DSPRelated.com

fred harris (@fharris)

Faculty at ECE Department at University of California San Diego specialty is DSP and Multirate Signal processing. Author of Book, Multirate Signal processing for Communication Systems, and co-author of Book Digital Communications

Re: How to calculate biquad cascade coefficient

Reply posted 1 year ago (05/15/2023)
If you have access to matlab use matlab to design the prototype filter, the use matlab's command tf2sos. see below! when using this script you may want to redistribute...
The alternating phases in the preamble is a perfect way to estimate the Doppler and then remove it. The matched filter for the alternating phase sequence then can...

Re: CIC implement question?

Reply posted 1 year ago (04/20/2023)
Dan,Traditionally the CIC is used to filter the output of a 1-bit ADC. There are no multiplies in a FIR filter operating on a 1-bit sequence. Thus a FIR filter can...

Re: CIC implement question?

Reply posted 1 year ago (04/18/2023)
Hello herbertLu,the advice I give to folks using the CIC filter is to not use the CIC filter. The integrators in the CIC are large and operate at the highest input...
hello six20star,First problem is what do you do if there is significant doppler offset due to moving platforms. Most FSK modems use a preamble to be used to detect...
There is an increase in delay starting sample to center (symmetry point) because you have gone through two filters. The workload has been reduced because the second...
Thanks for the nice feedback randy. I tell my students that my housekeeper doesn't do windows, but I do!have you played with the latest matlab script i put on line...
Hello all,Here it is: a tone signal in the spectral interval 1 kHz to 2 kHz. sampled at 100 kHz, Filter to extract tone, Passband 1-kHz to 2-kHz, stop band 0 to...
here is my script for the two stage polyphase filter. two versionsversion 1 a 10 path polyphase filter that performs a 10-to-1 down sampler from 100 kHz to 10 kHz....
A good rule to follow in DSP design is don't build filters when there is a large ratio of sample rate to BW. The response to this advice is to reduce the sample...
The IIR filter is even worse as an ill conditioned problem. Anytime there is a large ratio of sample rate to BW (100 to 1 here) the passband poles are packed very...
We need more constraints. Is the signal actually 1 kHz or is it an unknown frequency signal in the interval 1.0 to 2.0 kHz band? Is it a real signal or is it a...
John,The rabbit ears can by modified by controlled transitions in the pulses... controlled rise time and controlled fall time... makes amplifies less efficient....
Hello JohnnyM, The rabbit ears are the transient response of the channelizer bins (or offset frequency filters) to an input signal not in its bandwidth. Every filter...

Re: Calculating optimal loop point in audio sample

Reply posted 2 years ago (11/16/2022)
This is an interesting problem that is seen by people who operate an audio mixer system. You are trying to match continuity of amplitude and all orders of derivatives...
Measure instantaneous frequency from the complex signal obtained from output of hilbert transform filter... don't use x_hat=hilbert(x). form filter hilbert then...

Re: beamforming question.

Reply posted 2 years ago (08/21/2022)
I just thought of the third cause of your problem. The broadside beamwidth of a line anteanna is the ratio of antaenna length to wavelength (lamba) (L/lambda)....

Re: beamforming question.

Reply posted 2 years ago (08/21/2022)
What you may be running into is boundary conditions which we address with spacial windows called shading functions.The second problem in the near field the expanding...

Re: #OFDM material

Reply posted 2 years ago (07/25/2022)
Ok, i'll use it!fred

Re: #OFDM material

Reply posted 2 years ago (07/25/2022)
leonard, i can send you matlab script for full 802.11 ofdm modulator and demodulatorneed e-mail addressfred
hello puvanes,apparently you are doing IF sampling. the 120 MHz sample rate is a trip around the periodic circle. 60 MHz is half way around the circle so that an...

Re: FFT question

Reply posted 2 years ago (06/23/2022)
Paul,  Are the two sampling clocks the same phase and simultaneous? The phase shift between the ordered pair (I and Q) come from the cosine heterodyne and the sine...

Re: Determine Classical Phase Noise from EVM

Reply posted 3 years ago (03/01/2022)
here is paper Vector Signal Analyzer Implemented as a Synthetic Instrument.I can't attach paper,,, send me an email address and I'll pass it onfred h

Re: Determine Classical Phase Noise from EVM

Reply posted 3 years ago (03/01/2022)
see paper tagged below ( I coudn't attach it for some reason) It shows evm plots for different anomalies in modulation processVector Signal Analyzer Implemented...

Re: Viva Ukraine!

Reply posted 3 years ago (02/27/2022)
The entire world recognizes the actions of an misguided megalomaniac and the majestic and appraisable response of the Ukrainian nation and citizenship.  
raising the sample rate by inserting time domain zeros causes spectral replcates in the frequency domain. the low-pass filter suppresses the replicates leaving the...

Re: Extend FFT spectrum question

Reply posted 3 years ago (02/16/2022)
Questions!what is the sample rate? How did you pick up and paste spectral span to the offset span. What is the window used in the transform? How much overlapped...

Re: Multiframe FFT using single larger FFT

Reply posted 3 years ago (02/10/2022)
It is a pretty neat trick that i developed for the same reason you have.fred h

Re: Multiframe FFT using single larger FFT

Reply posted 3 years ago (02/10/2022)
Kris,yes you can do that!suppose you have two 64 point transforms you want to do but only have a 128 point FFT.imagine doing this: take the first time sequence of...
Hello hirnprinzI just finishd a search of my email and have not found the note from you. fjharris@eng.ucsd.edufred
hello hirnprinz,The design of the 2-path minimum phase filter came from discussions with Tony Constantinides. You have that copy in tonydes_2 (tony for anthony)....
hello Hirnprinz,me again (fred h) from my read -e file describing various all pass script filestwo are attachedtwo_path_compare.mtwo_path_demo.m tony_des_2 Graphical...
Hello Hirnprinz,The wave digtal filter is an all-pass filter. The various signal flow graph images in figure 9 will implement them. I have a fondness for the last...

Re: Optimal Cascaded FIR Decimator design

Reply posted 3 years ago (01/04/2022)
Hi Kaz,The adders in the polyphases are quite narrow, nowhere close to the width of the accumultors of a multistage CIC filter. The width of the coefficient sets...

Re: Optimal Cascaded FIR Decimator design

Reply posted 3 years ago (01/04/2022)
Hello andrew,I have looked at this problem a number of times and have reminded paeople tht when you are filting the output of a 1-bit sigm-delta, with a fir filter,...

Re: Fractionally Spaced Equalizer

Reply posted 3 years ago (12/21/2021)
If you have a burst transmission to a peer, you would use a preamble to teach the receiver what is the channel's impulse response or frequency response to initialize...

Re: Fractionally Spaced Equalizer

Reply posted 3 years ago (12/20/2021)
MitchSee if you can pick up a copy of my book, Multirate signal processing for communication systems.. second edition. I have a description of a graphical equalizer...

Re: Fractionally Spaced Equalizer

Reply posted 3 years ago (12/19/2021)
Hi Mitch,I think the original question addressed timing acquisition of a QAM receiver. Phase alignment of the receiver's clock sample times with the phase of the...

Re: Fractionally Spaced Equalizer

Reply posted 3 years ago (12/19/2021)
What you say about equalizers in your question is true. But the results won't be to your liking. A fractional spaced equalized can correct timing offsets as well...
Hello attskij, neirober is correct... what you are seeing is the side lobes from the spectral sinc in the negative frequency interacting with the main lobe of the...

Re: Using noise to increase resolution of ADC

Reply posted 3 years ago (11/22/2021)
Hello max maxfield try these matlab examples that show value of dither in time and frequencydds_dither.mmatlab demod_dither.m

Re: Using noise to increase resolution of ADC

Reply posted 3 years ago (11/22/2021)
hello maxfield,the noise you are adding is called dither. dither makes successive (correlated) measurements have different values in spite of the quantization.....
Rick,that's why we examine the signal and the spectrum on the circle where they are periodic.fred
Hello Graham,You never want to use a true even symmetric window. this is a window with a center point of symmetry and a matching left and right sample which gives...
What you want is part of the attached presentation... this is a parametric filter with independent control of amount of boos or cut, control of BW of the same boost...

Re: Nyquist frequency & odd even FFT

Reply posted 3 years ago (10/03/2021)
incidentally, My version of the Nyquist Theorem is that the sample rate should exceed the two sided bandwidth.... if that is interpreted correctly the sineusoid...

Re: Nyquist frequency & odd even FFT

Reply posted 3 years ago (10/03/2021)
Hello Kaz,The DFT works with an even or odd number of points.... Every sampled data sinewave has a continuous periodic frequency response. The DFT simply samples...

Re: Modulation in laser satellite communication

Reply posted 3 years ago (09/24/2021)
to demodulate phase information you have to phase lock to underlying carrier. It is fairly expensive to phase lock an optical carrier. Can use differential phase...

Re: FYI, Article on optical computation

Reply posted 3 years ago (09/24/2021)
Many years ago, the navy lab in SD, as well as the physics dept at SDSU, had active optical signal processing groups. They were using optics to perform convolution,...

Re: Complex IF

Reply posted 3 years ago (09/17/2021)
In a period a long time ago, we build modulators by adjusting the amplitude, phase, or frequency of a an analog carrier directly. We don't do that anymore! what...

Re: Help with polynomial zeros

Reply posted 3 years ago (08/07/2021)
Hello Napierm,Here is the problem: the factored form of the polynomial P(x) = product(x-root(k)) k=1,2,...., N  and the expanded form p(x) = sum a(n)x^(N-n), n=0,...

Re: ARM IIR Filter - Why No DAC Output?

Reply posted 3 years ago (07/16/2021)
the upper part of the sinewaves have overflowed and now appear at the lower range of the sinewaves. It looks as if you DC in your signal and not sufficient number...

Re: ARM IIR Filter - Why No DAC Output?

Reply posted 3 years ago (07/16/2021)
You have run into the problem that the biquadratric filter is ill conditioned for low pass filters. your bandwidth 1-millionth of the sample rate... you can not...

Re: PAPR vs Input Back-off in LTE

Reply posted 3 years ago (07/14/2021)
There are many ways to reduce PAPR in OFDM... SC-OFDM, which makes the cell links asymmetrical was invented to obtain reduced PAPR at the mobile.The reserve bin...

Re: Spreading Codes

Reply posted 3 years ago (07/14/2021)
You want to look at complementary code keying (CCK) and in particular non binary cck codes.The cck codes replaced the Barker spreading sequences in 802.11.the CCK...
Here is another unexpectedly simple approachsdr_2007_narrowband_IIR_filters.docfred h
here is paper on one way to build IIR low-pass filter with large ratio of sample rate to BWnarrowband IIR filters_again_2a.pdffred h
There is a way to build arbitrarily  narrow IIR filters.look at attached set of figures... see figure 5 to see sensitivity of root locations to coefficient bit...

Re: Digital IIR Parallel Implementation

Reply posted 3 years ago (06/26/2021)
this is another short paper comparing three versions of resampling filters in cascade. I am limited in what I can upload by page size restrictions... sorry about thatlook...

Re: Digital IIR Parallel Implementation

Reply posted 3 years ago (06/26/2021)
Sudarshan_onkar,On of the minor realities we often forget to teach undergraduate students is that the FIR filter can operate at rates that exceed the multiplier...
mtoddThe spectral product you describe with a brick wall filter has an interesting effect on the time domain versions of the spectrum. The two time functions are...

Re: FFT spectrum shift after time domain decimation

Reply posted 3 years ago (06/14/2021)
digital frequency is 2*pi*(fc/fs) with units of radians per sample if you reduce the sample rate by 2 the digital freq is 2*pi*(fc/(fs/2)) with units of radians/sample...

Re: Problem in Beam Steering

Reply posted 3 years ago (06/14/2021)
make the following changes% M=abs(M);M = 20*log10(abs ( M )) ;%polar ( t , M , '-r' ) ,plot(t,M,'r')fred h
Dear AiDrmerk,Small amounts of Doppler offset will be removed by the phase lock loop that tracks phase offsets from the timing sample positions identified by the...

Re: What is Self-noise in TED algorithm

Reply posted 3 years ago (06/04/2021)
Aidrmek,Self noise is the random transitions between states that can be seen in the eye diagram when you are offset in time from where the eye is maximally open....

Re: PDM - PCM CONVERSION AND THEIR CHARACTERISTICS

Reply posted 3 years ago (06/02/2021)
Naranguiz,I have worked on this problem for many years and have come to the conclusion that the sequence you described is not the correct approach to solve the...

Re: Filter approach for helicopter application

Reply posted 3 years ago (05/28/2021)
Dear coinmis, Is the signal you are modifying to remove spectral components at 4.3, 8.6, and 17.2 Hz part of the flight control system? If it is the transport...

Re: Overdriven Sine Wave through DSP Filter

Reply posted 3 years ago (05/28/2021)
Hi groger, I think you missed the thrust of my first response. You can not hard limit a sampled data signal and filter out the harmonics the same way you can an...

Re: Overdriven Sine Wave through DSP Filter

Reply posted 3 years ago (05/28/2021)
Hi Gary,You don't want to do what you say you are doing. A nonlinearity in the DSP domain will generate harmonics in the signal spectrum which will appear in the...

Re: Direct Form II

Reply posted 3 years ago (05/27/2021)
transfer functionY(Z)/X(Z) = [b0 + b1 Z^(-1) + b2 Z^(-2)]/[1 + a1 Z^(-1) +a2 Z^(-2)]add internal variable W(Z)[Y(Z)/(X(Z)] {W(Z)/W(Z)] =[Y(Z)/W(Z)] [W(Z)/X(Z)]where...

Re: Covariance Matrix diagonal or not?

Reply posted 3 years ago (05/25/2021)
The problem is you have not defined the interval over which you expect the sample values to be collected to form the correlation estimates.  Repeat your experiment...

Re: Joint Research in Base Band Signal Processing

Reply posted 3 years ago (05/09/2021)
mannai, start looking for papers by fred harris. One of my specialties in synchronization.also look for the 3rd edition id the sklar-harris book, Digital communications.we...

Re: DSP principles question

Reply posted 3 years ago (04/27/2021)
Rich,If you can find a copy of the "Multirate Signal processing for Communication Systems" on someone's desk, ask if you can borrow it and many of your questions...

Re: Recover signal from the real part of IFFT

Reply posted 3 years ago (04/23/2021)
Hi Frank_os,The spectrum of any real signal has a transform which is Hermetian symmetric. This means real part is even symmetric and the imaginary part is odd symmetric....

Re: parallel polyphase filters

Reply posted 3 years ago (04/18/2021)
hello cdprasad,I'm assuming you are building an analysis channelizer and you have designed the prototype with sufficiently narrow transition that it will not alias...

Re: SDR reception of legacy AM, FM waves

Reply posted 3 years ago (04/16/2021)
Hell0 DevBeeThe SDR radio can demodulate any analog modulation format as well as digital modulation options>If you have access to IEEE explore, look up paper...

Re: ADC can't perceive MLS (maximum length sequence)

Reply posted 4 years ago (04/06/2021)
Dear philipoakley,do you have some idea of the chip rate? you want to initially run your sampler at 4-samples per chip... not much faster or slower because the number...

Re: Measurement of frequency deviation in FM signals

Reply posted 4 years ago (03/28/2021)
Naumakila,It is easy to measure frequency deviation of a modulator if you can select the frequency and amplitude of a modulating sinewave. You can form the complex...

Re: phase error correction

Reply posted 4 years ago (03/18/2021)
Ali,are you phase aligning a sinewave of a local DDS with that of a received complex sinewave? Or are you phase locking to a modulated signal with out an underlying...
Rob,does your simulation include a channel probe and an equalizer? Also, have you examined the spectrum of the channel and of the transmitted and received signal?...

Re: How to test my FFT implementation?

Reply posted 4 years ago (03/07/2021)
Bittersweet,You have to perform Three tests. Each is very telling.test 1. Impulse responsein an array 0 to N-1,place the largest amplitude impulse in address +1...

Re: time recovery algorithm and CORDIC

Reply posted 4 years ago (03/07/2021)
Hello, all, this is a timing recovery loop using a polyphase matched filter and derivative matched filter works like a charm.fred hmodem_timing_32.m

Re: time recovery algorithm and CORDIC

Reply posted 4 years ago (03/05/2021)
hi neil, have a look at attachmentssdr_part_22.pdf Lets_assume_System_synchronized_2.pdfmodem_2012_timing_demo_1.mgood to see you (your picture at least)regards...

Re: time recovery algorithm and CORDIC

Reply posted 4 years ago (03/05/2021)
Hello AliYou have to program the front end of an IQ receiver. It has to have a pair of sqrt Nyquist matched filters running at 2-samples per symbol (same as shaping...

Re: Band-variable signals filtration - best approach

Reply posted 4 years ago (02/04/2021)
I have done similar filtering tasks using a a cascade of polyphase filter banks. First an analysis filter bank partitioning the positive frequencies into multiple...

Re: Dinner's ready!

Reply posted 4 years ago (01/31/2021)
try load handelsound(y)then try my little demofredhandel_demo

Re: Non-zero DFT components where zero is expected?

Reply posted 4 years ago (01/03/2021)
Hello bittersweet,What you are observing is that the trigonometric functions are transcendental, which means that they need very very wide width representations...
alex,Your attempt to obtain linear phase shift over the entire spectral span is doomed by Gibbs phenomena. You have to select a fractional BW, 80% and then use the...
OOPs, is your transform being implemented in fixed point or in floating point? If fixed point the amplitude should be 1/2 the largest allowable amplitude, such...
Did you do the single sinewave test on the cascade forward and inverse fft?use a single complex sinewave with one cycle per interval exp(j*2*pi*(0:N-1)/N) as the...
dear dszabo,when you perform fast convolution by two forward transforms and one inverse transform, you can save the address reversal process in the three transforms...

Re: PMCW Radar for cars

Reply posted 4 years ago (12/16/2020)
Hi mannai,Thanks for pointing out that I misinterpreted the DSP radar question. I am quite familiar with direct sequence spreading sequence radars. I worked for...

Re: PMCW Radar for cars

Reply posted 4 years ago (12/14/2020)
hello adaptivefilter,by phase modulated continuous wave radar, I'm assuming you mean linear FM sweep per transmitted pulse. This is the common wave shape for automotive...
We start with the observation that a sampled data sequence has a continuous periodic Fourier transform, periodic in 2pi, that is observed on the perimeter of the...

Re: How to synthesize band-limited noise?

Reply posted 4 years ago (11/27/2020)
Marcin,The most common approach is to take wide bandwidth white noise sequence and filter its bandwidth down to the desired bandwidth with a, IRR filter. That's...

Re: split Hilberts

Reply posted 4 years ago (10/25/2020)
Exactly,The sampled data domain has access to pure delay lines that the continuous domain does not. That simple access changes the rules and is one of the reason's...

Re: split Hilberts

Reply posted 4 years ago (10/24/2020)
% linear Phase FIR filter Hilbert Transform, two lengthsh=1:2:30;h1=1./h;h2=reshape([h1;zeros(1,15)],1,30);h4=[-fliplr(h2) 0 h2];hh=kaiser(61,8)';h3=(2/pi)*[0 h2...

Re: Anything like a “shift delay”?

Reply posted 4 years ago (10/24/2020)
You should expand the cos(wt+s(t)) to cos(s(t)) cos (wt)-sin(s(t) sin(wt)... what you have ere is QPM (quadrature phase modulation) or CPM (continuous phase modulation)....

Re: split Hilberts

Reply posted 4 years ago (10/23/2020)
in analog phase shifting hilbert transforms the two arms are +/-45 degrees to get a 90 phase difference. Much easier to do in DSP land.... are comfortable there....

Re: CFO correction in OFDM

Reply posted 4 years ago (10/01/2020)
You can correlate the CP with then of the OFDM symbol... same signal except for cannel distortion and phase offset due to residual frequency error... not a good...

Re: CFO correction in OFDM

Reply posted 4 years ago (09/30/2020)
Rob,Now you know why the 802.11 uses pilots. The initial estimate of frequency offset made with the preamble will always (repeat always) have a residual error. You...

Re: Will this FPGA be suitable for DSP purposes?

Reply posted 4 years ago (09/29/2020)
Corn1996,You first have to gather some skills in the design of phase lock loops. From your questions I gather you are trying to align the frequency and phase of...

Re: FSK demodulation with Doppler correction

Reply posted 4 years ago (09/24/2020)
loganathan,the preamble will give you enough information to estimate the dopple and the clock frequency as well as the phase. At the end of the preamble you should...

Re: Tilt correction in modulators

Reply posted 4 years ago (08/14/2020)
A good many modulators in transmitters are implemented in the DSP domain. Often the DSP based signal is interpolated to a much higher sample rate than the Nyquist...
asuchan,If your need is a reduction in BW and sample rate by 8000-to-1 you should be doing this with an 2000-to-1 CIC filter followed by 2- half band filters that...
You have 13 bits of growth per stage in your filter... how may stages do you have? if you have 4 stages you have 52 bits of growth... how many bits wide is the input...
Are you using 2's complement overflow  in your matlab simulation... I'm guessing from your question that you are concerned with your accumulators overflowing......
the ramp is the input-output relationship without quantizing.... the staircase is the instantaneous non linearity of a rounding quantizer... rela quantizers don't...
the adc always estimates the signal from below... the highest discrete level not higher than the measured signal... it performs truncation....successive approximation...
the self mixing properties of the ADC of the analog mixers down converts a DC line on both I and Q... the truncation transfer function of the ADC always injects...

Use this form to contact fharris

Before you can contact a member of the *Related Sites:

  • You must be logged in (register here)
  • You must confirm you email address