Filter Category Block Listing

The following blocks are included in the Filter category :

• Adaptive Equalizer
• File FIR Filter
• FIR Filter
• IIR Filter
• MagPhase Filter
• Pulse Shaping Filter
• Sampling File FIR Filter
• Sampling FIR Filter

Adaptive Equalizer

This block implements a conventional or fractionally-spaced adaptive equalizer suitable for use with both analog and digital waveforms. Two versions of this block exist, one complex and the other real.

This block uses a Least Mean Square (LMS) convergence algorithm to adapt the tap values with the goal of minimizing the average error. The user is responsible for providing a suitable error input to the block, for example the error vector between a received point in the IQ plane and the closest constellation point. Taps may be initialized either internally (sets all taps to zero except for a user specified tap) or by using the external vector input.

File FIR Filter

This block implements a Finite Impulse Response (FIR) filter based on user-supplied tap values provided in a file. Up to 32,767 taps can be specified. Because the input file represents the impulse response of the user-specified filter, this block can also be viewed as performing a convolution of the input signal with the truncated waveform specified by the input file.

FIR Filter

This block implements a Finite Impulse Response (FIR) filter. It employs the windowing method for filter design and allows the user to implement lowpass, highpass, bandpass, bandstop, raised cosine, root raised cosine, Hilbert and Gaussian filters with a choice of window function.

The effective sampling frequency of the filter can be specified to obtain a consistent filter response regardless of the simulation sampling rate. When the filter sampling frequency is a fraction of the simulation sampling rate, additional delay elements are introduced in the internal shift register between the filter’s active tap locations.

A Filter Viewer is provided to visualize the impulse response, gain, and phase characteristics of the filter.

IIR Filter

This block implements an Infinite Impulse Response (IIR) filter. The user can choose from several analog filter prototypes, including Butterworth, Chebyshev, Elliptical and Bessel designs. The desired filter is implemented using the bilinear transformation to map the s-domain analog design into the digital z-domain.

A Filter Viewer is provided to visualize the impulse response, gain, and phase characteristics of the filter.

MagPhase Filter

This block implements an arbitrary complex FIR filter based on user-specified magnitude and phase responses supplied via an external file. The magnitude response is specified in decibels, while the phase response may be provided in either degrees, radians, or as a group delay.

The filter is realized using the overlap-save method. The input signal is mapped to the frequency domain via FFT, multiplied by the specified frequency response, and then mapped back to the time domain via IFFT. The size of the FFT is twice that of the equivalent complex FIR filter tap length.

The input file should include points from -fs/2 to +fs/2 (0 to +fs/2 for a real filter) in increasing order, but the frequency spacing need not be uniform. Linear interpolation is used to compute intermediate points as required.

Pulse Shaping Filter

This block implements pulse shaping using a finite impulse response (FIR) approach. It supports the use of a variety of windowing shapes, including Nyquist type pulse shapes, such as raised cosine and root raised cosine forms, and Gaussian pulse shapes.

The Pulse Shaping Filter block expects an impulse pulse train representing the input symbol values. If a rectangular input signal is provided instead (e.g. NRZ data), an inverse sinc function can be applied to convert - internally to the block - the rectangular pulse train into an impulse one.

The block normally introduces a delay equal to N / 2 simulation steps, where N is the number of filter taps. The number of taps is equal to the pulse span interval times the number of samples per symbol.

Sampling File FIR Filter

This block implements a sampling Finite Impulse Response (FIR) filter based on user-supplied tap values provided in a file. Up to 32,767 taps can be specified. The block will sample the input signal at the specified rate and propagate these values through its internal shift register. The Sampling File FIR output signal can be either held or interpolated when the filter sampling rate is below the simulation rate. A filter delay adjustment is provided to allow fine tuning of the sampling instant.

Sampling FIR Filter

This block implements a sampling Finite Impulse Response (FIR) filter. The block will sample the input signal at the specified rate and propagate these values through its internal shift register. The Sampling FIR output signal can be either held or interpolated when the filter sampling rate is below the simulation rate. A filter delay adjustment is provided to allow fine tuning of the sampling instant.

The Sampling FIR block employs the windowing method for filter design and allows the user to implement lowpass, highpass, bandpass, bandstop, raised cosine, root raised cosine, Hilbert and Gaussian filters with a choice of window function.

The effective sampling frequency of the filter can be specified so as to obtain a consistent filter response regardless of the simulation sampling rate. When the filter sampling frequency is a fraction of the simulation sampling rate, additional delay elements are introduced in the internal shift register between the filter’s active tap locations.

A Filter Viewer is provided to visualize the impulse response, gain, and phase characteristics of the filter.