摘要:
A multiplexed FIR/IIR digital filter structure (300) which offers linear phase response and low group delay by switching on a FIR filter portion (31) or a IIR filter portion (32). To reduce the silicon area, the FIR/IIR filter (300) shares registers which is enabled because the FIR and IIR processing do not use the registers at the same time but rather consecutively. Further, the multiplexed FIR/IIR digital filter structure (300) can offer limit-cycle-free IIR operation using two's-complement truncation in combination with positive valued allpass coefficients.
摘要:
Die Erfindung bezieht sich auf ein digitales Rekursivfilter in Zustands-Raumstruktur, wobei das bewertete Eingangssignal (2) Verzögerungsgliedern zugeführt wird, deren Ausgangssignale bewertet jeweils auf ihren Eingang und jeweils auf den Eingang des anderen Verzögerungsgliedes zurückgeführt sind und bewertet zum Ausgangssignal zusammengefaßt sind (Filterblock 2. Ordnung), mit vorgegebener Signalwortlänge, in Fest- oder Gleitkomma-Arithmetik, und ist dadurch gekennzeichnet, daß nach jeder Bewertung das Signal einer Schneide- oder Rundungsoperation unterzogen wird, daß das Ein- oder Ausgangssignal der Verzögerungsglieder jeweils einer Sättigungskennlinie unterzogen werden und daß das Ein- oder Ausgangssignal der Verzögerungsglieder oder der Begrenzer oder die auf den Eingang der Verzögerungsglieder rückgeführten zusammengefaßten Signale jeweils einer Signalkorrektur unterworten werden, indem bei negativem Signal diesem seine Quantisierungsstufe q hinzuaddiert wird.
摘要:
A filter system (201, 301, 401, 501, 601) with infinite impulse response shall be created that has a particularly low sensitivity in the numerical representation of filter coefficients. This is achieved through the transfer function of the filter system comprising at least one pair of first order polynomial fractions.
摘要:
Recursive digital filter (3) circuitry which avoids persistent unstable conditions therein provides a serial clock signal, a synchronization signal, and a serial data input to corresponding inputs of a three- wire serial interface circuit (2) to produce a serial clock output signal, a synchronization output signal, and a parallel data output signal which are applied to corresponding inputs of a recursive digital filter. The serial clock signal and the synchronization signal are input to an auto-reset circuit (6) which detects a fault associated with the synchronization signal or the serial clock signal and produces a reset signal in res onse to detection of the fault for resettin the recursive digital filter.
摘要:
Limit-cycle oscillations are caused by the compounding of quantization errors that occurs when previous digital filter outputs are used as inputs to the digital filter for the current operation. Where a signal in a digital waveform has become a constant common value applied to the input of the digital filter (indicative that the digital waveform has suspended conveyance of data), limit-cycle oscillations often appear as "random" outputs, with values different from the common value, that occur long after the signal in the digital waveform has become the constant common value. Limit-cycle oscillations are manifested as noise in the filtered digital waveform. Such noise hampers the ability of the system to extract the signal from the filtered digital waveform. The present invention identifies the occurrence of a limit-cycle oscillation as an output different from the common value. The identified limit-cycle oscillation is set equal to the common value.
摘要:
An infinite impulse response (IIR) digital filter and method of performing the same is disclosed. The digital filter may be realized by way of a programmable logic device, such as a digital signal processor (75), or alternatively by way of dedicated logic including adders (44, 48, 50, 54, 58, 62, 66, 70, 72) and shifters (46, 52, 56, .60, 64). In either case, addition operations (34) are interleaved among first and second output sample values (y n-1 , y n-2 ), so that the resulting addition (30; 72; 215; 320) may be carried out with adder circuitry of the same precision as the signal input (x n ) and signal output (y n ). Carry control circuitry (76, 78, 80, 82, 84, 88; 217; 317) is provided to efficiently incorporate magnitude truncation quantization.
摘要:
A method for compensating the truncation error occurring in a sampled signal during processing in a digital filter. The mean error et is calculated for the filter, and before filtering the signal +et is added if the signal is positive and -et is added if the signal is negative. A device for carrying out the method is also described.