Pseudorandom sequences and the measurement of the frequency response [instrumentation notes]
Shlomo Engelberg, Haim Benjamin · IEEE Instrumentation & Measurement Magazine · 2005
A fundamental characteristic of a linear system is the system’s frequency response, T( f ). For each frequency, f, the frequency response T( f ) is a complex number. If Msin(2 π f t) is input to a system, the steady-state output of the system is |T( f )|Msin(2 π f t+ ∠ T( f )). There are many ways to measure the magnitude of the frequency response of a system. We consider three methods in the next section and explain why they are not optimal. Then we consider a method that uses pseudorandom sequences and explain why this method is optimal. We discuss pseudorandom sequences and show how their properties make them nearly ideal for measuring the magnitude of the frequency response of a system. We describe several implementations of pseudorandom sequence generators. Finally, we provide an example of a measurement made using a pseudorandom sequence. Pseudorandom sequences are used in audio applications to measure the properties of loudspeakers and of rooms. (See [7] and the references therein.) Additionally pseudorandom sequences are made use of in direct sequence spread spectrum [2].