Simulating a FIR filter

Hi,

I've found a lot of examples of how to use the fir1 function to calculate the FIR filter coeficients, but I need to define a input vector of samples, the coeficients and need as return the output vector.

I was able to use the filter function to simulate one sample, but I believe I should combine that with a conv function.

This is the diagram I need to simulate:

Where: x[n] is my input vector.

b[ tap_num ] is the coeficient vector.

y[n] is the output vector.

The "triangle" would be a multiplier between the input and b[ tap_num ];

As example, in a 5 tap filter, using an input like x = [ 0 0 0 1 1 1 1 1 ], and b = [ 1 1 1 1 1 ], the expected would be y = [ 1 2 3 4 5 4 3 2 ]

Whats the simpliest way to define the two input vectors ( x[n] and b[tap_num] ), and get an output vector?

Thank you!

Answer 1

[David Wilson]

Use the filter command. In this case your denominator, (A), coefficients are just 1.

x = [ 0 0 0 1 1 1 1 1 ]' % input  
b = [ 1 1 1 1 1 ];  

y = filter(b,1,x)

You will note that the solution above handles the initial conditions slightly differently than your proposed solution. However if this is important to you (hint: it shouldn't be unless this is a homework problem), then you need to use filtic to get them set to what you want.

[Rafael Ludtke]

The difference is because of the order I imagined the input vector would enter the filter, fixing that the output is what I was expecting.

Thank you!