addresses the trade-off between effective and timely discovery of optimal routes, and the desire to keep data traffic flowing smoothly. The approach utilizes a database of stored delay estimates on routes to build separate routing tables for ants and data packets. This decoupling results in better control of responsiveness to load or topology fluctuations since the tables can evolve at different rates, according to different rules. The presented architecture also features guaranteed loop-free routes for data packets.

We present one exact and two approximate methods for calculating blocking probabilities in an all-optical network with arbitrary placement of wavelength converters. The exact method is most useful for short paths (no more than 4 links), while the first approximate method is useful for slightly longer hop paths (no more than 6 links). The second approximate method is based on a recursion and can be applied to arbitrary length paths. A simple wavelength converter placement policy is described and implemented using the blocking models in an example using the CA*Net II topology.

We present two bounds (one lower bound and one upper bound) on the probability of a union of a finite number of events. The bounds-which are expressed in terms of only the individual event probabilities and the pairwise event probabilities-are applied to examine the symbol error (P/sub s/) and bit error (P/sub b/) probabilities of an uncoded communication system used in conjunction with M-ary PSK/QAM modulations and maximum a posteriori (MAP) decoding over AWGN channels. It is shown that both bounds provide a good, and often excellent, estimate of the error probabilities over the entire range of the signal-to-noise ratio E/sub b//N/sub 0/. The upper bound on P/sub s/ and the lower bound on P/sub b/ are particularly impressive as they agree with the simulation results even during very severe channel conditions.

Compartmental modelling, a technique that is widely used in areas such as bioscience, process control engineering, may serve as the basis for the performance evaluation of systems that are too complex to be handled by exact analytical/computational methods. With this technique, the dynamics of a system are described in terms of a set of differential or difference equations which permits the investigation of both the steady-state and transient behaviour of the system. In this paper, we describe how this technique can be used to model connection-oriented symmetric communication networks supporting multiple service classes, where each class is characterized by its bandwidth requirement, arrival and departure processes. We then demonstrate an application of this modelling technique to the steady-state analysis of a network and examine the accuracy of this modelling technique via simulations.