Phases

== =Project Phases=

**Project: Phase I -** **Software Simulation**
First a smart sensor system shall be simulated and demonstrated in a software environment showing how an input signal can be conditioned to ensure that proper information extracted from that signal does not contain too much noise and without the calibration and adjustment through human intervention. As an example of inputs to a smart sensor system, a temperature and pressure sensor shall be used in this project. The design and simulation of these however are only secondary objectives in our project. The output of those sensors, from a given input, will be fed back into a simulated smart data acquisition system (DAQ) which will do the conditioning. This conditioning smart DAQ is the primary focus.

**Temperature Sensor**
The temperature sensor was designed and simulated using Multisim�. It is a very simple design, exploiting the fact that the diode will exert a different voltage drop depending on an applied temperature. Hence the schematic below shows a simple temperature dependent circuit. //**Figure 1: Simple Temperature Sensor simulation schematic**//

**Pressure Sensor**
The pressure sensor on the other hand is much more complicated. It is a micro-electrical-mechanical-system (MEMS) device known as a comb-drive, for obvious reasons. The pressure sensor was simulated using a powerful MEMS software called Coventorware�. This software allowed for the process, layout, 3D design and simulation of the device. The comb-drive works as a pressure sensor by actuation and fluctuation. When a pressure is applied to the free moving comb (the comb with a "T" at the end) a change in capacitance can be noticed between the interlaced fingers. This capacitance is quite substantial and can be measured, hence an effective pressure sensing device. Below are 3D outputs of the conceptual pressure sensor design:
 * //Figure 2: MEMS Pressure sensor 3D design//**

**Data Acquisition System:**
This is the primary focus of our project, the smart conditioning of the signal. To simulate the data acquisition system, a powerful software called Proteus� was used as it has the ability to simulate a microcontroller. But due to budget considerations, we were only able to use a demonstration version, however we were able to show promising results. Below is the conceptual smart sensor system that takes an analogue input, appropriately conditions it and converts it to digital, ready for transmission or display. //**Figure 3: Smart DAQ simulation schematic**//

There were however a number of limitations of software simulation. The first of which is that each individual component and device simulated are based on ideal mathematical equations. This caused many errors and design alterations in our project. Another was the limited simulation models available, this impeded our success as we were unable to simulate various devices such like a switched-capacitor filter (programmable filter). In response, we chose alternatives that would simulate similarly so that our results would not deviate largely. To show that our smart sensor system was conventionally better than a standard sensor system, a comparison was made. A traditional/standard sensor is simply the output of the sensor with out any conditioning (setting gain, complex filtering etc.), in the our simulation however, we also included a standard noise filter and amplifier, similar to practical use. Below is a simulation output of a standard sensor system: //**Figure 4: Standard/traditional sensor output**//

In the above screenshot, in the VSM Oscilloscope of Proteus� the bottom waveform is the noisy input signal while the waveform above it is the output of the standard sensor. It can be seen that a lot of the noise has been eliminated but still present in the signal, the gain has also been increased. In the VSM Logic Analyser window, it shows the digital data that will be transmitted or stored. The standard sensor system does a less than sufficient job at removing noise and amplifying the signal. Below is our proposed smart sensor system output: //**Figure 5: Smart DAQ simulation comparison**//

By looking at the smart sensor system VSM Oscilloscope output, the output signal is much more clean and having a more appropriate amplification. The VSM Logic Analyser of the original and conditioned output also shows that more data can be accumulated from that signal (by comparison). This is a quick comparison that shows a smart sensor system performs better than a standard sensor system with regard to noise filtering, amplification, conditioning, data storage and transmission.

**Project: Phase II - Hardware demonstration**
To show that the above system is actually feasible, a physical electronic demonstration shall be done. This will demonstrate similar or better features of a smart microelectronic system. The device used to represent the simulation was a very intelligent microcontroller that had the all the abilities the simulated DAQ had (and more). The controller chosen was the ADuC834 microcontroller from **Analog Device**s. This was actually specifically designed for use with sensor inputs, thus very appropriate for the project.

Prototyping was first done using a chip converter, programmer, bread board and a PC. Below in figure 6 was the prototyping devices: //**Figure 6 Bread board prototyping of the sensor node**// Once a prototype was complete, a PCB design that integrates all the devices and components was designed and fabricated. All parts were then soldered and put together to form the final sensor node. Below in figure 7 shows the depiction of the wireless sensor node:

//**Figure 7 (left to right) Wireless sensor node, transmitter (and prototype) and Labview display interface**//

The wireless sensor node communicates through a wireless transceiver back to a PC, where **Labview** was utilised as a graphical user interface that visually shows the signals and information the on board sensors collects.

This encompasses the broad idea of the project which takes it a step further from the software simulation by representing a intelligent sensor system for healthcare applications into real-life terms. For more information, please see the contacts page and feel free to download the presentation slides.