Filters can be classified in several different groups, depending on what criteria are used for classification. The two major types of digital filters are finite impulse response digital filters (FIR filters) and infinite impulse response digital filters (IIR).
Both types have some advantages and disadvantages that should be carefully considered when designing a filter. Besides, it is necessary to take into account all fundamental characteristics of a signal to be filtered as these are very important when deciding which filter to use. In most cases, it is only one characteristic that really matters and it is whether it is necessary that filter has linear phase characteristic or not.
Speech signal, for example, can be processed in the systems with non-linear phase characteristic. The phase characteristic of a speech signal is not of the essence and as such can be neglected, which results in the possibility to use much wider range of systems for its prosessing.
There are also signals for which the phase characteristic is of the essence. A typical example are signals obtained from various sensors in industry. Therefore, it is necessary that a filter has linear phase characteristic to prevent loosing important information.
When a signal to be filtered is analysed in this way, it is easy to decide which type of digital filter is best to use. Accordingly, if the phase characteristic is of the essence, FIR filters should be used as they have linear phase characteristic. Such filters are of higher order and more complex, therefore. Otherwise, when it is only frequency response that matters, it is preferable to use IIR digital filters which have far lower order, i.e. are less complex, and thus much easier to realize.
The basic characteristics of Finite Impulse Response (FIR) filters are:
linear phase characteristic;
high filter order (more complex circuits); and
The basic characteristics of Infinite Impulse Response (IIR) are:
non-linear phase characteristic;
low filter order (less complex circuits); and
resulting digital filter has the potential to become unstable.