Road safety is one of the priorities for every government in Europe and including in the UK. One of the major concerns for road safety is the consumption of alcohol by drivers which seriously increases the risk of an accident. According to the department of transport in 2007, there were over 17,000 drink-drive casualties including 530 fatalities in England alone. For this reason, the police force needs a fast and easy way to assess the alcohol level of drivers. We propose a new method of detection, using microwaves, which would detect the alcohol-blood ratio from the finger/wrist of the driver. We also study the possibility of blood sugar detection for diabetes using microwave resonators. The system is based on microwave resonators using very low power sources (lmW). This PhD project explores the limits and feasibility of a rectangular cavity resonator and a microstrip suspended ring resonator for use with water/ethanol samples and water/glucose samples. Samples of water/ethanol are tested in three parts for each sensor, 0% (water) to 100% (ethanol) in 5% increments, 1% to 5% in 1% increments and samples less than to 1% in 0.2% increments. Samples of water/glucose are tested from Omol of glucose to 1mol per litre of water in O.lmol/1 increments. Both systems are studied, designed, simulated and tested for the full ranges of both mixtures. The data acquisition software has been written in C# in order to allow ease of data extraction and manipulation during the tests. Using variables such as the Qfactor, the resonant frequency and the reflection coefficient, the resonators can detect permittivity changes in the samples. The rectangular cavity is able to detect a lower limit of 1% of ethanol, and a tenth of a mole for water/glucose mixtures. The suspended ring resonator can detect down to 1% of ethanol using the Q-factor, the resonant frequency and the magnitude and down to 0.2% using the transmission magnitude at a fixed frequency. This method can also detect a lower limit of O.lmol/l for glucose/water mixtures.