A non-intrusive technique based on modeling the body as a transmission channel was tested in vitro and was shown to perform equally with a commercial OTS glucose meter on saline-glucose solutions of concentrations of glucose from 30 to 300 mg/dL. The technique uses an initial frequency sweep to locate a frequency where a resonant response occurs. At that position the phase is changing rapidly and can therefore be used more easily to measure a phase difference. This sweet spot can be natural or assisted. Present efforts are toward a system which uses a feedback amplifier with a low phase margin in order to easily find the the region of rapid phase change. Initial tests measuring known glucose concentrations with an OTS glucose meter and then using the proposed technique were shown to correlate with the actual concentrations with an R(2) = 0.9879 and R(2) = 0.9952 respectively (with one outlier data point removed). Further tests are being conducted in vitro on whether other interfering agents may disrupt results. The ultimate goal of the development of this technique is to fabricate a device that is worn externally on the upper arm and does not require blood for testing. A device is being built and is scheduled for human subject testing in Summer 2013. Some results of human testing may be available at the EMBC conference. Human testing will involve measurements done with our device against an OTS glucose meter and results compared. The future goal is to refine the device so that it can be worn 24 hours a day and will automatically test the wearer at a user specified interval such as 10 or 20 minutes by transmitting a low power signal for a few microseconds. Power levels are still being determined but, in simulation, nano-watts was sufficient for the distance needed to travel which is orders of magnitude less than present day cell phones. To evaluate safety of the system, 3D electromagnetic simulations are being carried out with the device positioned strapped to the upper arm. Simulations show transmission along the targeted paths which suggests that affective glucose sensing is possible with this device.