Like any other sensor the response time of ion mobility spectrometer (IMS) is important, especially when it is coupled to other analytical devices such as GC, HPLC or TLC. In this thesis, the time behavior of the IMS signals was evaluated under different conditions. At first, the IMS response to a gas sample (for anisole and ammonia), that was quickly injected by a mLit syringe was studied. The injection profile was assumed as a square pulse. However, the response was found in a Gaussian shape followed by a long tail. The Gaussian response was referred to distortion of the injected square pulse under diffusion in gas phase. The long tail was attributed to desorption of the sample from the internal surfaces of the injection port. Sample adsorbs on the walls of the injection port, as soon as it is injected, and releases after the injection is finished. In fact, the fresh carrier gas is flowing continuously and washes out the adsorbed sample slowly. The desorption kinetics is a first order reaction hence; it creates an exponential long tail in the response curve. The time behavior of solid samples (for papaverine and acridine) was also evaluated. Like gas sample, the response curve of the solid included a Gaussian peak, due to diffusion, and a long exponential decay due to desorption. However, the middle part of the response curve did not fit well. This new part, in comparison to the gas sample, was attributed to the evaporation of the solid sample from the injection needle. The evaporation process was assumed to be a first order desorption reaction. Based on the assumption, an extra fast exponential decay function was added to the response curve. In summary, the response curve of the solid sample included a Gaussian function, for the rise time, a fast exponential decay for the desorption of the sample from the injection needle, and a slow exponential decay for the desorption of the sample from the internal surfaces of the injection port. Although, in this model, the evaporation process was considered at constant temperature, the experimental data well agreed with the model. In reality the needle temperature rises exponentially to reach the injection port temperature. A more complete model was also presented considering a first order evaporation or desorption kinetics under an exponentially rising temperature, plus a long decay function. Experimental data, for different final temperatures were fitted to this new model and the activation energy of desorption and pre exponential factors were obtained. The fittings were very good and the results agreed well with the values reported in the literature. In the second part of this work the influence of the experimental factors such as; sample type, oven temperature, type and length of the carrier tubes and the amount of sample, on the response curve, were evaluated. The response time is shorter for samples with lower melting points. The length of the tail was also found to be proportional to the sample amount. The cell temperature had no considerable effect on the response curve. In the third part of this work, a new method was proposed to extend the working and dynamic range of the calibration curve. The method is based on collecting the signal over the whole acquisition time, even when the signal is saturated. The saturated part was corrected by extrapolating the exponentially decaying signal. This increased the linear dynamic range of the calibration curve from 10-400 ng to 10-6000 ng for papaverine.