Conventionally, a memoryless analog repeater at the relay of a multihop transmission system amplifies the signal received from its incoming link, and retransmits the amplified signal to its outgoing link. In the frequency domain, such an amplification essentially is an ideal bandpass filtering, treating all the frequency components uniformly. For multicarrier systems like orthogonal frequency division multiplexing (OFDM) over frequency-selective channels, such a frequency-flat amplification is inadequate to exploit the benefits of adaptive processing at the relay. This paper analyzes the potential performance gain of non-uniform frequency-domain relay amplification, in which the gain coefficients for subcarriers are adapted from the frequency responses of both the incoming and outgoing links. The end-to-end achievable rate optimization problem is formulated and shown to lack the desired concavity property and thus is not amenable to Karush-Kuhn-Tucker approaches like water-filling. A heuristic power allocation and relay processing algorithm is proposed, which adaptively selects and matches a subset of the subcarriers for each link, according to the realization of the link frequency responses. Numerical results show that the heuristic algorithm achieves considerable performance gains, compared to conventional amplify-and-forward relay processing without adaptive power allocation.

Research in the field of underwater sensor networks exists from the need for scientific data collection, pollution monitoring, offshore oil exploration, tsunami warnings, ocean mapping, and tactical surveillance without human assistance. Aided with acoustic communication systems, autonomous underwater vehicles (AUVs) can work along with fixed nodes to enhance the capablities of sensor networks. With channel estimation algorithms, AUVs can detect connectivity loss and position sensors to self-configure for optimal network efficiency. However, one of the biggest challenges to underwater networks is multipath fading, where the reflection and scattering from the bottom and surface of the ocean result in severe intersymbol interference (ISI) of transmitted signals. Secondly, sound absorption loss increases with the increase in frequency, dramatically limiting the bandwidth available. Orthogonal frequency division multiplexing (OFDM) is proposed as a possible method of communication, by transmitting signals over multiple subcarriers simultaneously.

A formulation for a joint communication and estimation problem is proposed: simultaneous communication over a noisy channel and estimation of certain channel parameters are desired. We are interested in quantifying the tradeoff between the achievable rate and distortion in estimating the channel parameters. Two particular sample channels are considered and achievable rates for these channels are determined. First, the binary symmetric channel is examined; an achievable capacity-distortion tradeoff is derived for both joint and time-orthogonal protocols. For the flat fading, additive, white, Gaussian noise channel, a novel joint communication and estimation scheme using low correlation sequences is presented. It is observed that in most situations, joint communication and estimation performs better than a scheme where communication and estimation are performed individually, furthermore, the gains of joint communication and estimation over individual communication and estimation can be significant as the distortion tolerance increases. Finally, it is observed that even a slight tolerance to errors in the channel parameters close to the theoretical lower bounds yield significant improvements in the rate at which reliable communication is achievable.

Unmanned surface vehicles (USVs) are subjected to external forces such as wind, water currents and waves which make their control challenging. Typical problems in USV control involve navigation, trajectory tracking and station keeping. Station keeping implies dynamic positioning in order to reject disturbances while holding position. This is necessary when sensors need a sufficiently large warm-up and/or dwell time, to ensure that the data collected pertains to the appropriate location where measurements are being made. This poster describes ongoing work on the robotic boat RoboDuck-II in addressing these issues.