Generated reference › Polynomial — Control Systems/Base Blocks
kind: generated#block#control-systems-base-blocks

Polynomial — Control Systems/Base Blocks

n

Control_Systems/Base_Blocks/Polynomial · 1 input / 1 output port(s) at insert · exports to Python, MATLAB, Java, Rust, C, C++, VHDL, Verilog, SystemVerilog, PLC Structured Text

Description#

The block's own DESCRIPTION_HTML, rendered verbatim — the same text the config dialog's info panel and the library navigator show. Fix a wrong sentence in the block's .cpp (R-D9), never here.

Polynomial

Control Systems / Base Blocks

Evaluates a polynomial at every entry of its input, entry by entry. The coefficients are given in descending powers, so [1 -2 0.5] is y = u² − 2u + 0.5.

Ports

  • Input – the signal u, of any size [m,n]. The polynomial is applied to each entry independently; this is not a matrix polynomial.
  • Output – the evaluated signal y, of the SAME size [m,n]. The block never reshapes a signal.

Parameters

  • Polynomial Coefficients – a row vector in descending powers, highest first, with the constant term last. Its length is the order plus one, so [1 0 0] is u², [3 -1] is 3u − 1, and a single value is a constant. At least one coefficient is required; leading zeros are allowed and simply lower the effective order.
  • Sampling Time (s) – zero or less inherits the solver's rate; a positive value runs the block at that period.

Code export

All ten targets: Python, MATLAB, Java, Rust, C, C++, VHDL, Verilog, SystemVerilog and PLC Structured Text. The coefficients are inlined into the generated arithmetic at export time rather than exposed as a tunable parameter: the number of them is what the emitted expression is shaped by, so changing one on the generated core would mean re-emitting it anyway.

Every target evaluates by Horner's rule – the nested form ((c₀u + c₁)u + c₂)u + c₃ – and not by summing powers. The two are not the same computation in floating point, so this is what keeps the ten targets agreeing with each other and with the block's own simulation.

The three HDL targets are simulation-only. Horner needs nothing but multiplies and adds, which the Q16.16 datapath has, but a polynomial multiplies the signal by itself once per order and that format saturates at ±32768 while resolving only 1.5×10⁻⁵. Rather than ship a core that silently overflows on a cubic, the generated HDL converts at the port boundary and evaluates in real – correct in simulation, but not offered as synthesizable.

Simulink bridge

Import and export, mapped to simulink/Math Operations/Polynomial. "Polynomial Coefficients" to Coefs, passed through as the vector it is – the descending-power convention is the same on both sides, so nothing is reordered crossing over.

The Simulink block has no SampleTime parameter, so the rate is not carried across: it stays on the ICore side, and a block configured with an explicit positive rate reports that the rate did not cross.

The two blocks' defaults differ, deliberately. Simulink's is a six-coefficient thermocouple fit; this block defaults to [1 0 0], which is u². Nothing is lost by that: the coefficients are always written out explicitly on export, so a block that crosses carries its own values either way.

Notes

  • Algebraic, with no state: the output depends only on the current input.
  • Deliberately carries no state space at any order. A polynomial of order two or more is not linear; a first-order one would be, but the order is a config the user can change at any time, and a state space that was valid for one setting and stale for the next is worse than none. Model reduction reports the block as unmergeable throughout.
  • The polynomial is applied entry by entry. A matrix polynomial – one using matrix multiplication for the powers – is a different operation and is not what this block or Simulink's computes.

Code facts#

FactValue
registered typeControl_Systems/Base_Blocks/Polynomial
familyControl_Systems/Base_Blocks
solver environment classICoreBlock_0_Control_Systems_1_Base_Blocks_2_Polynomial
sourcesrc/ICoreSDK/ICoreBlockLibrary/Blocks/Control_Systems/Base_Blocks/Polynomial/ICoreBlock_0_Control_Systems_1_Base_Blocks_2_Polynomial.cpp
headersrc/ICoreSDK/ICoreBlockLibrary/Blocks/Control_Systems/Base_Blocks/Polynomial/ICoreBlock_0_Control_Systems_1_Base_Blocks_2_Polynomial.h
default size on canvas90 × 70 px
ports at insert1 in, 1 out
code generators implementedPython, MATLAB, Java, Rust, C, C++, VHDL, Verilog, SystemVerilog, PLC Structured Text

Ports#

#DirectionSignal typeDescription label
1inICoreDouble
2outICoreDouble

Ports the constructor creates. A block whose port list changes with its configuration adds or removes ports at load time; the count above is the one a freshly inserted block has.

Configuration variables#

Config variableDefaultSimulink parameter
Polynomial Coefficients[1 0 0]Coefs

Every block also carries Sampling Time (s) from ICoreBlockSolverEnvironment: zero or less inherits the solver's rate, a positive value runs the block at that period.

supportSupport::Both
Simulink pathsimulink/Math Operations/Polynomial
port-count rulePortsParam::None
SampleTime parameterno — the counterpart defines none; the rate stays on the ICore side
ICore configSimulink parameterValue translation
Polynomial CoefficientsCoefspasses through

Caveat (shown to the user): the coefficients are in descending powers on both sides, so they cross unchanged; the two blocks' DEFAULTS differ - Simulink's is a six-coefficient thermocouple fit and this block's is [1 0 0] - which costs nothing, since the coefficients are always written out explicitly on export

Catalog contract: src/ICoreSDK/ICoreCoder/ICoreCommandSystem/SimulinkBridge/ICoreSimulinkBlockCatalog.h

Description vs code#

The lists agree. check_block_descriptions.py finds no disagreement between the description's Ports, Parameters, Code export and Simulink bridge lists and the code's.

The verdict above is tools/docs/check_block_descriptions.py (P7.1), which compares LISTS. It cannot read a sentence: "stateless" on a block with a state, an initial-value semantic the recursion does not implement, a "not synthesizable" caveat the HDL banner contradicts. That is the agent audit (P7.3) on BLOCK_DESCRIPTION_AUDIT.md, and this tool's green is not a substitute for one.

File banner (developer view)#

The top comment of the block's .cpp — the maths, the realization and the export strategy, addressed to whoever changes it. It must not contradict the description above (P7.5).

Polynomial -- a polynomial evaluated at every entry of the input The coefficients are in DESCENDING powers, exactly as MATLAB's polyval takes them, so [1 -2 0.5] is u^2 - 2u + 0.5. Measured against Simulink R2026a on [2 -1; 0.5 3] rather than assumed: its output and polyval's agree entry for entry, isequal true.

EVERY BACKEND USES HORNER'S RULE, and that is a correctness decision rather than a speed one. Nesting

((c0*u + c1)*u + c2)*u + c3

and summing c0*u^3 + c1*u^2 + c2*u + c3 are not the same computation in floating point. The reference this suite compares against is the C++ compute_h below, which nests; a target that summed powers instead would differ by a rounding error that grows with the order and with |u|, and on the HDL targets - where the quantum is 1.5e-5 rather than an ULP - it would not stay small. hornerExpr() writes the nest once and every target renders it.

HDL IS SIMULATION-ONLY here. Horner is only multiplies and adds, which the Q16.16 datapath can do - but a polynomial of order n multiplies the signal by itself n times, and Q16.16 saturates at +/-32768 while carrying only 1.5e-5 of resolution. A cubic on an input of a few units is already pushing both ends of that, so the generated cores convert at the port boundary and evaluate in real, which is correct in simulation but not offered as synthesizable. Reference for the same choice: Recursive IIR, Variable Transport Delay.

Algebraic and stateless. No state space -- see the header for why, including why the first-order case does not get one either.

Sample results#

Polynomial — Repeating Sequence Stair: [-2 -1 -0.5 0 0.5 1 2 3], one entry per samplePolynomial — Repeating Sequence Stair: [-2 -1 -0.5 0 0.5 1 2 3], one entry per sample05-2-10123inputoutput
tin ICoreDouble-Out-0out ICoreDouble-Out-0
0-24
0.40.50.25
0.8-24
1.20.50.25
1.6-24
20.50.25
2.4-24
2.80.50.25
3.2-24
3.60.50.25
4-24
4.40.50.25
4.8-24
5.20.50.25

Every 4th of 60 samples, from the table stimulus.

The same rig also ran:

StimulusWhat it isOutput range
impulseImpulse: one sample of 1 at k = 5, 0 elsewhere (Repeating Sequence Stair)0 … 1
rampRamp: slope 1 from t = 00 … 33.64
sineSine Wave: amplitude 1, 2 rad/s, no phase, no bias0 … 1
stepStep: 0 -> 1 at t = 1 s0 … 1

Plotted: table — Repeating Sequence Stair: [-2 -1 -0.5 0 0.5 1 2 3], one entry per sample

Category static · sample time 0.1 · 60 steps · commit ccf005c8 · produced by docsSample --out <folder> --steps 60 · data docs/generated/samples/Control_Systems__Base_Blocks__Polynomial.json · the SVG is generated from those numbers by tools/docs/plot_svg.py, so it is a run and not a drawing (R-D10).