Enhanced decoder for a radio telephone

• US 5,557,639 A
• Filed: 10/05/1994
• Issued: 09/17/1996
• Est. Priority Date: 10/11/1993
• Status: Expired due to Term

• Current Assignee: Nokia Technologies Oy (Nokia Corporation)
• Original Assignee: Nokia Mobile Phones UK Limited (Nokia Corporation)
• Inventors: Heikkila, Ilkka, Ranta, Jukka, Jokinen, Harri
• Primary Examiner(s): Tse, Young T.
• Assistant Examiner(s): WEBSTER, BRYAN E
• Application Number:
  US08/322,445
• Time in Patent Office: 713 Days
• Field of Search: 375/224, 375/225, 375/227, 375/359, 375/285, 375/346, 375/348, 455/226.1, 455/226.2, 455/226.3, 455/63, 455/67.3, 455/226.1, 455/226.2, 455/226.3, 371/43, 371/44, 371/45
• US Class Current: 375/224
• CPC Class Codes: H04L 1/20 using signal quality detector
• CPC Class Codes: H04L 1/201 Frame classification, e.g. bad, good or erased frame indication per se H04L1/0082
• CPC Class Codes: H04L 1/208 involving signal re-encoding

Abstract

The invention provides an improved method and apparatus for identifying a bad GSM speech frame. Both the estimated signal-to-noise ratio (ESNR) of the received signal and the pseudo bit error rate (PBER) of a recorded speech signal are used to determine whether or not a speech frame is bad.

Claims

1. A method for identifying a bad GSM speech frame in digital mobile phone system, characterized in that the method comprises the steps of:

receiving a signal;

estimating the quality of the received signal using a function formed from an estimated signal-to-noise ratio ESNR, where the ESNR is a vector comprising the estimated signal-to-noise ratio of each received burst or a scalar describing their approximate signal-to-noise ratio, and a pseudo bit error rate PBER of a received and recorded speech frame, the ESNR and PBER being functionally related to a chosen threshold; and

removing a speech frame if it fails to meet the threshold.

2. A method according to claim 1, wherein the step of removing a speech frame if f(ESNR, PBER)>

T, where f is a function of variables ESNR and PBER, and where T is a chosen threshold value.

3. A method according to claim 1, wherein the function is so formed that a number of faulty decisions to remove speech frames can be minimized.

4. A method according to claim 1, wherein the function is implemented in the form of a table so that the ESNR and PBMR are indexed variables of the table.

5. A method for identifying a bad GSM speech frame in digital mobile phone system, characterized in that the method comprises the steps of:

receiving a signal;

estimating the quality of the received signal using an estimated signal-to-noise ratio ESNR, where the ESNR is a vector comprising the estimated signal-to-noise ratio of each receiveg burst or a scalar describing their approximate signal-to-noise ratio, and a pseudo bit error rate PBER of a received and recorded speech frame, the ESNR and PBER being functionally related to a chosen threshold; and

removing a speech frame if it fails to meet the threshold, and wherein

space="preserve" listing-type="equation">f(a,b)=(C1*a)+(C2*b)

is chosen for a function so that the speech frame is removed if

space="preserve" listing-type="equation">(C1*ESNR)+(C2*PBER)>

T

wherein C1 and C2 are constants and the function is a linear combination of variables ESNR and PBER.

6. A method for identifying a bad GSM speech frame in digital mobile phone system, characterized in that the method is comprised of the steps of:

receiving a signal;

estimating the quality of the received signal using an estimated signal-to-noise ratio ESNR, where the ESNR is a vector comprising the estimated signal-to-noise ratio of each received burst or a scalar describing their approximate signal-to-noise ratio, and a pseudo bit error rate PBER of a received and recorded speech frame, the ESNR and PBER being functionally related by a function f to a chosen threshold; and

removing a speech frame if it fails to meet the threshold, and wherein the function f and a threshold limit T are made adaptive.

7. A method for identifying a bad GSM speech frame in digital mobile phone system, characterized in that the method is comprised of the steps of:

receiving a signal;

estimating the quality of the received signal using an estimated signal-to-noise ratio ESNR, where the ESNR is a vector comprising the estimated signal-to-noise ratio of each received burst or a scalar describing their approximate signal-to-noise ratio, and a pseudo bit error rate PBER of a received and recorded speech frame, the ESNR and PBER being functionally related to a chosen threshold; and

removing a speech frame if it fails to meet the threshold, wherein the speech frame is removed if

space="preserve" listing-type="equation">f(ESNR, PBER)>

T

where f is a function of variables ESNR and PBER and T is a chosen threshold value, and wherein the threshold value is calculated

space="preserve" listing-type="equation">BFI.sub.-- LIM=(SNR.sub.-- VAR)-C1)*C2+C3

wherein SNR_-- VAR is a variance of the values of the ESNR-vectors; and

where the obtained result BFI_-- LIM is limited in the following way

space="preserve" listing-type="equation">C4≦

BFI.sub.-- LIM≦

C5

so that the speech frame is removed if

space="preserve" listing-type="equation">PBER>

BFI.sub.-- LIM,

where C1, C2, C3, C4 and C5 are constants defined by simulation.

8. A circuit for a radio telephone for providing speech frames from a modulated RF signal transmitted to the radio telephone as a plurality of digitally encoded bursts, comprising:

an antenna for receiving said modulated RF signal;

a demodulator for demodulating said received RF signal so as to extract digital information representing speech frames from said modulated RF signal;

signal quality determination means having an input coupled to an output of said demodulator for providing, for each received burst, a first burst quality measure and a second burst quality measure, said first burst quality measure being indicative of a signal-to-noise ratio and said second burst quality measure being indicative of a bit error rate; and

means, having inputs coupled to outputs of said signal quality determining means, and responsive to a function formed in accordance with a combination of said first and second burst quality measures, for rejecting the digital information of a burst having a function that fails to meet a burst quality threshold value.

9. A circuit for a radio telephone for providing speech frames from a modulated RF signal transmitted to the radio telephone as a plurality of digitally encoded bursts, comprising:

an antenna for receiving said modulated RF signal;

a demodulator for demodulating said received RF signal so as to extract digital information representing speech frames from said modulated RF signal;

signal quality determination means having an input coupled to an output of said demodulator for providing, for each received burst, a first burst quality measure and a second burst quality measure, said first burst quality measure being indicative of a signal-to-noise ratio and said second burst quality measure being indicative of a bit error rate; and

means, having inputs coupled to outputs of said signal quality determining means, and responsive to a comparison of said second burst quality measure with a burst quality threshold value, for rejecting the digital information of a burst having a second burst quality measure that fails to meet the burst quality threshold value;

wherein said burst quality threshold value is derived at least in part from said first burst quality measure.