We introduce an adaptive pacing scheme to overcome the drawbacks of TCP in wireless mesh networks with Internet connectivity. The pacing scheme is implemented at the wireless TCP sender as well as at the mesh gateway, and reacts according to the direction of TCP flows running across the wireless network and the Internet. TCP packets are transmitted rate-based within the TCP congestion window according to the current out-of-interference delay and the coefficient of variation of recently measured round-trip times. Opposed to the majority of previous work which builds on simulations, we implement a Linux prototype of our approach and evaluate its feasibility in a real 20-node mesh testbed. In an experimental performance study, we compare the goodput and fairness of our approach against the widely deployed TCP NewReno. Experiments show that our approach, which we denote as Mesh Adaptive Pacing (MAP), can achieve up to 150% more goodput than TCP NewReno and significantly improves fairness between competing flows. MAP is incrementally deployable since it is TCP-compatible, does not require cross-layer information from intermediate nodes along the path, and requires no modifications in the wired domain.

Feedback is essential to guide performance in simulation-based training (SBT) and to refine learning. Generally outcomes improve when feedback is delivered with personalized tutoring that tailors specific guidance and adapts feedback to the learner in a one-to-on environment. Therefore, emulating by automation these adaptive aspects of human tutors in SBT systems should be an effective way to train individuals. This study investigates the efficacy of automating different types of feedback in a SBT system. These include adaptive bottom-up feedback (i.e., detailed feedback, changing to general as proficiency develops) and adaptive top-down feedback (i.e., general feedback, changing to detailed if performance fails to improve). Other types of non-adaptive feedback were included for performance comparisons as well as to examine the overall cognitive load. To test hypotheses, 130 participants were randomly assigned to five conditions. Two feedback conditions employed adaptive approaches (bottom-up and top-down), two used non-adaptive approaches (constant detailed and constant general), and one functioned as a control group (i.e., only a performance score was given). After preliminary training on the simulator system, participants completed four simulated search and rescue missions (three training missions and one transfer mission). After each training mission, all participants received feedback relative to the condition they were assigned. Overall performance on missions, knowledge post-test scores, and subjective cognitive load were measured and analyzed to determine the effectiveness of the type of feedback. Results indicate that: (1) feedback generally improves performance, confirming prior research; (2) performance for the two adaptive approaches (bottom-up vs. top-down did not differ significantly at the end of training, but the bottom-up group achieved higher performance levels significantly sooner; (3) performance for the bottom-up and constant detailed groups did not differ significantly, although the trend suggests that adaptive bottom-up feedback may yield significant results in further studies. Overall, these results have implications for the implementation of feedback in SBT and beyond for other computer-based training systems.

No description available.

This thesis demonstrates how a Fast Gradient approximation to a Lattice Filter can be used as an Adaptive Line Enhancer for sampled data consisting of multiple switching tones in correlated noise. A tradeoff analysis is performed with four methods of digital filtering including a conventional Digital Fourier Transform (DFT) algorithm, a Least Mean Squares (LMS) adaptive algorithm, a Fast Recursive Least Squares (Fast RLS) adaptive algorithm, and the Fast Gradient adaptive algorithm. The DFT algorithm is incapable of removing correlations from the incoming noise, and the LMS and Fast RLS algorithms become unstable when a dynamic switching environment is being filtered. The Fast Gradient adaptive algorithm simulated on a computer is robust and capable of converging to an optimal set of FIR filter weights with minimum Mean Squared Error. Parameters for the Fast Gradient algorithm are optimized to provide good filter performance with a minimum number of computations. / Master of Science

The increasing demand for mobile communication services without a corresponding increase in RF spectrum allocation motivates the need for new techniques to improve spectrum utilization. The CDMA and adaptive antenna array are two approaches that shows real promise for increasing spectrum efficiency. In this research, we investigate the performance of different blind adaptive array algorithms in the CDMA systems. Two novel algorithms, least-squares despread respread multitarget array (LS-DRMTA) and least-squares despread respread multitarget constant modulus algorithm (LS-DRMTCMA), are developed, and a MATLAB simulation testbed is created to compare the performance of these two novel algorithms with those of the multitarget least-squares constant modulus algorithm (MT-LSCMA) and multitarget steepest-descent decision-directed (MT-SDDD) algorithm. It is shown from the simulation results that these two novel algorithms can outperform the other algorithms in all the test situations (e.g., AWGN channel, timing offset case, frequency offset case, and multipath environment). It is also shown that these two algorithms have less complexity and can converge faster than the other algorithms. / Master of Science

The usage of Unmanned Aerial Systems (UASs), such as quadcopters and hexacopters, has steadily increased over the past few years in both recreational and commercial use. This increased availability to purchase such systems has also given rise to many safety and security concerns. A common concern is that the misuse of a UAS can cause damage to airplanes and helicopters in and around airports. Another growing concern is the use of UASs for terrorist intentions such as using the UAS as a remote controlled bomb. There is clearly a need to be able to detect the presence of unwanted UASs in restricted areas. This thesis work presents the beginning work towards a method to detect the presence of these UASs using the blade pass frequency (BPF) of the motors and rotors of a home made quadcopter. A low cost uniform linear microphone array is first used to perform a simple delay-and-sum beamformer to spatially filter out noise sources. The beamformer output is then divided into sub-bands using three bandpass filters centered on the expected location of the fundamental BPF and its 2nd and 3rd harmonics. For each sub-band, an adaptive filter called an adaptive line enhancer is used to extract and enhance the narrowband signals. The response of the adaptive filters are then used to detect the quadcopter by looking for the presence of the 2nd and 3rd harmonics of the fundamental BPF. Static tests of the quadcopter out in a field showed promising results for this method with the ability to detect up to the 3rd harmonic 90ft away and the 2nd harmonic 130 ft away. / Master of Science / The usage of Unmanned Aerial Systems (UASs), such as quadcopters and hexacopters, has steadily increased over the past few years in both recreational and commercial use. This increased availability to purchase such systems has also given rise to many safety and security concerns. A common concern is that the misuse of a UAS can cause damage to airplanes and helicopters in and around airports. Another growing concern is the use of UASs for terrorist intentions such as using the UAS as a remote controlled bomb. There is clearly a need to be able to detect the presence of unwanted UASs in restricted areas. This thesis work presents the beginning work towards a method to detect the presence of a home made quadcopter based on the sound it produces. A series of microphone are first used to remove surrounding sounds that could interfere with the quadcopter’s sound. The output of this processes is then divided into smaller sections using three filters centered on the expected location of the most important and information rich parts of the quadcopter’s sound. For each section, a final filter is used to extract and enhance the signals of interest produced by the quadcopter. The response of these filters are then used to detect whether the quadcopter is present or not. Static tests of the quadcopter out in a field showed promising results for this method with the ability to detect the quadcopter 90 to 130 ft away.

This work is a study of a hybrid adaptive controller that blends fixed feedback control and adaptive feedback control techniques. This type of adaptive controller removes the requirement that information about the disturbance is known apriori. Additionally, the control structure is implemented in such a way that as long as the adaptive controller is stable during adaptation, the system consisting of the controller and plant remain stable. The objective is to design and implement an adaptive controller that damps the structural vibrations induced in a multi-modal structure. The adaptive controller utilizes an adaptive infinite impulse response lattice filter for improved damping over the fixed feedback controller alone. An adaptive finite impulse response LMS filter is also implemented for comparison of the ability for both algorithms to reject harmonic, narrow bandwidth and wide bandwidth disturbances. It is demonstrated that the lattice filter algorithm performs slightly better than the LMS filter algorithm in all three disturbance cases. The lattice filter also requires less than half the order of the LMS filter to get the same performance. / Master of Science

No description available.

International Telemetering Conference Proceedings / October 26-29, 1992 / Town and Country Hotel and Convention Center, San Diego, California / The uncertainty in the gain of voltage controlled crystal oscillators (VCXOs) used in the implementation of certain analog phase-locked loops (PLLs) suggests some form of automatic tuning algorithm, both for pretuning and during operation. This paper proposes an adaptive PLL (APLL) algorithm to fill this need for PLLs used in the recovery of tones in noise. This algorithm makes use of a resonant error algorithm to remove the effects of VCXO noise, measurement noise, and parasitic poles. Both classical convergence theorems and robustness theorems that indicate the functionality of the proposed algorithm are given. Finally, the implementation of this algorithm is considered.

International Telemetering Conference Proceedings / October 22-25, 2001 / Riviera Hotel and Convention Center, Las Vegas, Nevada / We explore the idea of matching a scaling function - the basic building block of a wavelet function - to a desired spectrum. This would allow the scaling function to be used as the signal pulse for a digital communication system that is matched to the channel, avoiding problems such as energy loss or noise amplification due to spectral nulls. An unconstrained parameterization of the scaling function coefficients represents the scaling functions. This parameterization is adapted using gradient descent. Tests indicate that the adaptation is able to capture major features of a desired spectrum, including spectral nulls and major lobes.