FIR Compiler 7.2 - 2022.1 English

Vitis Model Composer User Guide (UG1483)

Document ID
UG1483
Release Date
2022-05-26
Version
2022.1 English

This Xilinx FIR Compiler block provides users with a way to generate highly parameterizable, area-efficient, high-performance FIR filters with an AXI4-Stream-compliant interface.

AXI Ports that are Unique to this Block

This block exposes the AXI CONFIG channel as a group of separate ports based on sub-field names. The sub-field ports are described as follows:

Configuration Channel Input Signals:

config_tdata_fsel A sub-field port that represents the fsel field in the Configuration Channel vector. fsel is used to select the active filter set. This port is exposed when the number of coefficient sets is greater than one. Refer to the FIR Compiler V7.2 Product Guide for an explanation of the bits in this field.

Block Parameters

The block parameters dialog box can be invoked by double-clicking the icon in your Simulink® model.

Filter Specification tab
Parameters specific to the Filter Specification tab are as follows.
Filter Coefficients
Coefficient Vector
Specifies the coefficient vector as a single MATLAB® row vector. The number of taps is inferred from the length of the MATLAB® row vector. If multiple coefficient sets are specified, then each set is appended to the previous set in the vector. It is possible to enter these coefficients using the FDATool block as well.
Number of Coefficients Sets
The number of sets of filter coefficients to be implemented. The value specified must divide without remainder into the number of coefficients.
Use Reloadable Coefficients
Check to add the coefficient reload ports to the block. The set of data loaded into the reload channel will not take action until triggered by a re-configuration synchronization event. Refer to the FIR Compiler V7.2 Product Guide for a more detailed explanation of the RELOAD Channel interface timing. This block supports the xlGetReloadOrder function. See the Model Composer Utility function xlGetReloadOrder for details.
Filter Specification
Filter Type
Single_Rate
The data rate of the input and the output are the same.
Interpolation
The data rate of the output is faster than the input by a factor specified by the Interpolation Rate value.
Decimation
The data rate of the output is slower than the input by a factor specified in the Decimation Rate Value.
Hilbert
Filter uses the Hilbert Transform.
Interpolated
An interpolated FIR filter has a similar architecture to a conventional FIR filter, but with the unit delay operator replaced by k-1 units of delay. k is referred to as the zero-packing factor. The interpolated FIR should not be confused with an interpolation filter. Interpolated filters are single-rate systems employed to produce efficient realizations of narrow-band filters and, with some minor enhancements, wide-band filters can be accommodated. The data rate of the input and the output are the same.
Rate Change Type
This field is applicable to Interpolation and Decimation filter types. Used to specify an Integer or Fixed_Fractional rate change.
Interpolation Rate Value
This field is applicable to all Interpolation filter types and Decimation filter types for Fractional Rate Change implementations. The value provided in this field defines the up-sampling factor, or P for Fixed Fractional Rate (P/Q) resampling filter implementations.
Decimation Rate Value
This field is applicable to the all Decimation and Interpolation filter types for Fractional Rate Change implementations. The value provided in this field defines the down-sampling factor, or Q for Fixed Fractional Rate (P/Q) resampling filter implementations.
Zero pack factor
Allows you to specify the number of 0’s inserted between the coefficient specified by the coefficient vector. A zero packing factor of k inserts k-1 0s between the supplied coefficient values. This parameter is only active when the Filter type is set to Interpolated.
Channel Specification tab
Parameters specific to the Channel Specification tab are as follows.
Interleaved Channel Specification
Channel Sequence
Select Basic or Advanced. See the FIR Compiler LogiCORE IP Product Guide (PG149) for an explanation of the advanced channel specification feature.
Number of Channels
The number of data channels to be processed by the FIR Compiler block. The multiple channel data is passed to the core in a time-multiplexed manner. A maximum of 64 channels is supported.
Sequence ID List
A comma delimited list that specifies which channel sequences are implemented.
Parallel Channel Specification
Number of Paths
Specifies the number of parallel data paths the filter is to process. As shown below, when more than one path is specified, the data_tdata input port is divided into sub-ports that represent each parallel path.
Figure 1. Number of Paths

Hardware Oversampling Specification
Select format
Maximum_Possible
Specifies that oversampling be automatically determined based on the din sample rate.
Input_Sample_Period/Output_Sample_Period
Activates the Sample period dialog box below. Enter the Sample Period specification. Selecting this option exposes the s_axis_data_tvalid port (called ND port on earlier versions of the core). With this port exposed, no input handshake abstraction and no rate-propagation takes place.
Hardware Oversampling Rate
Activates the Hardware Oversampling Rate dialog box. Enter the Hardware Oversampling Rate specification below.
Hardware Oversampling Rate
The hardware oversampling rate determines the degree of parallelism. A rate of one produces a fully parallel filter. A rate of n (resp., n+1) for an n-bit input signal produces a fully serial implementation for a non-symmetric (resp., symmetric) impulse response. Intermediate values produce implementations with intermediate levels of parallelism.
Implementation tab
Parameters specific to the Implementation tab are as follows.
Coefficient Options
Coefficient Type
Specify Signed or Unsigned.
Quantization
Specifies the quantization method to be used for quantizing the coefficients. This can be set to one of the following:
  • Integer_Coefficients
  • Quantize_Only
  • Maximize_Dynamic_Range
  • Normalize_to_Centre_Coefficient
Coefficient Width
Specifies the number of bits used to represent the coefficients.
Best Precision Fractional Bits
When selected, the coefficient fractional width is automatically set to maximize the precision of the specified filter coefficients.
Coefficient Fractional Bits
Specifies the binary point location in the coefficients datapath options.
Coefficients Structure
Specifies the coefficient structure. Depending on the coefficient structure, optimizations are made in the core to reduce the amount of hardware required to implement a particular filter configuration. The selected structure can be any of the following.
  • Inferred
  • Non-Symmetric
  • Symmetric
  • Negative_Symmetric
  • Half_Band
  • Hilbert

The vector of coefficients specified must match the structure specified unless Inferred from coefficients is selected in which case the structure is determined automatically from these coefficients.

Datapath Options
Output Rounding Mode
Choose one of the following.
  • Full_Precision
  • Truncate_LSBs
  • Non_Symmetric_Rounding_Down
  • Non_Symmetric_Rounding_Up
  • Symmetric_Rounding_to_Zero
  • Symmetric_Rounding_to_Infinity
  • Convergent_Rounding_to_Even
  • Convergent_Rounding_to_Odd
Output Width
Specify the output width. Edit box activated only if the Rounding mode is set to a value other than Full_Precision.
Detailed Implementation tab
Parameters specific to the Detailed Implementation tab are as follows.
Filter Architecture
The following two filter architectures are supported.
  • Systolic_Multiply_Accumulate
  • Transpose_Multiply_Accumulate
    Note: When selecting the Transpose Multiply-Accumulate architecture, these limitations apply:
    • Symmetry is not exploited. If the Coefficient Vector specified on the Filter Specification tab is detected as symmetric, the FIR Compiler 7.2 block parameters dialog box will not allow you to select Transpose Multiply Accumulate.
    • Multiple interleaved channels are not supported.
Optimization Options

Specifies if the core is required to operate at maximum possible speed (“Speed” option) or minimum area (“Area” option). The “Area” option is the recommended default and will normally achieve the best speed and area for the design, however in certain configurations, the “Speed” setting might be required to improve performance at the expense of overall resource usage (this setting normally adds pipeline registers in critical paths).

Goal
  • Area
  • Speed
  • Custom
List
A comma delimited list that specifies which optimizations are implemented by the block. The optimizations are as follows.
Data_Path_Fanout
Adds additional pipeline registers on the data memory outputs to minimize fan-out. Useful when implementing large data width filters requiring multiple DSP slices per multiply-add unit.
Pre-Adder_Pipeline
Pipelines the pre-adder when implemented using fabric resources. This may occur when a large coefficient width is specified.
Coefficient_Fanout
Adds additional pipeline registers on the coefficient memory outputs to minimize fan-out. Useful for Parallel channels or large coefficient width filters requiring multiple DSP slices per multiply-add unit.
Control_Path_Fanout
Adds additional pipeline registers to control logic when Parallel channels have been specified.
Control_Column_Fanout
Adds additional pipeline registers to control logic when multiple DSP columns are required to implement the filter.
Control_Broadcast_Fanout
Adds additional pipeline registers to control logic for fully parallel (one clock cycle per channel per input sample) symmetric filter implementations.
Control_LUT_Pipeline
Pipelines the Look-up tables required to implement the control logic for Advanced Channel sequences.
No_BRAM_Read_First_Mode
Specifies that Block RAM READ-FIRST mode should not be used.
Increased speed
Multiple DSP slice columns are required for non-symmetric filter implementations.
Other
Miscellaneous optimizations.
Note: All optimizations maybe specified but are only implemented when relevant to the core configuration.
Memory Options
The memory type for MAC implementations can either be user-selected or chosen automatically to suit the best implementation options. Note that a choice of “Distributed” might result in a shift register implementation where appropriate to the filter structure. Forcing the RAM selection to be either Block or Distributed should be used with caution, as inappropriate use can lead to inefficient resource usage - the default Automatic mode is recommended for most applications.
Data Buffer Type
Specifies the type of memory used to store data samples.
Coefficient Buffer Type
Specifies the type of memory used to store the coefficients.
Input Buffer Type
Specifies the type of memory to be used to implement the data input buffer, where present.
Output Buffer type
Specifies the type of memory to be used to implement the data output buffer, where present.
Preference for other storage
Specifies the type of memory to be used to implement general storage in the datapath.
DSP Slice Column Options
Multi-Column Support
For device families with DSP slices, implementations of large high speed filters might require chaining of DSP slice elements across multiple columns. Where applicable (the feature is only enabled for multi-column devices), you can select the method of folding the filter structure across the multiple-columns, which can be Automatic (based on the selected device for the project) or Custom (you select the length of the first and subsequent columns).
Column Configuration
Specifies the individual column lengths in a comma delimited list. (See the data sheet for a more detailed explanation.)
Inter-Column Pipe Length
Pipeline stages are required to connect between the columns, with the level of pipelining required being depending on the required system clock rate, the chosen device and other system-level parameters. The choice of this parameter is always left for you to specify.
Interface tab
Data Channel Options
TLAST
TLAST can either be Not_Required, in which case the block will not have the port, or Vector_Framing, where TLAST is expected to denote the last sample of an interleaved cycle of data channels, or Packet_Framing, where the block does not interpret TLAST, but passes the signal to the output DATA channel TLAST with the same latency as the datapath.
Output TREADY
This field enables the data_tready port. With this port enabled, the block will support back-pressure. Without the port, back-pressure is not supported, but resources are saved and performance is likely to be higher.
Input FIFO
Selects a FIFO interface for the S_AXIS_DATA channel. When the FIFO has been selected, data can be transferred in a continuous burst up to the size of the FIFO (default 16) or, if greater, the number of interleaved data channels. The FIFO requires additional FPGA logic resources.
TUSER
Select one of the following options for the Input and the Output.
Not_Required
Neither of the uses is required; the channel in question will not have a TUSER field.
User_Field
In this mode, the block ignores the content of the TUSER field, but passes the content untouched from the input channel to the output channels.
Chan_ID_Field
In this mode, the TUSER field identifies the time-division-multiplexed channel for the transfer.
User and Chan_ID_Field
In this mode, the TUSER field will have both a user field and a chan_id field, with the chan_id field in the least significant bits. The minimal number of bits required to describe the channel will determine the width of the chan_id field, e.g. 7 channels will require 3 bits.
Configuration Channel Options
Synchronization Mode
On_Vector
Configuration packets, when available, are consumed and their contents applied when the first sample of an interleaved data channel sequence is processed by the block. When the block is configured to process a single data channel configuration packets are consumed every processing cycle of the block.
On_Packet
Further qualifies the consumption of configuration packets. Packets will only be consumed once the block has received a transaction on the s_axis_data channel where s_axis_data_tlast has been asserted.
Configuration Method
Single
A single coefficient set is used to process all interleaved data channels.
By_Channel
A unique coefficient set is specified for each interleaved data channel.
Reload Channel Options
Reload Slots
Specifies the number of coefficient sets that can be loaded in advance. Reloaded coefficients are only applied to the block once the configuration packet has been consumed (Range 1 to 256).
Control Options
ACLKEN
Active-high clock enable. Available for MAC-based FIR implementations.
ARESETn (active low)
Active-low synchronous clear input that always takes priority over ACLKEN. A minimum ARESETn active pulse of two cycles is required, since the signal is internally registered for performance. A pulse of one cycle resets the control and datapath of the core, but the response to the pulse is not in the cycle immediately following.
Advanced tab
Block Icon Display
Display shortened port names

On by default. When unchecked, data_tvalid, for example, becomes m_axis_data_tvalid.

Other parameters used by this block are explained in the topic Common Options in Block Parameter Dialog Boxes.

LogiCORE™ Documentation

FIR Compiler LogiCORE IP Product Guide (PG149)