Slope drift and offset compensation in zero-if receivers

• US 6,473,471 B2
• Filed: 08/13/1996
• Issued: 10/29/2002
• Est. Priority Date: 03/09/1995
• Status: Expired due to Term

• Current Assignee: Ericsson, Inc. (Telefonaktiebolaget LM Ericsson)
• Original Assignee: Ericsson, Inc. (Telefonaktiebolaget LM Ericsson)
• Inventors: Dent, Paul W., Lindquist, Björn
• Primary Examiner(s): Vo, Don N.
• Application Number:
  US08/699,774
• Publication Number:
  US 20020071506A1
• Time in Patent Office: 2,268 Days
• Field of Search: 375/341, 375/371, 375/262, 375/344, 375/346, 375/350, 455/324, 455/63, 455/296
• US Class Current: 375/341
• CPC Class Codes: H03D 3/008 Compensating DC offsets
• CPC Class Codes: H03D 7/165 at least two frequency changers being located in different paths, e.g. in two paths with carriers in quadrature combined with amplitude demodulation H03D1/2245, combined with angle demodulation H03D3/007; N-path filters H03H19/002
• CPC Class Codes: H04L 2025/03401 PSK
• CPC Class Codes: H04L 2025/03407 Continuous phase
• CPC Class Codes: H04L 25/03273 with carrier recovery circuitry
• CPC Class Codes: H04L 27/233 using non-coherent demodulation

Abstract

A method for processing a signal modulated with information symbols to account for an additive offset and slope is disclosed. First, an initial estimate of offset and slope is made and hypotheses of all possible values of a sequence of one or more information symbols are then made. For each of said hypotheses, the associated data symbol sequence is used to make an improved estimate of offset and slope and the improved estimate of offset and slope are stored against each of the hypotheses. For each hypothesis, the improved estimate of offset and slope is used in calculating an expected signal value and a mismatch between a sample of the modulated signal and the expected value is computed. The hypotheses are then sequentially extended by one symbol, the slope and offset estimates are updated and the mismatches are accumulated to form a path metric value for each extended hypothesis, and resolving between said hypotheses based on said path metric values using a Viterbi Sequential Maximum Likelihood Sequence Estimation process to produce a most likely hypothesis of said modulated information symbols substantially unimpaired by said additive slope and offset.

Claims

1. An improved radio receiving apparatus for direct conversion of signals to a baseband for processing, comprising:

direct conversion means for converting a radio input signal to a complex baseband signal having a real waveform and an imaginary waveform; and

slope and offset compensation means for estimating offsets and systematic drifts in said real and imaginary waveforms and compensating for said drifts and offsets such that decoding of information modulated on said radio input signal is substantially unimpaired, said slope and offset compensation means comprising;

integrating capacitors which are driven to follow the input real and imaginary waveforms by a charge or discharge current pulse from charge pumps;

comparator means for comparing the real and imaginary waveforms with a voltage on said integrating capacitors to generate a positive/negative indication;

processing means for determining correction values for said real and imaginary waveforms wherein said correction values are used to select current magnitudes to be applied to said charge pumps, said current magnitudes being stored in one of register pairs and pairs of memory positions and said current magnitudes, for each pair of said one of register pairs and pairs of memory positions, are independent for positive and negative stepsign.

2. A receiver according to claim 1, further comprising means for further processing said corrected waveforms to demodulate and decode information.

3. A receiver according to claim 2, wherein said further processing performs demodulation of an analog voice-modulated signal.

4. A receiver according to claim 3, wherein said analog voice modulation is frequency modulation.

5. A receiver according to claim 2, wherein said information is digital information.

6. A receiver according to claim 5, wherein said digital information represents a digitized voice signal.

7. A receiver according to claim 5, wherein said digital information is manchester-code frequency modulated on to said radio input signal.

8. A receiver according to claim 5, wherein said digital information is modulated on to said radio input signal using Gaussian Minimum Shift Keying.

9. A receiver according to claim 5, wherein said digital information is modulated on to said radio input signal using DQPSK or Pi/4-DQPSK.

10. A method for processing a signal modulated with information symbols to account for an additive offset and slope in said signal, comprising:

making an initial estimate of offset and slope errors in said modulated information symbols;

making hypotheses of possible values of a sequence of one or more information symbols within a predefined period of time;

for each of said hypotheses a data symbol sequence associated with said hypotheses can be used to make an improved estimate of offset and slope and storing the improved estimate of offset and slope against each of said hypotheses;

for each hypothesis, using said improved estimate of offset and slope in calculating an expected signal value and computing a mismatch between a sample of said modulated signal and said expected value;

sequentially adding one symbol to said hypotheses, updating said slope and offset estimates and accumulating said mismatches to form a path metric value for each extended hypothesis; and

selecting between said hypotheses based on said path metric values using a Viterbi Sequential Maximum Likelihood Sequence Estimation process to produce a most likely hypothesis of modulated information symbols which are unimpaired by said additive slope and offset.

11. A method according to claim 10, wherein said digital information is manchester-code frequency modulated on to said radio input signal.

12. A method according to claim 10, wherein said digital information is modulated on to said radio input signal using Gaussian Minimum Shift Keying.

13. A method according to claim 10, wherein said digital information is modulated on to said radio input signal using DQPSK or Pi/4-DQPSK.

14. A method according to claim 10, wherein said information is digital information.

15. A method according to claim 14, wherein said digital information represents a digitized voice signal.