Signal characteristic state detector using interval-count processing method

• US 4,330,862 A
• Filed: 07/09/1980
• Issued: 05/18/1982
• Est. Priority Date: 07/09/1980
• Status: Expired due to Term

• Current Assignee: Bell Telephone Laboratories, Inc. (Nokia Corporation)
• Original Assignee: Bell Telephone Laboratories, Inc. (Nokia Corporation)
• Inventors: Smolik, Kenneth F.
• Primary Examiner(s): Britton, Howard
• Application Number:
  US06/167,383
• Time in Patent Office: 678 Days
• Field of Search: 375/75, 375/84, 375/114, 375/118, 375/119, 375/120
• US Class Current: 375/316
• CPC Class Codes: H04L 1/20 using signal quality detector
• CPC Class Codes: H04L 25/4904 using self-synchronising codes, e.g. split-phase codes

Abstract

Received baseband data signals, in the form of signals having at least one level transition per symbol indicative of symbol information state, are sampled (17) at a rate much higher than the symbol rate. The samples are used to control the direction of counting (20) of local clock (12) signals recurring at that higher rate during symbol subintervals of predetermined duration. Counts produced during adjacent subintervals are processed (FIGS. 4 and 5) to produce several different results such as recognition of the start of an asynchronously occurring data message, acquisition and tracking of symbol phase in the message, and detection of symbol information states.

Claims

1. In a system for receiving binary coded data signals having at least one predetermined characteristic, as well as having at least one signal level transition per symbol in a direction that is indicative of the binary information state of the symbol, the method for detecting the state of said predetermined characteristic of such data signals CHARACTERIZED BY the steps of

generating (10) binary coded words representing, respectively, for recurring symbol subintervals of predetermined length, the approximate proportion of each subinterval that the information signal state is in one of the binary signal states, and

processing (11) said words to detect the state of said characteristic in said signals.

2. The data signal characteristic detecting method in accordance with claim 1 wherein the system is responsive to a local clock signal (12), and the method is further CHARACTERIZED IN THAT the word generating step includes the steps of

sampling (17) said data signal at the rate of the local clock signal, which clock signal is at a higher rate than the data bit rate, and

during each subinterval, reversibly counting (20) pulses in said local clock signal in a counting direction which is a function of the data signal information state of each data sample.

3. The data characteristic detecting method in accordance with claim 2 in which the method is further CHARACTERIZED BY

deriving from said local clock signal a bit rate clock signal, and

the inclusion in said processing step of the steps of

calculating the sum and difference of time-adjacent subinterval count words, and

adjusting the phase of said bit rate clock to a predetermined phase relationship with the received data signals.

4. The data signal characteristic detecting method in accordance with claim 3 and further CHARACTERIZED BY the inclusion in the processing step of the steps of

establishing a symbol information state discriminating threshold value,

comparing at least one subinterval count word per symbol to said threshold value, and

outputting, in response to the results of the foregoing comparing step, a regenerated version of the data signal in synchronism with said local clock.

5. The data signal characteristic detecting method in accordance with claim 1 in which said characteristic is the leading or lagging phase state of a bit rate clock signal with respect to symbols of said data signals and CHARACTERIZED IN THAT

said subinterval predetermined length is approximately one-half of a symbol interval in said data signals, and

said processing step includes the steps of

selecting a threshold value for said phase state, and

testing a value proportional to the magnitude of one of said subinterval count words for one of said symbols with respect to said threshold value to indicate leading or lagging phase as a function of whether the value is larger or smaller than said threshold.

6. The data signal characteristic detecting method in accordance with claim 1 or 2 further CHARACTERIZED IN THAT the processing step includes the steps of

selecting a threshold value and a count word combining rule for said characteristic,

combining selected ones of said count words according to said rule, and

comparing the result of said combining step to said threshold value to indicate the state of said characteristic in said signals.

7. The data signal characteristic detecting method in accordance with claim 6 in which said characteristic is the binary information state of successive symbols in said data signals, and the method is further CHARACTERIZED IN THAT said combining step includes the steps of

calculating the difference between time-adjacent subinterval count word values, and

said comparing step comprises the step of testing said difference with respect to an information threshold value such that the test result has a first binary state if the difference is greater than said threshold, and a second binary state if the difference is less than said threshold value.

8. The data signal characteristic detecting method in accordance with claim 6 in which said characteristic is the leading or lagging phase state of a bit rate clock signal with respect to symbols of said data signals, and the method is further CHARACTERIZED IN THAT

said combining step includes the steps of, for each pair of time-adjacent subinterval count words,

calculating the sum of and difference between the words of said pair, and

calculating a further difference value of the first mentioned count word difference minus said count word sum, and

said comparing step comprises the step of testing the magnitude of said further difference with respect to a predetermined phase state threshold value such that if said further difference is larger than said such threshold value, an advance-clock-phase signal is produced, and if said further difference is smaller than said such threshold, a retard-clock-phase signal is produced.

9. The data signal characteristic detecting method in accordance with claim 8 CHARACTERIZED IN THAT it includes the further steps of

repeating said combining and comparing steps for successive symbols of said data signal,

accumulating the number of said advance signals reduced by the number of said retard signals,

testing the result of said accumulating step with respect to a predetermined range of values, and

when said result is outside of said range, changing the phase of said bit rate clock by one predetermined delay unit in a direction indicated by the sign of said result.

10. The data signal characteristic detecting method in accordance with claim 6 in which said characteristic is the presence or absence state of a predetermined frequency component in said data signals, and said method is further CHARACTERIZED IN THAT said comparing step further includes the step of

testing, as to successive pairs (A_i, A_i+1) of alternate ones of said words, the magnitude of the difference between the words of a pair with respect to the magnitude of a predetermined frequency threshold value chosen so that if said difference is at least equal to said value, said frequency component is present.

11. The data signal characteristic detecting method in accordance with claim 10 and further CHARACTERIZED IN THAT said testing step includes the steps of

calculating the difference between subinterval count words of a pair occurring at the beginning of time-adjacent symbols in said data signals,

calculating the difference between subinterval count words of a pair occurring at the end of said time-adjacent symbols,

testing at least one of said calculated differences with respect to a said frequency threshold value, and

if the tested difference is at least equal to said frequency threshold, producing a signal indicating the presence of said frequency component.

12. The data signal characteristic detecting method in accordance with claim 11 CHARACTERIZED IN THAT said comparing step further includes the steps of

accumulating the number of occurrences of said component indicating a signal over a plurality of successive sets of time-adjacent symbols, and

testing the result of said accumulating step for attainment of a value predetermined to signify a sequential symbol interval of predetermined minimum duration and including said frequency component.

13. The data signal characteristic detecting method in accordance with claim 11 in which said method is further CHARACTERIZED IN THAT during the next symbol interval following said time-adjacent symbols, said method includes the steps of

calculating the sum and difference of time-adjacent subinterval count words thereof,

translating the values of said sum and difference into a corresponding bit rate clock phase error value, and

adjusting the phase of said bit rate clock substantially to eliminate said phase error value.

14. The data signal characteristic detecting method in accordance with claim 13 further CHARACTERIZED IN THAT said adjusting step includes the step of

offsetting the phase adjustment value for said symbol rate clock by an amount which is sufficient in a predetermined phase error range to cause every sum and difference value combination to represent a unique phase error value in said range.

15. The data signal characteristic detecting method in accordance with claim 14 in which an additional characteristic is the binary information state of said data signals, and the method is further CHARACTERIZED IN THAT said processing step includes the steps of

selecting one of said interval count words occurring in a symbol interval,

testing said one word with respect to a predetermined information threshold value, and

outputting a signal in a first binary state if said word is greater than said information threshold, and outputting a signal in a second binary state if said word is less than said information threshold.

16. In a system for receiving binary coded data signals having at least one predetermined characteristic as well as having at least one signal level transition per symbol in a direction that is indicative of the binary information state of the symbol, a circuit for detecting said characteristic and being CHARACTERIZED IN THAT it includes

means for generating (10) binary coded words representing, respectively, for recurring symbol subintervals of predetermined length, the approximate proportion of each interval that the information signal state is in one of the binary signal states, and

means for processing (11) successive ones of said words for detecting said data signal characteristic.

17. The characteristic detecting circuit in accordance with claim 16 in which such circuit is responsive to a local clock (12) signal and is CHARACTERIZED IN THAT said word generating means comprises

means (17) for sampling said data signal at the rate of said local clock signal, which rate is higher than the symbol rate of said data signal, and

means (20) for reversibly counting pulses in said lock clock signal during respective ones of said subintervals in a counting direction which is a function of the data signal information state of each sample.

18. The characteristic detecting circuit in accordance with claim 17 CHARACTERIZED IN THAT said processing means comprises

means (11) for combining selectable sets of said subinterval count words in accordance with a predetermined logical rule for representing said predetermined characteristic of said data signals, and

means for comparing combined count word results with respect to a predetermined threshold value for indicating the state of said characteristic.

19. The characteristic detecting circuit in accordance with claim 17 further CHARACTERIZED IN THAT there are provided

means (13), responsive to an output of said processing means, for adjusting the phase of operation of said counting means.