FIR Filters Explained

Finite Impulse Response (FIR) filters are a fundamental type of digital filter used extensively in signal processing. They are named 'Finite Impulse Response' because their output signal eventually returns to zero after a finite number of time steps when subjected to an impulse input. This characteristic is key to understanding their structure and behavior.

Understanding Non-Recursive vs. Recursive Filters

Digital filters can be broadly classified into two categories based on their structure: non-recursive and recursive.

• Non-recursive filters: The output at any given time is calculated solely based on the current and previous input samples. They do not use past output values in their computation. This structure implies a finite impulse response.
• Recursive filters: The output at any given time depends not only on the current and previous input samples but also on previous output samples. This creates a feedback loop, where the output influences future outputs. Recursive filters typically have an infinite impulse response (IIR).

The Role of Feedback in Filters

Feedback is a mechanism where a portion of the output signal is fed back and combined with the input signal. This is essential for recursive filter structures (like IIR filters) to achieve their characteristic infinite impulse response. Filters that do not employ feedback are known as feedforward filters.

Analyzing FIR Filter Characteristics

Let's evaluate the given statements concerning FIR filters:

• 1. FIR filters are non-recursive: This statement is true. FIR filters compute their output as a weighted sum of past and present input samples only. They do not use past output values, making them non-recursive. The general equation for an FIR filter is:

  $ y[n] = \sum_{i=0}^{N-1} b_i x[n-i] $

  where $y[n]$ is the output at time $n$, $x[n]$ is the input at time $n$, $b_i$ are the filter coefficients, and $N$ is the filter order. Notice the absence of $y[n-k]$ terms.
• 2. FIR filters use feedback: This statement is false. As explained above, feedback involves using past outputs. FIR filters, being non-recursive, do not incorporate feedback loops.
• 3. FIR filters are recursive: This statement is false. Recursion implies the use of past outputs, which FIR filters do not employ. Filters that use past outputs are typically IIR filters.
• 4. FIR filters do not adopt any feedback: This statement is true. This is a direct consequence of their non-recursive structure. All computations rely solely on the input sequence.

Conclusion on FIR Filter Properties

Based on the analysis, the correct characteristics of FIR filters are that they are non-recursive and do not use feedback. Therefore, statements 1 and 4 are accurate.