Different regions of electromagnetic spectrum can be used in remote sensing and earth surface imaging. Synthetic aperture Radar (SAR) using microwave region brings out several benefits including capability to all weather imaging even in fog and cloud, day and night imaging without need to external source, and high resolution imaging even in long ranges. Therefore it is widely used whether in civil or military applications. Bistatic SAR (BSAR) is characterized by different locations for transmitter and receiver. There are several benefits with respect to monostatic SAR systems like increased reliability and flexibility, reduced vulnerability for military applications, reduced costs using existing illuminators of opportunity with several receive-only systems, more scattering information acquiring, and capacity of forward- or backward-looking SAR imaging. Even for objects that show a low Radar Cross Section (RCS) in monostatic SAR, one can find distinct bistatic angles to increase their bistatic RCS. All the advantages above are however paid with the price of increased processing complexity. BSAR system geometry can be categorized in the three main configurations: the azimuth invariant BSAR with platforms of the same velocity along parallel directions, the parallel BSAR with platforms of different velocity and the general BSAR with arbitrary directions and velocities of transmitter and receiver. Degree of complexity to process these configurations increases respectively. In SAR raw data, the energy of a point target spreads in both range and azimuth, and no information can be obtained without further processing. Thus, processing these raw data and image formation is a significant subject. Image formation algorithms can work in time or frequency. Time domain methods can focus BSAR data well; however these methods suffer from inefficiency problem. Efficiency can be improved by performing the focusing in the frequency domain. Omega-k is a known focusing algorithm characterized by its precision. It is preferred by many SAR researchers because of its capability to process wide apertures and highly squinted cases. The Bistatic Point Target Reference Spectrum (BPTRS) is the basis for most frequency domain algorithms. The accuracy of an algorithm is limited by the accuracy of its spectrum. There are different implementations of an algorithm based on the spectrum used. MELBF is an accurate bistatic spectrum, presented recently by Loffeld and his group. It can be applied within any image formation algorithm. In this thesis, we propose a new method by implementing omega-k algorithm based on MELBF spectrum that is applicable for all bistatic configurations. We present two different approaches based on accurate Stolt interpolation and efficient approximate interpolation-free ISFT. First, we show the validity of our method in different BSAR configurations. Then we show that our method gives better results in terms of image quality measurements compared to an existing omega-k algorithm based on ELBF, and simulation results confirms this claim. In another contribution, we investigated the use of Chebyshev orthogonal polynomials in deriving the bistatic spectrum. As using Taylor and Legendre expansion is reported in literature, we propose to add a third expansion to this group, that in some cases gives better results. Key Words SAR, Bistatic, Image formation, Spectrum, Omega-k, Orthogonal polynomials, Chebyshev