Datasheet

dsPIC30F
DS70043F-page 52 © 2005 Microchip Technology Inc.
TABLE 12-4: FUNCTION EXECUTION TIMES
12.4 dsPICworks™ Data Analysis and
DSP Software
dsPICworks is a free data analysis and signal
processing package for use with Microsoft Windows
9x, NT, 2000 and XP platforms. It provides an extensive
number of functions encompassing:
• Wide variety of signal generators – Sine, Square,
Triangular, Window functions, Noise
• Extensive DSP functions – FFT, DCT, Filtering,
Convolution, Interpolation
• Extensive arithmetic functions – Algebraic
expressions, data-scaling, clipping etc.
• 1-D, 2-D and 3-D displays
• Multiple data quantization and saturation options
• Multi-channel data support
• Automatic “script file”-based execution options
available for any user-defined sequence of
dsPICworks functions
• File Import/Export interoperable with MPLAB IDE
• Digital filtering options support filters generated by
dsPIC Filter Design
• ASM30 assembler file option to export data tables
into dsPIC30F RAM.
FIGURE 12-1: dsPICworks Data
Analysis and DSP
Software
12.4.1 SIGNAL GENERATION:
dsPICworks™ Data Analysis and DSP Software
supports an extensive set of signal generators
including basic sine, square and triangle wave
generators as well as advanced generators for window
functions, unit step, unit sample, sine, exponential and
noise functions. Noise, with specified distribution, can
be added to any signal. Signals can be generated as 32
bit floating-point or as 16-bit fractional fixed point
values for any desired sampling rate. The length of the
generated signal is limited only by available disk space.
Signals can be imported or exported from or to MPLAB
file-register windows. Multi-channel data can be
created by a set of multiplexing functions.
Function
Cycle Count
Equation
Conditions
(1)
Number of
Cycles
(2)
Execution Time
@30 MIPS
Complex FFT
(3)
- N=64 3739 124.6 μs
Complex FFT
(3)
- N=128 8485 282.8 μs
Complex FFT
(3)
- N=256 19055 635.2 μs
Block FIR 53+N(4+M) N=32, M=32 1205 40.2 μs
Block FIR Lattice 41+N(4+7M) N=32, M=32 7337 244.6 μs
Block IIR Canonic 36+N(8+7S) N=32, S=4 1188 39.6 μs
Block IIR Lattice 46+N(16+7M) N=32, M=8 2350 78.3 μs
Matrix Add 20+3(C*R) C=8, R=8 212 7.1 μs
Matrix Transpose 16+C(6+3(R-1)) C=8, R=8 232 7.7 μs
Vector Dot Product 17+3N N=32 113 3.8 μs
Vector Max 19+7(N-2) N=32 229 7.6 μs
Vector Multiply 17+4N N=32 145 4.8 μs
Vector Power 16+2N N=32 80 2.7 μs
Note 1: C = #columns, N = # samples, M = #taps, S = #sections, R = #rows
2: 1 cycle = 33 nanoseconds @30 MIPS
3: Complex FFT routine inherently prevents overflow