During vibration of engineering structures, fatigue cracks may exhibit repetitive crack open-close breathing like phenomenon which ultimately result in a distinct crack type, breathing cracks.This breathing phenomenon generates bi-linearity and irregularities in vibration signals of the cracked structure which carry useful information about the crack occurrence. In this thesis, the concept of entropy is employed to quantify this bi-linearity/irregularity of the vibration response so as to evaluate crack severity. To increase the sensitivity of the entropy calculation to detect the damage severity, sample entropy and quantized approximation of sample entropy are merged with wavelet transformation (WT) which is capable of amplifying the weak irregularities in vibration signal caused by small and initial breathing cracks. A cantilever beam with a breathing crack is studied to asses proposed crack identification method under two vibration conditions with sinusoidal and random excitations. An iterative numerical model is established to generate accurate time domain vibration responses of the cantilever with a breathing crack. Through both numerical simulations and experimental testing, the breathing crack identification with entropy under sinusoidal excitation is studied first and proven to be effective. Then, the crack identification sensitivity under lower excitation frequencies is further improved by parametric optimization of sample entropy and WT. Finally, effective breathing crack identification under general random excitations are experimentally studied and realized using frequency response functions (FRFs) which adapts the proposed crack identification technique to the incurred extra complexity due to random nature of the excitation and structural response.iii