In control systems, system identification methods are a powerful tool for analyzing and improving the performance of industrial processes. In most control methods, a model of the system must first be available to control the system. But unfortunately sometimes the system model is not specified or it is expressed in a parametric form in which the parameters are uncertain. Therefore, we need some techniques to identify the parameters of the system. For this reason, parametric identification of systems is one of the most important issues in control systems theory and for uncertain systems it has become a standard tool for control engineers. In practice, some physical systems such as heating furnaces, hydraulic ducts, rotor skin effect of induction machines, elastoplastic behavior of metals and electrical characteristics of solid oxide fuel cells are described by fractional order models. So to better control of such systems, fractional order identification and control methods are required. However, in recent years, some progress has been made in identifying fractional order systems, there are very few references that can cover this topic well. Also the proposed methods often have limitations such as limitation on the input signal, limitation on identifying the system delay and limitation on being commensurate order system. In this research, the parametric identification of linear systems with time delay, which are described by fractional order transfer functions or fractional order parametric differential equations, is investigated. For this purpose, the delayed fractional order system is first approximated by a delayed integer order system using the Pade approximation method for the fractional order operators in the transfer function. This approximation greatly reduces the complexity of the identification process and then by applying this approximation and block pulse functions a method is proposed to determine the unknown parameters of the transfer function so as to overcome the above limitations. Keywords : system identification, fractional order, time delay, parameter estimation, Pade approximation