CA2089154C - Automatic frequency control by an adaptive filter - Google Patents
Automatic frequency control by an adaptive filterInfo
- Publication number
- CA2089154C CA2089154C CA002089154A CA2089154A CA2089154C CA 2089154 C CA2089154 C CA 2089154C CA 002089154 A CA002089154 A CA 002089154A CA 2089154 A CA2089154 A CA 2089154A CA 2089154 C CA2089154 C CA 2089154C
- Authority
- CA
- Canada
- Prior art keywords
- signal
- mean square
- error
- response
- generating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 claims abstract description 17
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 9
- 238000001914 filtration Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims 4
- 238000009825 accumulation Methods 0.000 claims 1
- 230000008569 process Effects 0.000 abstract description 10
- XUKUURHRXDUEBC-KAYWLYCHSA-N Atorvastatin Chemical compound C=1C=CC=CC=1C1=C(C=2C=CC(F)=CC=2)N(CC[C@@H](O)C[C@@H](O)CC(O)=O)C(C(C)C)=C1C(=O)NC1=CC=CC=C1 XUKUURHRXDUEBC-KAYWLYCHSA-N 0.000 description 3
- 241000282320 Panthera leo Species 0.000 description 3
- 230000001413 cellular effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000000523 sample Substances 0.000 description 2
- UDIODXADSSQKTM-GUHNCMMLSA-N (5ar,8ar,9r)-5-[[(2r,4ar,6r,7r,8r,8as)-7,8-dihydroxy-2-methyl-4,4a,6,7,8,8a-hexahydropyrano[3,2-d][1,3]dioxin-6-yl]oxy]-9-(4-hydroxy-3,5-dimethoxyphenyl)-5a,6,8a,9-tetrahydro-5h-[2]benzofuro[6,5-f][1,3]benzodioxol-8-one;(7s,9s)-7-[(2r,4s,5s,6s)-4-amino-5- Chemical compound ClCCN(CCCl)P1(=O)NCCCO1.O([C@H]1C[C@@](O)(CC=2C(O)=C3C(=O)C=4C=CC=C(C=4C(=O)C3=C(O)C=21)OC)C(=O)CO)[C@H]1C[C@H](N)[C@H](O)[C@H](C)O1.COC1=C(O)C(OC)=CC([C@@H]2C3=CC=4OCOC=4C=C3C(O[C@H]3[C@@H]([C@@H](O)[C@@H]4O[C@H](C)OC[C@H]4O3)O)[C@@H]3[C@@H]2C(OC3)=O)=C1 UDIODXADSSQKTM-GUHNCMMLSA-N 0.000 description 1
- 101150111329 ACE-1 gene Proteins 0.000 description 1
- HRPVXLWXLXDGHG-UHFFFAOYSA-N Acrylamide Chemical compound NC(=O)C=C HRPVXLWXLXDGHG-UHFFFAOYSA-N 0.000 description 1
- 241000380873 Algon Species 0.000 description 1
- 101100162210 Aspergillus parasiticus (strain ATCC 56775 / NRRL 5862 / SRRC 143 / SU-1) aflM gene Proteins 0.000 description 1
- 101100102500 Caenorhabditis elegans ver-1 gene Proteins 0.000 description 1
- 240000000233 Melia azedarach Species 0.000 description 1
- 150000001768 cations Chemical class 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 150000002500 ions Chemical class 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03L—AUTOMATIC CONTROL, STARTING, SYNCHRONISATION OR STABILISATION OF GENERATORS OF ELECTRONIC OSCILLATIONS OR PULSES
- H03L7/00—Automatic control of frequency or phase; Synchronisation
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
- H03H21/00—Adaptive networks
- H03H21/0012—Digital adaptive filters
- H03H21/0043—Adaptive algorithms
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03J—TUNING RESONANT CIRCUITS; SELECTING RESONANT CIRCUITS
- H03J7/00—Automatic frequency control; Automatic scanning over a band of frequencies
- H03J7/02—Automatic frequency control
- H03J7/04—Automatic frequency control where the frequency control is accomplished by varying the electrical characteristics of a non-mechanically adjustable element or where the nature of the frequency controlling element is not significant
Landscapes
- Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
- Digital Transmission Methods That Use Modulated Carrier Waves (AREA)
- Channel Selection Circuits, Automatic Tuning Circuits (AREA)
- Circuits Of Receivers In General (AREA)
- Radio Relay Systems (AREA)
- Feedback Control In General (AREA)
Abstract
The process of the present invention generates an optimum automatic frequence co ntrol signal in an apparatus that has a plurality of adaptive algorithms, each adaptive algorithm having a reference sig nal with an associated frequency dither. The process starts by comparing the performance of each of the plurality of adaptive algorithms. This difference is input to a compartor (6) where it is compared to zero. This delta signal then modifies the numericall y controlled oscillator frequency (8). After several iterations, the frequency offset is reduced to substantially zero.
Description
W O 93/00747 2 ~ 8 9 1 5 ~ PC~r/US92/03763 AUTOMATIC FREQUENCY CONTROL BY AN ADAPTIVE
FrLTER
F~do~~ L..
The present iu~,eulion relates ..c~ally to the field of co..~ -..ic~tio~c and particularly to olltQ nq*c frequency con-trol.
1 0 B~do~t~1~venlion U.S. ~ tol cellular (USDC) comml~nico*on8 uses digi-tized voice and data ~i~ol~ for comm~nirot;~n bc~- ee~ a mo-bile telephone and a base station. The mobiles and bases em-l 5 ploy time division multiple acces~ (TDMA) mo~l~lotiQ- for the tronnmi~sj~orl of these sign-l~. A typical format for the TDMA
data bursts is i]lu~L_ted in E IG. 5. This data burst, as well as USDC in general, is ~ in more detail in the USDC
sperifirati~n EIA/TIA IS-54 available from hlc~,lrv~c 20 Tn~ ~ies ~oci~tjon, TP~neeling Del,a.h.lent, 2001 Eye Street, N.W., W~hington, D.C. 20006.
When the mobile mo~es, it may enco~tc~ de- ~ de~l c~.... ir~t;Qn channels due to noise and mllltir~th distor-tion; both noise and distortion va~ ~u~g with time. The multi-2 5 path distortion is due to a signal being ~ecc;ved by the mobile at ~ . t times when it bouuces off btlil~ling~ and terrsin.
Multipath channels csn cause inter-symbol interference (ISI) that can be removed with an adaptive rh~.~..ol eqll~li7er, a s~fic type of adaylive filter. An adayLve channel cFl~ tQr 3 0 is ~n~tl~er type of ad~y1.ive filter.
A typicsl adsptive filter is illustrated in FIG. 1. The in-put signal (106) is p.ocf,sse~l by the adaptive filter (101), pro-rlllring the ad~,1ive filter output signal (102). The o.-ly~ of the filter is then subl~ rte~l (105) from a ~e~e~e..ce signal (103), 3 5 typically the llnfiltered input signal (106), to produce an error signal (104). This error signal (104) is used by an adaptive al-'- ~0~9 ~ ~ 4 gorit~tm with an ~~pAStte coeffi~çnt, ~, in the adaptive filter to l-pA~e the filter coeffi~ipnts~ The nr~Stte coeffiriPnt is also re-ferred to as a ~rL ;~ oçf~i~içnt or msmQry coeffi~ ient~ The ~e~o. ~ of the adaptive algorithm increases as the value of S increa~es.
The adaptive algorithm may be a ~stlmstn~ Rec.,.~i~,e Least Square, or Least Mean Square (LMS) algon~tm The typical goal of the ad~pli~e algonthm is to ~--;~-;---i7e the mean square value of the error signal (104), fised ~p~lote co~ffi~Pn~
10 Thi~ value is typically ~le~i~n~te~l mean square error (MSE).
A detriment~l c~ e,~ristic of adaptive rh~nn~l equal-izer~ i~ that they can esperience degraded ~-~r.---S~nce in the p~e~~ -~ce of a frequency offset of more than a~.o~;...~tely 10 Hz (in a a~al4h- with a 24 kHz symbol rate). While the specifi-l 5 cation for a tr~n~mias;Qn &~lem l~q~ es a certain frequencyv~ t;~ limit, the adaptive ch~nnel eq~li7er may .eq~e a ~t~icte- limit. An esample is t_e USDC ~lem. USDC re-quires the ~ecc;v~- ope.al~ frequency to be locked within 200 Hz of the ~ ;tteY. The adaptive ~h~nn~l eq~ i7er will not 2 0 ~e.fo-m properly in this e..vi.ol.me~t~ A coarse automD~tic frequency control (AFC) is typically used to ~ o~ove most of the offset. Any rem~ining offset, however, can detrim-~nt~lly af-fect a dete~on al~w;~ -- and i..c,osse the detected bit erTor rate. There i8 a res~ ;ng need for an AFC that can reduce the 2 5 frequency offset to an acceptably small level and track any vanation of the offset as the environment changes when the mobile moves.
8umma~y of ~he Iu~L;o~
The l,-ocess of the present invention generates an opti-mum ~tom~;c frequency control signal in an al,l,a~atus that has a plurality of adaptive filters. The process starts by generating a difference signal by colllpa~ g the perforrn~n~es of a subset number of ~' ., ~0~9 ~ 54 adaptive filters with each other. Then the automatic frequency control signal is adjusted in response to the dirre~ ce signal.
B~iefDesc~lion~eDraw~
FIG. 1 shows a block diagram of a typical ~ ve fil-ter.
FIG. 2 shows a block diagram of the y.OU~58 of the pre-10 sent iuve.llion.
FIG. 3 shows a graph of mean square error versusresi~l~Pl frequency offset in accordance with the y.vcess of the y~ee~t iuve~lion.
FIG. 4 shows a graph of re~ ol frequency offset ver-1 5 sus time in accordance with the yrocea3 of the present inven-tion.
FIG. 5 shows the fo~ ~t of a TDMA data burst used in the U.S. digital cellular co~ o-t;Qn ~ e~.
FIG. 6 shows an alternate emho~liment of the y.~cea~ of 2 0 the yreeel,t invçnt;Qn Det~iled Descri~on of t~e Ple&ned Embodin~
The process of the y~e~e~lt invention provides fine au-2 5 tomot;c frequency control in a device using adayl ive filters.The di~ ..ce in pe~ ance of the adaptive filters is used to ~Gdil,~ the AFC ~iEnDl, thereby re~ ng the frequency offset.
A linear coherent digital radio leceivêr typically de-mo~ t~s the ;-~co~ g signal by mi~ the signal to base-3 0 band using a local osrillAtor. The frequency of the local oscil-lator must be kept re~con~l~ly close to the frequency of the tr-o-n~mitter. Af~cer the signal has been mised down to base-band, further analog or digital sign.,l y~OCe5~ is performed to fec~J~er an estim~te of the tr~n~mitte~l data. In the follow-3 5 ing description of the ploce88 of the present invention~ it is as-~llmPd that the b~ebAn~ signal has been co.lve~ l,ed by an ana-~ ~.
20~9 ~ 54 log to digital collvel ler to a fonn suitable for further digital signal proce~ g The preferred emho~limpnt of the process of the present invent;on, as ilh,&~al,ed in FIG. 2, is cG,u~r;sed of t_ree adap-tive filters (1- 3) configured as adaptive ch~nnPl es1;m~tQrs (ACE). The three ACE's (1- 3) _ave an l~p~l~ta- coeffi~ent, ~L, that varies vrith the c-.vh~ent of the dence. The process for determinin~ ,u is described in ~n~ n patent application no. 2,086,548 filed on May 8, 1992 and titled "A Method for Optimi7~tion of Adaptive Pilter Update Coefficient" to Kevin Baum and ~sign~qd to Motorola, Inc.
The update coefficient will remain constant during a single TDMA data burst.
The three ACE's (1 - 3) are i~en1;~1 escept for having di~e~.-t frequency offset dither generators, a dither generator 1 5 being the source of the ~efer~nce ~ l. ACE 2 uses the base-band ~-,c~;~,ed signal that has been mised (9) with the numeri-cally controlled o~cill~tQr (NC0) (8) signal as the ~eferehce s~ l ACE's 1 and 3 mis (11 and 12) this lefe.cl ce signsl with frequency offsets before using it as a ~cfe.e.,ce ~
2 0 The frequency off~et~, ei~ and ei~', and their rPl~t;Q~-ship to the des*ed resi~ln~l frequency of~et, a~, are illustrated g~p~iCplly in FIG. 3. These offsets are located on either side of the frequency offset e~t;m~te, -~. ACE's 1 and 3 act as resid~ frequency offset ~ bes"; relative to ACE 2, ACE 1 2 5 acts as a ~high~ frequency probe and ACE 3 acts as a ~low" fre-quenQ probe. ACE's 1 and 3 esti_ate two points on the MSE
curve. ACE 2 ye.fo-., s the actual desired adaptive filtering fi~nct;on .
The k tenn of the frequency offset denotes the time in-desofthe~mple. TheC)dtermisapplic~t;ondependent CDd should be chQsçn as small as poss-b1e while still allowing a difference in the mean square errors (MSE) to be detsct~
could also vary with time by setting it to a larger value in~ 11y to speed acquisition and then re~lt)ce~l to get the most accurate 3 5 frequency offset c~t;m~te. In the ~-efe~-c~ ~mbo~liment, ~d i8 set to a value of 5 ~ (2t~) r~ n~/secQntl ~.. . .
.~; .
WO 93/00747 ~ 2 0 8 9 1 5 ~ PCl'/US92/03763 ~_ -- 5 --In operation, the process of the ~.ee_~-t invention ini-tially ~e~o~es the c~.e.~t e~ te 1 frequency offset, -~, from the ~--ehS~n~l le~;~,ed signal by mi~ing (9) the signal with the NCO (8) output, ei~. Trli~;sllly, the NCO (8) fre-5 quency, ~, is set to zero if there is no prior knowledge of theinitial frequency offset. This is in~lir~te~l by the initial accu-mnlsttor (7) value being zero.
This signal is then operated on by a ~l~t~ct;n~ algorithm (10) that is driven by the ACE 2 output. The resulting symbol 1 0 ~'e~iQviQ-~ ~Qign~l, a, is input to the three ACE's (1 - 3).
The ACE's (1 - 3) ~;~..e.ste error signs~l-s that are the dif-fe~ the filtered output and the S~Q~c~Slte~l refer-ence sigllals that are ~ 5 etl above. Two of the error sig-nals, errorl ant error3, are input to MSE çstim~tors (4 and 5) that operate s~8 follows:
Ic I n ET1 = j = ~' Ic I n ~ Ic~ r3l2 ET3 = j = ~' where Ic is the same as in the frequency offset and n is the m-m~er of samples of the error signsll As an P~Slmple~ if k = 1 and n = 10 for the first e~ 1;Qr~ cycle, k will stS~rt at 12 for the nest cycle.
2 5 The di~re.once bel~een the e~ ,qte<l MSE~8, Et = ET1 - ET3 . provides an in~lic~ n of which direc~on to move along the frequency offset a~is, illusL~ated in PIG. 3, to get closer to the ...;~;...l.... MSE point (re~i-ln~l offset = O). For e~ e, if the residual frequency offset is greater than 0, ET1 3 0 will be larger than ET3 thus m~ki~ Ed < O. The negative value of Ed in~licPte~ that ~Dv is too large and should be decre-mented.
wo93/00747 ~ 9~ L Pcr/us92/03763 ID the l,.ofer~ed emho~imPnt, Ed i8 input to a cc ~ a-tor (6) where it is compared to 0. In this case, the co~p~ator has an outl,u~ filnet;on, f~Ed), as follows:
S f~Ed) = ~ when Ed > ~-fl~Ed) = ~A when Ed ~ ~.
f~Et)=owhenEd=o~
where A is app~ t;on ~'eFendent and dete,~es the resolu-1 0 tion of the AFC and also the ada~t~l ion speed of the AFC.
can be c~osen as a very small value for a s~ate~ with a coarse AFC. In an alternate eml~QAim"nt, ~ could vary with time bg eet';n& it to a larger value ini*Ally to speed r ~ql~isition and then re~lllc~ to get the most accu~ate frequency offset esti-1 5 mate. In the ~refe~l en~ nent~ ~ is set t~D a value of 21 rp~liAn~/8ec D.~
In an alternate emk)~ r~t, Et is input to a filter in-stead of a aDml,~ato,. The filter prDvides a time ~/~~ step size (CQ-~ d to the f~ed step size of ~) that is responsive to 2 0 the size of the errDr din'e,e,lce si~n~l For e~ample, when the err~Dr di~rè~CnCe 8ignal becom~ large, the gtep gize AlltQ.~AI _ caUy increases resulting in faster co.l~e,~..ca of the algo-rithm. Using the filter, however, ~c.,eases the cQ~nple~ity of the invention and _ay cause st~bility problem~ if a higher 2 5 order filtrer is used. A first order digital ;~ të impulse re-spon~e (IrR) filter i8 lJ~efe~-l 'd due to s~h;lity and simplicity c~nsiterations. The o~ t of the filter is used to ~ te the frequency offset e~Rm~te The output of the co ~-~ stor (6) (or filter) i8 input to an 3 0 ~cc-~ tQr (7) that adds the new input value to the previ-ously stored value. The ~cc--~ te~ value is then used to control the frequency, ~Dv, of the NC0 (8). Since the M~'s ET1 and ET3 are e~;m~te~ over blocks of n s~mple~, Ed and the outputs of the CQ- ~-l'A ' ator (7) and acc ~~ tQr (8) are cal-3 5 ~ t~l every n iterations. The NC0 (8) frequency, therefore,is ~lp~l~te~3 once evely n iter~Ro~ ~.
wo 93/00747 2 0 8 9 1 5 ~ Pcr/uss2/03763 _~ 7 As illustrated gr~p~ir~lly in FIG. 3, after several NC0 (8) ~ te cycles, ~b will be &~lJlo~;m~tely equal to ~- and the rç~;At~l fieque~ offset will be alJ".o~ Ptely zero. If the frequency offset changes, the ~ Ce~3 of t_e ~ ~e,lt invention S Aetect~ and tracks the change.
The operation of the ~locea3 of the ~ ce,lt invention can be seen in graphically FIG. 4. The process is using an LMS
adaptive rh~nnPl estimator with an initial frequency offset of 50 Hz. The bit error rate of the ~letector in t_is e-~ -le is 19~o.
1 0 Note that the rçsi~ frequency offset quickly ~l~rline~ to nearly 0 Hz. The slope of the initial change from 50 Hz can be changed by modil~u~ A. A larger value for ~ will cause a faster acq~ tion and, thelefolo, a stee~er slope.
In the ~.~f~.,~l embQAiment, the ~ cess of the l"oee~t iu~,e.lLion is impl~Pm ~nte A as an alg~.l;t,l~. Alternate çm~oA-iments of the invention can be implementet in hart~ ~d or combinations of ha~l- ~e ant eoIl . ~e; each block of the pro-cess being either an algonthm or a hardware circuit eq~uva-lent of that block.
2 0 Another alternate emboA;~ -.t can use only two adap tive filters by not using the second adaptive filter. In this em-bodiment, the o~ t of one of the filters replaoes the seCQn~l filter's o~ t. The re~llting AFC value will be biased by ~12.
Still another alternate emboAinlent, illus~llsted in PIG.
2 5 6, can use ataptive eql~-li7ers in place of the rh-nnP~l estima-tors. In this PmhoAin~ent, the ~efO~euce signal and the symbol tecision ~ign~l, a, are input to the equalizer. The ataptive equalizers (601- 603) are ol~lati~,e to loh~o~e the ISI from the ,e~ titheret ~ece;v~,t ~ignol~. The adaylive eq~oli7er 3 0 may have some inherent telay until an output le~i~ to t_e current input is avoil~Ahle. The symbol decisions, a, are delayed (604 - 606) until the eqn-li77Or o~ t coms~ L to that Ae~iPiQ- is avAilohle T_e diffO~uce between the symbol decisions and the coll~pQnAing eq~-ou~-or o~ t forms an 3 5 e~or ~ignol l~e error signal is used in the same way as the ~lLfô~l~d emboAimont to ~pAote the NCO fi~quo,lcy.
W O 93/00747 - : : PC~r/US92/03763 2 0 8 9 ~ S ~ -8-In ~ , a ~)rOCel38 of ~llt~m~tic frequency control in a rh~ngpn~ e~ o-~m ~nt has been descnbed. By ~ r.
ing the l,e~fo,~a~ce of each ad~li~,e algorithm to dete.~ne how to change the osr~ tQr frequency, the frequency offset 5 can be ~e,~,r~l to almost zero. The l,~oceas of the l ro~e,~t in-vention is not ~ le-l by inter-symbol inte.fe~e~ce since the adaptive rh~nn-,! equalizers take the ISI into PccQ ~nt, in their e~t;m~te~ Comml~nir~tjQn devices using the process of the ~.cc_..t i~re~Lion can out-l,e.fo.~ device~ using only coar~e 10 AFC.
FrLTER
F~do~~ L..
The present iu~,eulion relates ..c~ally to the field of co..~ -..ic~tio~c and particularly to olltQ nq*c frequency con-trol.
1 0 B~do~t~1~venlion U.S. ~ tol cellular (USDC) comml~nico*on8 uses digi-tized voice and data ~i~ol~ for comm~nirot;~n bc~- ee~ a mo-bile telephone and a base station. The mobiles and bases em-l 5 ploy time division multiple acces~ (TDMA) mo~l~lotiQ- for the tronnmi~sj~orl of these sign-l~. A typical format for the TDMA
data bursts is i]lu~L_ted in E IG. 5. This data burst, as well as USDC in general, is ~ in more detail in the USDC
sperifirati~n EIA/TIA IS-54 available from hlc~,lrv~c 20 Tn~ ~ies ~oci~tjon, TP~neeling Del,a.h.lent, 2001 Eye Street, N.W., W~hington, D.C. 20006.
When the mobile mo~es, it may enco~tc~ de- ~ de~l c~.... ir~t;Qn channels due to noise and mllltir~th distor-tion; both noise and distortion va~ ~u~g with time. The multi-2 5 path distortion is due to a signal being ~ecc;ved by the mobile at ~ . t times when it bouuces off btlil~ling~ and terrsin.
Multipath channels csn cause inter-symbol interference (ISI) that can be removed with an adaptive rh~.~..ol eqll~li7er, a s~fic type of adaylive filter. An adayLve channel cFl~ tQr 3 0 is ~n~tl~er type of ad~y1.ive filter.
A typicsl adsptive filter is illustrated in FIG. 1. The in-put signal (106) is p.ocf,sse~l by the adaptive filter (101), pro-rlllring the ad~,1ive filter output signal (102). The o.-ly~ of the filter is then subl~ rte~l (105) from a ~e~e~e..ce signal (103), 3 5 typically the llnfiltered input signal (106), to produce an error signal (104). This error signal (104) is used by an adaptive al-'- ~0~9 ~ ~ 4 gorit~tm with an ~~pAStte coeffi~çnt, ~, in the adaptive filter to l-pA~e the filter coeffi~ipnts~ The nr~Stte coeffiriPnt is also re-ferred to as a ~rL ;~ oçf~i~içnt or msmQry coeffi~ ient~ The ~e~o. ~ of the adaptive algorithm increases as the value of S increa~es.
The adaptive algorithm may be a ~stlmstn~ Rec.,.~i~,e Least Square, or Least Mean Square (LMS) algon~tm The typical goal of the ad~pli~e algonthm is to ~--;~-;---i7e the mean square value of the error signal (104), fised ~p~lote co~ffi~Pn~
10 Thi~ value is typically ~le~i~n~te~l mean square error (MSE).
A detriment~l c~ e,~ristic of adaptive rh~nn~l equal-izer~ i~ that they can esperience degraded ~-~r.---S~nce in the p~e~~ -~ce of a frequency offset of more than a~.o~;...~tely 10 Hz (in a a~al4h- with a 24 kHz symbol rate). While the specifi-l 5 cation for a tr~n~mias;Qn &~lem l~q~ es a certain frequencyv~ t;~ limit, the adaptive ch~nnel eq~li7er may .eq~e a ~t~icte- limit. An esample is t_e USDC ~lem. USDC re-quires the ~ecc;v~- ope.al~ frequency to be locked within 200 Hz of the ~ ;tteY. The adaptive ~h~nn~l eq~ i7er will not 2 0 ~e.fo-m properly in this e..vi.ol.me~t~ A coarse automD~tic frequency control (AFC) is typically used to ~ o~ove most of the offset. Any rem~ining offset, however, can detrim-~nt~lly af-fect a dete~on al~w;~ -- and i..c,osse the detected bit erTor rate. There i8 a res~ ;ng need for an AFC that can reduce the 2 5 frequency offset to an acceptably small level and track any vanation of the offset as the environment changes when the mobile moves.
8umma~y of ~he Iu~L;o~
The l,-ocess of the present invention generates an opti-mum ~tom~;c frequency control signal in an al,l,a~atus that has a plurality of adaptive filters. The process starts by generating a difference signal by colllpa~ g the perforrn~n~es of a subset number of ~' ., ~0~9 ~ 54 adaptive filters with each other. Then the automatic frequency control signal is adjusted in response to the dirre~ ce signal.
B~iefDesc~lion~eDraw~
FIG. 1 shows a block diagram of a typical ~ ve fil-ter.
FIG. 2 shows a block diagram of the y.OU~58 of the pre-10 sent iuve.llion.
FIG. 3 shows a graph of mean square error versusresi~l~Pl frequency offset in accordance with the y.vcess of the y~ee~t iuve~lion.
FIG. 4 shows a graph of re~ ol frequency offset ver-1 5 sus time in accordance with the yrocea3 of the present inven-tion.
FIG. 5 shows the fo~ ~t of a TDMA data burst used in the U.S. digital cellular co~ o-t;Qn ~ e~.
FIG. 6 shows an alternate emho~liment of the y.~cea~ of 2 0 the yreeel,t invçnt;Qn Det~iled Descri~on of t~e Ple&ned Embodin~
The process of the y~e~e~lt invention provides fine au-2 5 tomot;c frequency control in a device using adayl ive filters.The di~ ..ce in pe~ ance of the adaptive filters is used to ~Gdil,~ the AFC ~iEnDl, thereby re~ ng the frequency offset.
A linear coherent digital radio leceivêr typically de-mo~ t~s the ;-~co~ g signal by mi~ the signal to base-3 0 band using a local osrillAtor. The frequency of the local oscil-lator must be kept re~con~l~ly close to the frequency of the tr-o-n~mitter. Af~cer the signal has been mised down to base-band, further analog or digital sign.,l y~OCe5~ is performed to fec~J~er an estim~te of the tr~n~mitte~l data. In the follow-3 5 ing description of the ploce88 of the present invention~ it is as-~llmPd that the b~ebAn~ signal has been co.lve~ l,ed by an ana-~ ~.
20~9 ~ 54 log to digital collvel ler to a fonn suitable for further digital signal proce~ g The preferred emho~limpnt of the process of the present invent;on, as ilh,&~al,ed in FIG. 2, is cG,u~r;sed of t_ree adap-tive filters (1- 3) configured as adaptive ch~nnPl es1;m~tQrs (ACE). The three ACE's (1- 3) _ave an l~p~l~ta- coeffi~ent, ~L, that varies vrith the c-.vh~ent of the dence. The process for determinin~ ,u is described in ~n~ n patent application no. 2,086,548 filed on May 8, 1992 and titled "A Method for Optimi7~tion of Adaptive Pilter Update Coefficient" to Kevin Baum and ~sign~qd to Motorola, Inc.
The update coefficient will remain constant during a single TDMA data burst.
The three ACE's (1 - 3) are i~en1;~1 escept for having di~e~.-t frequency offset dither generators, a dither generator 1 5 being the source of the ~efer~nce ~ l. ACE 2 uses the base-band ~-,c~;~,ed signal that has been mised (9) with the numeri-cally controlled o~cill~tQr (NC0) (8) signal as the ~eferehce s~ l ACE's 1 and 3 mis (11 and 12) this lefe.cl ce signsl with frequency offsets before using it as a ~cfe.e.,ce ~
2 0 The frequency off~et~, ei~ and ei~', and their rPl~t;Q~-ship to the des*ed resi~ln~l frequency of~et, a~, are illustrated g~p~iCplly in FIG. 3. These offsets are located on either side of the frequency offset e~t;m~te, -~. ACE's 1 and 3 act as resid~ frequency offset ~ bes"; relative to ACE 2, ACE 1 2 5 acts as a ~high~ frequency probe and ACE 3 acts as a ~low" fre-quenQ probe. ACE's 1 and 3 esti_ate two points on the MSE
curve. ACE 2 ye.fo-., s the actual desired adaptive filtering fi~nct;on .
The k tenn of the frequency offset denotes the time in-desofthe~mple. TheC)dtermisapplic~t;ondependent CDd should be chQsçn as small as poss-b1e while still allowing a difference in the mean square errors (MSE) to be detsct~
could also vary with time by setting it to a larger value in~ 11y to speed acquisition and then re~lt)ce~l to get the most accurate 3 5 frequency offset c~t;m~te. In the ~-efe~-c~ ~mbo~liment, ~d i8 set to a value of 5 ~ (2t~) r~ n~/secQntl ~.. . .
.~; .
WO 93/00747 ~ 2 0 8 9 1 5 ~ PCl'/US92/03763 ~_ -- 5 --In operation, the process of the ~.ee_~-t invention ini-tially ~e~o~es the c~.e.~t e~ te 1 frequency offset, -~, from the ~--ehS~n~l le~;~,ed signal by mi~ing (9) the signal with the NCO (8) output, ei~. Trli~;sllly, the NCO (8) fre-5 quency, ~, is set to zero if there is no prior knowledge of theinitial frequency offset. This is in~lir~te~l by the initial accu-mnlsttor (7) value being zero.
This signal is then operated on by a ~l~t~ct;n~ algorithm (10) that is driven by the ACE 2 output. The resulting symbol 1 0 ~'e~iQviQ-~ ~Qign~l, a, is input to the three ACE's (1 - 3).
The ACE's (1 - 3) ~;~..e.ste error signs~l-s that are the dif-fe~ the filtered output and the S~Q~c~Slte~l refer-ence sigllals that are ~ 5 etl above. Two of the error sig-nals, errorl ant error3, are input to MSE çstim~tors (4 and 5) that operate s~8 follows:
Ic I n ET1 = j = ~' Ic I n ~ Ic~ r3l2 ET3 = j = ~' where Ic is the same as in the frequency offset and n is the m-m~er of samples of the error signsll As an P~Slmple~ if k = 1 and n = 10 for the first e~ 1;Qr~ cycle, k will stS~rt at 12 for the nest cycle.
2 5 The di~re.once bel~een the e~ ,qte<l MSE~8, Et = ET1 - ET3 . provides an in~lic~ n of which direc~on to move along the frequency offset a~is, illusL~ated in PIG. 3, to get closer to the ...;~;...l.... MSE point (re~i-ln~l offset = O). For e~ e, if the residual frequency offset is greater than 0, ET1 3 0 will be larger than ET3 thus m~ki~ Ed < O. The negative value of Ed in~licPte~ that ~Dv is too large and should be decre-mented.
wo93/00747 ~ 9~ L Pcr/us92/03763 ID the l,.ofer~ed emho~imPnt, Ed i8 input to a cc ~ a-tor (6) where it is compared to 0. In this case, the co~p~ator has an outl,u~ filnet;on, f~Ed), as follows:
S f~Ed) = ~ when Ed > ~-fl~Ed) = ~A when Ed ~ ~.
f~Et)=owhenEd=o~
where A is app~ t;on ~'eFendent and dete,~es the resolu-1 0 tion of the AFC and also the ada~t~l ion speed of the AFC.
can be c~osen as a very small value for a s~ate~ with a coarse AFC. In an alternate eml~QAim"nt, ~ could vary with time bg eet';n& it to a larger value ini*Ally to speed r ~ql~isition and then re~lllc~ to get the most accu~ate frequency offset esti-1 5 mate. In the ~refe~l en~ nent~ ~ is set t~D a value of 21 rp~liAn~/8ec D.~
In an alternate emk)~ r~t, Et is input to a filter in-stead of a aDml,~ato,. The filter prDvides a time ~/~~ step size (CQ-~ d to the f~ed step size of ~) that is responsive to 2 0 the size of the errDr din'e,e,lce si~n~l For e~ample, when the err~Dr di~rè~CnCe 8ignal becom~ large, the gtep gize AlltQ.~AI _ caUy increases resulting in faster co.l~e,~..ca of the algo-rithm. Using the filter, however, ~c.,eases the cQ~nple~ity of the invention and _ay cause st~bility problem~ if a higher 2 5 order filtrer is used. A first order digital ;~ të impulse re-spon~e (IrR) filter i8 lJ~efe~-l 'd due to s~h;lity and simplicity c~nsiterations. The o~ t of the filter is used to ~ te the frequency offset e~Rm~te The output of the co ~-~ stor (6) (or filter) i8 input to an 3 0 ~cc-~ tQr (7) that adds the new input value to the previ-ously stored value. The ~cc--~ te~ value is then used to control the frequency, ~Dv, of the NC0 (8). Since the M~'s ET1 and ET3 are e~;m~te~ over blocks of n s~mple~, Ed and the outputs of the CQ- ~-l'A ' ator (7) and acc ~~ tQr (8) are cal-3 5 ~ t~l every n iterations. The NC0 (8) frequency, therefore,is ~lp~l~te~3 once evely n iter~Ro~ ~.
wo 93/00747 2 0 8 9 1 5 ~ Pcr/uss2/03763 _~ 7 As illustrated gr~p~ir~lly in FIG. 3, after several NC0 (8) ~ te cycles, ~b will be &~lJlo~;m~tely equal to ~- and the rç~;At~l fieque~ offset will be alJ".o~ Ptely zero. If the frequency offset changes, the ~ Ce~3 of t_e ~ ~e,lt invention S Aetect~ and tracks the change.
The operation of the ~locea3 of the ~ ce,lt invention can be seen in graphically FIG. 4. The process is using an LMS
adaptive rh~nnPl estimator with an initial frequency offset of 50 Hz. The bit error rate of the ~letector in t_is e-~ -le is 19~o.
1 0 Note that the rçsi~ frequency offset quickly ~l~rline~ to nearly 0 Hz. The slope of the initial change from 50 Hz can be changed by modil~u~ A. A larger value for ~ will cause a faster acq~ tion and, thelefolo, a stee~er slope.
In the ~.~f~.,~l embQAiment, the ~ cess of the l"oee~t iu~,e.lLion is impl~Pm ~nte A as an alg~.l;t,l~. Alternate çm~oA-iments of the invention can be implementet in hart~ ~d or combinations of ha~l- ~e ant eoIl . ~e; each block of the pro-cess being either an algonthm or a hardware circuit eq~uva-lent of that block.
2 0 Another alternate emboA;~ -.t can use only two adap tive filters by not using the second adaptive filter. In this em-bodiment, the o~ t of one of the filters replaoes the seCQn~l filter's o~ t. The re~llting AFC value will be biased by ~12.
Still another alternate emboAinlent, illus~llsted in PIG.
2 5 6, can use ataptive eql~-li7ers in place of the rh-nnP~l estima-tors. In this PmhoAin~ent, the ~efO~euce signal and the symbol tecision ~ign~l, a, are input to the equalizer. The ataptive equalizers (601- 603) are ol~lati~,e to loh~o~e the ISI from the ,e~ titheret ~ece;v~,t ~ignol~. The adaylive eq~oli7er 3 0 may have some inherent telay until an output le~i~ to t_e current input is avoil~Ahle. The symbol decisions, a, are delayed (604 - 606) until the eqn-li77Or o~ t coms~ L to that Ae~iPiQ- is avAilohle T_e diffO~uce between the symbol decisions and the coll~pQnAing eq~-ou~-or o~ t forms an 3 5 e~or ~ignol l~e error signal is used in the same way as the ~lLfô~l~d emboAimont to ~pAote the NCO fi~quo,lcy.
W O 93/00747 - : : PC~r/US92/03763 2 0 8 9 ~ S ~ -8-In ~ , a ~)rOCel38 of ~llt~m~tic frequency control in a rh~ngpn~ e~ o-~m ~nt has been descnbed. By ~ r.
ing the l,e~fo,~a~ce of each ad~li~,e algorithm to dete.~ne how to change the osr~ tQr frequency, the frequency offset 5 can be ~e,~,r~l to almost zero. The l,~oceas of the l ro~e,~t in-vention is not ~ le-l by inter-symbol inte.fe~e~ce since the adaptive rh~nn-,! equalizers take the ISI into PccQ ~nt, in their e~t;m~te~ Comml~nir~tjQn devices using the process of the ~.cc_..t i~re~Lion can out-l,e.fo.~ device~ using only coar~e 10 AFC.
Claims (6)
PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. A method for generating an automatic frequency control signal in an apparatushaving a plurality of adaptive filters, each adaptive filter having an adaptive algorithm, a subset number of adapter filters of the plurality of adaptive filters having a reference signal with an associated frequency dither, the method comprising the steps of:
generating a difference signal by comparing performances of the subset number of adaptive filters with each other; and adjusting the automatic frequency control signal in response to the difference signal.
generating a difference signal by comparing performances of the subset number of adaptive filters with each other; and adjusting the automatic frequency control signal in response to the difference signal.
2. A method for automatic frequency control in a system having a baseband received signal, an oscillator signal, and a symbol decision signal, the method comprising the steps of:
multiplying the baseband received signal with the oscillator signal to produce afirst reference signal;
multiplying the first reference signal by a first offset signal, having a first magnitude, to produce a second reference signal;
multiplying the first reference signal by a second offset signal, having a second magnitude to produce a third reference signal;
generating a first error signal in response to the symbol decision signal and the first reference signal;
generating a second error signal in response to the symbol decision signal and the second reference signal;
generating a third error signal in response to the symbol decision signal and the third reference signal;
estimating a first mean square error in response to the first error signal;
estimating a second mean square error in response to the third error signal;
subtracting the second mean square error from the first mean square error to produce a mean square error difference; and modifying the oscillator signal in response to the mean square error difference.
multiplying the baseband received signal with the oscillator signal to produce afirst reference signal;
multiplying the first reference signal by a first offset signal, having a first magnitude, to produce a second reference signal;
multiplying the first reference signal by a second offset signal, having a second magnitude to produce a third reference signal;
generating a first error signal in response to the symbol decision signal and the first reference signal;
generating a second error signal in response to the symbol decision signal and the second reference signal;
generating a third error signal in response to the symbol decision signal and the third reference signal;
estimating a first mean square error in response to the first error signal;
estimating a second mean square error in response to the third error signal;
subtracting the second mean square error from the first mean square error to produce a mean square error difference; and modifying the oscillator signal in response to the mean square error difference.
3. An apparatus for automatic frequency control in a device having a received signal, oscillating means for generating an oscillator signal, symbol detection means for generating a symbol decision signal, and a plurality of multiplying means for generating a plurality of reference signals, the apparatus comprising:
first adaptive filtering means, coupled to the symbol detection means and first multiplying means of the plurality of multiplying means, for generating a first error signal in response to the symbol decision signal and a first reference signal of the plurality of reference signals;
second adaptive filtering means, coupled to the symbol detection means and second multiplying means of the plurality of multiplying means, for generating a second error signal in response to the symbol decision signal, the oscillator signal, and the received signal;
third adaptive filtering means, coupled to the symbol detection means and third multiplying means of the plurality of multiplying means, for generating a third error signal in response to the symbol decision signal and a second reference signal of the plurality of reference signals;
first means for estimating mean square error, coupled to the first adaptive filtering means, for generating a first mean square error estimate;
second means for estimating mean square error, coupled to the third adaptive filtering means, for generating a second mean square error estimate;
summing means, coupled to the first and second means for estimating mean square error, for producing a difference signal;
means for producing a delta signal coupled to the summing means;
accumulation means, coupled to the means for producing a delta signal and the oscillating means, for accumulating the delta signal.
first adaptive filtering means, coupled to the symbol detection means and first multiplying means of the plurality of multiplying means, for generating a first error signal in response to the symbol decision signal and a first reference signal of the plurality of reference signals;
second adaptive filtering means, coupled to the symbol detection means and second multiplying means of the plurality of multiplying means, for generating a second error signal in response to the symbol decision signal, the oscillator signal, and the received signal;
third adaptive filtering means, coupled to the symbol detection means and third multiplying means of the plurality of multiplying means, for generating a third error signal in response to the symbol decision signal and a second reference signal of the plurality of reference signals;
first means for estimating mean square error, coupled to the first adaptive filtering means, for generating a first mean square error estimate;
second means for estimating mean square error, coupled to the third adaptive filtering means, for generating a second mean square error estimate;
summing means, coupled to the first and second means for estimating mean square error, for producing a difference signal;
means for producing a delta signal coupled to the summing means;
accumulation means, coupled to the means for producing a delta signal and the oscillating means, for accumulating the delta signal.
4. The apparatus of claim 3 wherein the first, second and third adaptive filtering means are adaptive channel estimators.
5. The apparatus of claim 3 wherein the first, second and third adaptive filtering means are adaptive equalizers.
6. A method for automatic frequency control in a system having a baseband received signal, an oscillator signal, and a symbol decision signal, the method comprising the steps of:
multiplying the baseband received signal with the oscillator signal to produce afirst reference signal;
multiplying the first reference signal by a first offset signal, having a first magnitude, to produce a second reference signal;
multiplying the first reference signal by a second offset signal, having a second magnitude, to produce a third reference signal;
generating a first error signal in response to the symbol decision signal and the second reference signal;
generating a second error signal in response to the symbol decision signal and the third reference signal;
estimating a first mean square error in response to the first error signal;
estimating a second mean square error in response to the second error signal;
subtracting the second mean square error from the first mean square error to produce a mean square error difference; and modifying the oscillator signal in response to the mean square error difference.
multiplying the baseband received signal with the oscillator signal to produce afirst reference signal;
multiplying the first reference signal by a first offset signal, having a first magnitude, to produce a second reference signal;
multiplying the first reference signal by a second offset signal, having a second magnitude, to produce a third reference signal;
generating a first error signal in response to the symbol decision signal and the second reference signal;
generating a second error signal in response to the symbol decision signal and the third reference signal;
estimating a first mean square error in response to the first error signal;
estimating a second mean square error in response to the second error signal;
subtracting the second mean square error from the first mean square error to produce a mean square error difference; and modifying the oscillator signal in response to the mean square error difference.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US72282891A | 1991-06-28 | 1991-06-28 | |
| US722,828 | 1991-06-28 | ||
| PCT/US1992/003763 WO1993000747A1 (en) | 1991-06-28 | 1992-05-08 | Automatic frequency control by an adaptive filter |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CA2089154A1 CA2089154A1 (en) | 1992-12-29 |
| CA2089154C true CA2089154C (en) | 1999-08-03 |
Family
ID=24903558
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA002089154A Expired - Fee Related CA2089154C (en) | 1991-06-28 | 1992-05-08 | Automatic frequency control by an adaptive filter |
Country Status (7)
| Country | Link |
|---|---|
| JP (1) | JP2830471B2 (en) |
| KR (1) | KR960005386B1 (en) |
| CA (1) | CA2089154C (en) |
| DE (2) | DE4292274C2 (en) |
| GB (1) | GB2263829B (en) |
| MX (1) | MX9203307A (en) |
| WO (1) | WO1993000747A1 (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5394071A (en) * | 1993-02-19 | 1995-02-28 | Mts Systems Corportion | Control network with on-line iteration and adaptive filter |
| JPH09219693A (en) * | 1996-02-09 | 1997-08-19 | Mitsubishi Electric Corp | Digital broadcast receiver |
| JP3556047B2 (en) * | 1996-05-22 | 2004-08-18 | 三菱電機株式会社 | Digital broadcast receiver |
| KR19990069891A (en) * | 1998-02-13 | 1999-09-06 | 전주범 | Center frequency tracking method using adaptive regression digital filter algorithm |
| JP3846546B2 (en) | 2000-08-29 | 2006-11-15 | 日本電気株式会社 | Frequency offset estimator |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4466108A (en) * | 1981-10-06 | 1984-08-14 | Communications Satellite Corporation | TDMA/PSK Carrier synchronization without preamble |
| US4691176A (en) * | 1986-03-17 | 1987-09-01 | General Electric Company | Adaptive carrier tracking circuit |
| US4852090A (en) * | 1987-02-02 | 1989-07-25 | Motorola, Inc. | TDMA communications system with adaptive equalization |
| CA1332450C (en) * | 1987-06-23 | 1994-10-11 | Shousei Yoshida | Carrier-to-noise detector for digital transmission systems |
| US4896336A (en) * | 1988-08-29 | 1990-01-23 | Rockwell International Corporation | Differential phase-shift keying demodulator |
| US5230007A (en) * | 1991-06-28 | 1993-07-20 | Motorola, Inc. | Method for optimizing an adaptive filter update coefficient |
-
1992
- 1992-05-08 CA CA002089154A patent/CA2089154C/en not_active Expired - Fee Related
- 1992-05-08 WO PCT/US1992/003763 patent/WO1993000747A1/en active Application Filing
- 1992-05-08 DE DE4292274A patent/DE4292274C2/en not_active Expired - Lifetime
- 1992-05-08 GB GB9303663A patent/GB2263829B/en not_active Expired - Lifetime
- 1992-05-08 KR KR1019930700393A patent/KR960005386B1/en not_active IP Right Cessation
- 1992-05-08 DE DE4292274T patent/DE4292274T1/de active Pending
- 1992-05-08 JP JP5501452A patent/JP2830471B2/en not_active Expired - Fee Related
- 1992-06-25 MX MX9203307A patent/MX9203307A/en unknown
Also Published As
| Publication number | Publication date |
|---|---|
| GB2263829A (en) | 1993-08-04 |
| KR930701869A (en) | 1993-06-12 |
| DE4292274T1 (en) | 1993-07-15 |
| GB2263829B (en) | 1995-07-26 |
| DE4292274C2 (en) | 1999-05-27 |
| WO1993000747A1 (en) | 1993-01-07 |
| JPH05508989A (en) | 1993-12-09 |
| JP2830471B2 (en) | 1998-12-02 |
| GB9303663D0 (en) | 1993-05-05 |
| MX9203307A (en) | 1992-12-01 |
| CA2089154A1 (en) | 1992-12-29 |
| KR960005386B1 (en) | 1996-04-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7937058B2 (en) | Controlling the bandwidth of an analog filter | |
| KR100466471B1 (en) | Integrated receiver | |
| US7994870B2 (en) | Tunable filter with gain control circuit | |
| CN112118199B (en) | High dynamic signal frame structure and high dynamic signal tracking method | |
| AU748203B2 (en) | Apparatus and methods for tuning bandpass filters | |
| US7932844B1 (en) | Circuits and methods for calibrating a frequency response of a filter | |
| US8295798B2 (en) | Tunable filters with lower residual sideband | |
| US5233633A (en) | Automatic frequency control by an adaptive filter | |
| CA2357491A1 (en) | Filter tuning using direct digital sub-sampling | |
| US6057735A (en) | Amplifier for continuous high gain, narrowband signal amplification | |
| CA2089154C (en) | Automatic frequency control by an adaptive filter | |
| US3757221A (en) | Automatic equalizer system for phase modulated data signals | |
| US9231634B2 (en) | Compensation apparatus and wireless communication equipment | |
| US6593828B1 (en) | System and method for filter tuning | |
| US5789993A (en) | Amplitude/frequency correcting device and corresponding frequency equalizer | |
| US8724751B2 (en) | Method for a single radio aggregated spectrum receiver, computer program, receiver, and terminal | |
| EP1154604A1 (en) | DC-offset compensation method and apparatus | |
| US7714643B1 (en) | Tuning of analog filters | |
| US4363131A (en) | Built-in group delay testing arrangement for an FM radio system | |
| JP3964346B2 (en) | FM signal receiver and radio communication apparatus using the same | |
| US5706221A (en) | Mehtod and apparatus for recovering digital data from baseband analog signal | |
| KR20080072790A (en) | Filter calibration | |
| CA1095593A (en) | Active filter | |
| CN114024521B (en) | Broadband variable frequency link group delay equalization method and system | |
| Wölfel et al. | Self-calibrating highly integrated radio frequency front-end for parallel π/4 DQPSK decoding |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request | ||
| MKLA | Lapsed |