Research On The Cancer Detection Technique With The UWB Microwave Based On Dispersive Algorithm

In recent years, the incidence and mortality rates of breast cancer are rising year byyear, seriously endangering the lives and health of women. The early detection of breastcancer can greatly extend the lives of patients. The physical basis for microwavedetection of breast cancer is the significant contrast in the dielectric properties of normaland malignant breast tissues. The large contrast exists at the earliest stage of tumordevelopment and draws the attention of microwave imaging technology. Confocalimaging algorithm is a kind of active microwave imaging. The algorithm is simple inprinciple and efficient in computation. But studies on the algorithm are built onnon-dispersive model, which is not reliable. On the basis of confocal imaging algorithm,this paper proposes tumor detection algorithm for dispersive tissues.The dispersion characteristics of the breast and the tumor are discussed. And thetwo-dimensional dispersive breast model in the XFDTD is demonstrated for simulatingthe microwave propagation in breast. This paper measures the optimistic workfrequency and the bandwidth of microwave source in dispersive breast model andshows how to fit the parameters of the single-pole Debye equation from the knowledgeof the dielectric parameters of the tissue at the different frequencies.The dispersion compensation algorithm is introduced on the basis of simpleconfocal imaging algorithm. According to the number of channels to which thealgorithm is applied, two signal processing methods are proposed based on thedispersion processing algorithm, namely, single-channel processing method andmulti-channel processing method. Both the algorithms and confocal imaging algorithmare practiced on the dispersive signal, respectively, to verify the feasibility of thedispersion processing algorithms. The imaging results of the various situations areobserved, including the difference of the number, the size of tumors and the dielectricparameters of the tissues. The robustness of the algorithm is discussed. In the abovemodels, the algorithm that this paper puts forward gets better imaging results in thesingle-tumor models. The interaction of signals from multi-tumor results in notaccurately detecting tumors in the multi-tumor models.