CA1294073C - Tdma radio system employing bpsk synchronization for qpsk signals subject to random phase variation and multipath fading - Google Patents
Tdma radio system employing bpsk synchronization for qpsk signals subject to random phase variation and multipath fadingInfo
- Publication number
- CA1294073C CA1294073C CA000575635A CA575635A CA1294073C CA 1294073 C CA1294073 C CA 1294073C CA 000575635 A CA000575635 A CA 000575635A CA 575635 A CA575635 A CA 575635A CA 1294073 C CA1294073 C CA 1294073C
- Authority
- CA
- Canada
- Prior art keywords
- data signal
- tdma
- phase variation
- multipath
- qpsk
- 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 - Lifetime
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/18—Phase-modulated carrier systems, i.e. using phase-shift keying
- H04L27/22—Demodulator circuits; Receiver circuits
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/24—Radio transmission systems, i.e. using radiation field for communication between two or more posts
- H04B7/26—Radio transmission systems, i.e. using radiation field for communication between two or more posts at least one of which is mobile
- H04B7/2643—Radio transmission systems, i.e. using radiation field for communication between two or more posts at least one of which is mobile using time-division multiple access [TDMA]
Abstract
TDMA RADIO SYSTEM EMPLYING BPSK SYNCHRONIZATION
FOR QPSK SIGNALS SUBJECT TO RANDOM PHASE
VARIATION AND MULTIPATH FADING
Abstract of the Invention A time division multiple access (TDMA) radio system conveying a quadrature phase shift keying (QPSK) modulated data signal is disclosed. In order to overcome multipath distortion and phase variation introduced by the radio channel, a synchronizing portion of a TDMA
timeslot is transmitted in only one veotor of the QPSK
modulation. The system utilizes the one vector synchronizing portion to develop a channel profile estimate which is applies to the received message portion of the QPSK data signal to compensate for the multipath distortion and phase variation during one TDMA timeslot.
FOR QPSK SIGNALS SUBJECT TO RANDOM PHASE
VARIATION AND MULTIPATH FADING
Abstract of the Invention A time division multiple access (TDMA) radio system conveying a quadrature phase shift keying (QPSK) modulated data signal is disclosed. In order to overcome multipath distortion and phase variation introduced by the radio channel, a synchronizing portion of a TDMA
timeslot is transmitted in only one veotor of the QPSK
modulation. The system utilizes the one vector synchronizing portion to develop a channel profile estimate which is applies to the received message portion of the QPSK data signal to compensate for the multipath distortion and phase variation during one TDMA timeslot.
Description
~ - \
4Q~3 TDMA RADIO SYSTEM EMPLOYING BPSK SYNCHRONIZATION
FOR QPSK SI&NALS SUBJECT TO RANDOM PHASE
VARIATION AND MULTIPATH FADING
5Backqround of the Invention This invention relates generally to digital radio systems and more specifically to a system transmitting, receiving, and demodulating TDMA QPSK modulation in a multipath fading environment such as an environment where 10the transceivers may be in motion and are served from fixed site transmitters and receivers such as in a radiotelephone system. This invention is related to instant assignee's Canadian Patent Applications 576,174 "Rapid Reference Acquisition and Phase Error Compensation ~15for Radio Transmission of Data", filed August 31, 1988 ; (the inventors being Labedz et al) and 576,640 "Phase-Coherent TDMA Quadrature Receiver for Multipath Fading Channels", filed on September 7, 1988 (the inventors being Borth et al) and containing related subject matterO
20In a typical environment, a UHF or microwave radio channel exhibits a multipath structure in addition to ~:; Rayleigh fading. Thus, a radio transceiver for a mobile or portable TDMA system operating at high data rates must accommodate reception of multiple replicas of the 25transmitted signal, each with a random magnitude, phase, '' ~
~' ~.~
, ~
: . ,,f ~I
:' ~IL2~40 73 and time delay with respect to the transmitted signal.
Without corrective measures, the data message can be obliterated by the multipath signals. As early as 1958, a receiver capable o~ accommodating these impairments was 05 described for the use o~ either Differential Binary Phase-Shift Keying (DBPSK) or noncoherent Fre~uency-Shift Keying (FSK). It employed a channel sounding method to estimate the channel impulse response or channel profile, and a transversal equalizer having taps which were adjusted in response to the estimated channel profile.
By 1960 the multipath channel had been exhaustively studied and simulated, and optimum reception had been de~ined, but largely avoiding phase coherent techni~ues.
Such demodulation techniques do not permit the use of higher spectral e~ficiency modulation methods which employ two-dimensional signaling techniques such as shaped Quadratuxe Phase-Shift Keying (QPSK) and its variations.
By 1983, TDMA (Time Division Multiple Access) receivers for digital telephony using Binary Phase-Shift Keying (BPSK) phase coherent detection had been described in U.S. Patent No. 4,587,662. In 1985 this was extended to include QPSK, but the receiver was only described in general terms.
In 1986 an MSK receiver, with possible application to QPSK, was reported which could accommodate two rays of multipath and which used an adaptive equalizer employing both ~eedforward and feedback filtering. (See Krister Raith et al., "Multi-Path Equalization for Digital 30 Cellular Radio Operating at 300 kbits/s", 36th IEEE
Vehicular ~onference, pp.268-272, May 1986). Although this adaptive e~ualizer apparently has never been thoroughly described in the literature, it is different ; than the multipath correlation employed in the present invention since it requires decisions to be made on the output in order to adjust the equalizer.
1294~)~3 Adaptive egualization generally operating continuously on the data being received has been utilized in digital microwave receivers receiving continuous data streams. Such continuous receivers can equallze over a 05 relatively long period o~ time. TD~A, due to its burst-like characteristics, demands rapid determination of the channel pr~file including the significant multipath contributions. Even though the data transceiver may be moving, the channel profile can be a~sumed to undergo negligible changes in a given TDMA timeslot if the timeslot is sufficiently short in duration.
Summary of the Invention Therefore, it is one ob~ect of the present invention to synchronize a TDMA receiver to a synchronization sequence transmitted during a timeslot to compensate for multipath signals and random phase introduced by a radio channel to transmitted digital information.
It is another object of the pre~ent invention to transmit the synchronization saquence on a single vector of the QPSK modulated signal.
It is a ~urther ob~ect of the present invention to create a radio channel multipath profile from the synchronization sequence and utilize that model during only on~ time~lot period.
Accordingly, these and other objects are achieved in the present invention which encompasses a time division multiple access (TDMA) radio system which conveys a data message using quadrature phase shift keying (QPSK) ~ modulation and a synchronization sequence in a timeslot ; over a radio channel. The transmitter transmits the synchronization seguence in only one vector of the QPSK
modulation and the receiver receives and employs the one vector synchronization to compensate the data message for multipath and phase vaxiation during one TDMA timeslot.
~2~0~73 Brief_Descri~ion of the Drawin~s Figure 1 is a block diagram of a data transmission system employing quadrature digital transmission and reception.
Figures 2A and 2B are, together, a block diagram of a TDMA receiver which may receive QPSK signals.
Figure 3 is a block diagram of a TDNA receiver signal processor which may advantageously employ the present invention.
Figure 4 is a block diagram of the peak detector circuit of the receiver of Fig. 3.
Figure 5~ is a graph of the synchronization correlator outputs CI(t) and CQ(t) plotted against time and showing a possible set of outputs including a correlati~n detection.
Figure 6 is a block diagram of a QPSK transmitter which transmits a synchronization sequence during a portion of a TDMA time slot in accordance with the present invention.
Figure 7 is an illustration of the TDMA signal format which may be transmitted from the transmitter of Fig. 6.
Description of the Preferred Embodiment A radio frequency system conveying a data signal from a transmitter 101 to a receiver 103 is shown in Fig.
1. In a radiotelephone system, transmitter 101 would be a fixed site transmitter serving a radio coverage area which would be populated by mobile or portable transceivers, the receiver of which is shown as receiver 103. Additionally, more than one radio coverage area can be linked in such a way that continuous coverage may be provided over a wide area, i.e., a cellular radio telephone system or an integrated area such as an office building or shopping mall. (See, for example, instant ... .
. .
1~4(~73 assignee ' s Canadian Patent Application No. 560, 416 "Microcellular Communication System Using Macrodiversity", filed March 3, 19~8, the inventor being Gerald P . Labedz ) .
In the preferred embodiment, quadrature phase shift keying (QPSK) is employed to increase the through put of the channel although other multi-dimensional signaling may equivalently be employed. Further, the well-known time division multiple access (TDMA) technique of sharing a limited channel resource among a large number of users is employed in the present invention. Each of the users is assigned a brief period of time (a timaslot) during which a message may be transmitted to or received from the user. The advantages of such a TDMA technique over other techniques (such as frequency division multiple access l'DMA) are: a) no duplexer is required for full duplex communications, b) variable data rate transmission may be accommodated through the use of multiple adj acent time slots, c) a common radio frequency power amplifier may be used to amplify multiple channels at any power level without the combining losses or intermodulation distortion present with FDMA, and d) a capability of scanning other "channels (timeslots) without requiring separate receivers may be providPd.
The high data rate employed in the present invention (200Kbps to 2Mbps) exceeds the channel coherence bandwidth of the mobile radio channel for many urban and suburban environments. As a result, the channel exhibits a multipath structure in addition to the expected Rayleigh fading. The receiver of the present invention enables TDMA quadrature signals to be coherently received over a multipath fading channel. This embodiment will demodulate a 2-megachip/sec QPSK radio signal, the only 3LZ~ 3 constraint being that the acquisition sequence be transmitted as a binary phase shi~t keying (BPSK) signal with a predetermined phase relative to the QPSK data.
Figs. 2A and 2B are a block diagram of a TDMA
quadrature phase shift keying data and is described in instant assignee's PCT International Publication W0 88/05981 "TDMA Communications System with Adaptive Equalization" published on August 11, 1988 and filed on behalf of David E. Borth.
The digital signal outputs of the A/D converters 209 and 211, respectively, are applied to in-phase (I) time slot correlator 213 and quadrature (Q) correlator 215, respectively, as well as to their respective signal buffers 217 and 219. I correlator 213 performs a correlation function between all received bits of the input signal and a pre-loaded synchronization word (I
sync word) corresponding to the in-phase time slot sync word.
The output of I correlator 213 is a digital bit stream representing the sample-by-sample correlation of the received data with the stored synchronization word replica for the timeslot. The correlation function exhibits a peak when the I sync word is located in the received sample data. In the same way, Q correlator 215 performs a correlation function between the pre-stored guadrature Q sync word from memory 221 and the sampled quadrature (Q) input.
The outputs of correlators 213 and 215 are applied to squaring blocks 223 and 225, respectively. I~he squaring blocX output signals represent the squared values of the separate I and Q correlation operations respectively. The squaring block outputs are then applied to summing block 227. The I and Q correlation .
- 6a -signals are summed together to form a squared envelope signal which represents the sum of squares of the correlation signal. The squared envelope of the correlation signal makes an explicit determination of the phase ambiguity unnecessary. Thus, without resolving any ~29~73 ambiguity, a large amplitude signal output from summing block 227 repres~nts a possible start location for a particular timeslot.
The output of summing block 227 is then routed to 05 time slot detector 229, wherein the summed correlation signal is compared with a predetermined threshold value.
This threshold value represents the minimum allowable correlation value which would represent a detected time-slot. If the summed output is greater than the threshold value, a time slot detect signal is generated and applied to system timing controller 231.
Timing controller 231 functions as a phase-locked loop (PLL), using a stable timing reference to validate the timeslot detect signal and provide a validated detect output signal. The validated timeslot detect signal is applied to AND gate 233 along with a bit clock output.
The combined timeslot detect/bit clock signal i8 then routed to the I and Q signal buffers 217 and 219, re~pectively. Data ~ignals are clocked lnto 6ignal buffers 217 and ~19 using the combined detect/bit clock signal.
In the implementation shown in Figs. 2A and 2B, a conventional baseband synchronous decision feedback equalizer (D~E) 234 is employed for data signal recovery.
The DFE 234 basically consists of two parts: a forward linear transversal filter 235 and a feedback linear transversal filter 237. The forward filter 235 attempts to minimize the mean-square-error (MSE~ due to intersymbol interference (ISI), while the feedback filter 237 attempts to remove the ISI due to previously detected symbols.
The decision feedback equalizer 234 structure is adapted at least once each time slot in order to compensate for the effects of the time-varying multipath profile. The equalized and quantized complex data output from quantizer 238 is applied to multiplexer 239 for 2:1 ~Z~ 3 multiplexing together with the data clock and output as an output data word.
Returning to Fig. 1, in a QPSK communication system, a transmitted signal x(t) may be expre sed as:
x(t)=a(t)cos~ct+b(t)sin~ct (1) ~ where a(t) and b(t) are the in-phase and quadrature : information signals and ~c is the carrier frequency of the QPSK signal in radians/sec.
~ A frequency-selective (or delay-spread) channel that : is, a radio channel subject ~o multipath inter~erence, may be ch~racterized by an equivalent channel impulse ; response given by:
h(t)~:xo6(t-ro)+~l6(t-r )+~ 6(t-r )~ . . .
m =~ Ixi6(t-ri)~ (2) i-0 : where ~i is the amplitude o~ the i-th resolvable path, ri is the (excess) path delay associated with the th resolvable path, and m+l is the total number of resolvable paths.
For a channel input given by eguation (1), the output of the equivalent delay-spread channel having the impulse response of equation (2) is essentially constant during any given timeslot, and given by.
m y(t)-x(t)*h(t)=~rx(r)h(t-r)dr=J`x(r)~ 6(t-r-ri)dr i=O
m , =~i J;~(r)~(t-r-ri)dr i~O - a~
~, m y~t)=~iX(t Ti) O
=~ i [a(t-r; )Cos~c(t-r~ b(t-ri )Sin~c(t~ri )] (3) It is this signal, ylt), which is input to rsceiver ;103. When the local oscillator reference 105 in the receiver has a phase of~set of y with respect to the (direct-path) received QPSX transmiscion, the receiver local oscillator reference may b2 given by cos (~ct+~) and is es~entially constant during a TDMA timeslot.
(Although the antenna is shown connected to the mixers 107 and 111, it is likely t~at additional signal processing will be required for higher frequ~ncy radio ignals. If down-conversion to an intermediate frequency is used, the output frequency of local oscillator may be different~. Let UI(t) denote the output of the mixer 107 2~ in the uncorrected in-phase branch of the receiver and lat UI'(t) denote the low-pa~ ~iltered version of UI(t)output from low pas~ filter 109. Similarly, let UQ(t) denote the output of the mixer 111 in the uncorrected quadratur~ phase branch of the receiver and ~25 let UQ'(t) denote the low-pas~ filtered version of UQ(t) -~from filter 113. UI'(t) and UQ'(t) ar~ subsequently input to signal processor 115 for resolution into I and Q
data and then cou'pled to data signal recovery 117.
UI(t) i9 give~ by:
c ~i~0 i [ ( i ) c ( i ) ( i ) in~c ( t ri ) ]
:~ m ~ i [(l/2)a(t-ri)(cos{~+ocr~)+Cost2~)ct~ cr ))+
3 5 i 0 1 i (1/2)b(t-ri ) (sint21~ct+~-~cri ) -sint~+~cri ))] . (4) .
The low-pass filtered ver~ion UI'(t) of UI(t) is given by:
m 05 UI (~ /2)~i[a(t-ri)CS(y+~cri)-b(t-ri)sin(~*~)cri)] (5) Similarly UQ(t) is given by:
m c Yi~ oi [ a ( t - ri ) Cos~c ( t - ri ) +b ( t - ri ~ s in~ ( t - r ) ]
m ='~ i [ (1/2)a(t-ri ) (sin~20ct+y-ocri ~+sin~y+~)cri })+
i--O
(1/2)b(t-ri ) (Cos~y+~ocri ) -Cos~2~ct+y-~cri ~) ] (6) and UQ'(t) i8 given by:
m i~ ( / ) i la(t-ri )9in(Y+~cri )+b(t-ri )Cos(y+locr ) ] (7) :~ 20 Con~idering the operation of the present invention `~ in mathematical form, it is an important feature that the : transmitted slgnal xT(t) duriny the synchronization (or ~: training~ phase of the equalizer 115 i9 a BPSK signal.
;~ ~
~: 30 ; 35 When kransmitted in the I phase it is given by:
XT(t)-aT(t)COs~Ct (a) 05 where signal aT(t) (not shown) is a predetermined synchronization sequence with good aperiodic autocorrelation properties, such as one of the Barker sequencss.
The uncorrected in-phase and guadrature receiver branch outputs corresponding to the synchronizing transmitted signal xT(t) may be found by substituting the signal of equation (8) in the received and low pass filtered signals UII(t) and UQ'(t) of equations (5) and (7) respectively, yielding:
1~
m ,~ UI'T(t)~(l/2)~i[aT(t-ri)CS(~+~c~i)] (9) i=0 .~
and, : 20 UQ~T(t) ~(l/2)c~i[aT(t-ri)sin(~+~cri)](10) Thus UI'(t) and UQ'(t) are defined during the : 25 training phase as IlT" as shown in eguations (9) and (10).
: Referring now to Figuxe 3 which illustrates the pre~erred embodiment of the present invention in block diagram ~orm, the signals UIIT(t) and UQ'T(t) are ~IL2~ 73 applied to synchronization correlators (303 and 305, respectively) via conventional fast A/D converters 307 and 309. In the preferred embodiment, synchronization correlators 303, 305 are 4 by 32 bit digital finite 05 impulse response (FIR) filters programmed to provide signed weighted correlation outputs. Synchronization correlator~ 303 and 305 are realized by an IMS A100 Cascadable Signal Processor available from Inmos Corp., Colorado Springs, Colorado. The outputs of aorrelators CI(t) and CQ(t) which are, in simple terms, weighting factors for each i-th resolvable path, generated during reception of the acquisition ~equence, may have the appearance as shown in Fig. 5 and are given by:
m CI(t)~ (l/2)'~iCOs( Y~>cri)~(t-ri) (11) and, Q(t)~ /2)~iSin(~+~Cri)~(t-ri)- (12) The ~ function in equations (11) and (12) determine when to sample the in-phase and quadrature receiver ; branch outputs and the ~ factor provides a weighting for each i-th resolvable pas~ contribution. In the preferred e~bodiment, a sequen¢e controller 311 is realized using a : conventional microprocessor (such as an ~C68020 microprocessor available from Motorola, Inc.3 and associated memory and timing dividers. The sequence 3L~94~73 controller 311 loads a predetermined normalized replica of the acquisition sequence (32 each 4-bit words) stored in the memory of sequence controller 311 into synchronization correlators 303 and 305 prior to the 05 desired TDMA timeslot to be demodulated. TDMA frame ; timing is determined by the sequence controller 311 employing a conventional framing al~orithm to confirm and maintain timeslot acquisitionO
Synchronization correlators 303 and 305 2ach correlate the stored acquisition ~eguence against the last 32 received A/D ~amples, and for each new sample perform another complete correlation. While receiving noise or random data, the outputs CI(t) and CQ(t) of synchronization correlators 303 and 305 ars small numbers of either polarity, emerging at the same rate as the A/D
sampling rate (4 per chip interval). If the radio channel were free of noise and not degraded by multipath, then when an acquisition sequence has been received and digitized and entered into the correlator~ 303 and 305, CI(t) and CQ(t) would simultaneously manifest a pair (or sometimes two ad~acent pairs) of signed numbers signi~icantly larger than those produced by noise or random data, ~uch that the root sum o~ squares of these numbers would be proportional to the ~agnitude of the received signal, and the phase angle ~ relative to the local re~erence oscillator is:
~ arc tan [CQ(t)/CI(t)]- (13) ::
,~
~L2~4C~3 - 14 - CE00027~
In the presence of multipath, each ~igni~icant path will result in the presence of ~uch a peak pair appearing on CI(t) and CQ(t), the signs and magnitudes of each pair of outputs at each peak defining the delay, phase 05 angle, and amplitude contribution of that path to the : total, fulfilling the equations (11) and (12). Thus, each sequence of number~ CI(t) and CQ(t) are bipolar multipath channel profile e~timates, which re~emble a cla~sic multipath channel profile, except that they are bipolar.
Each of the M/PATH correlators 312, 313, 315, and 317 are FIR filters of at least 32 taps. In the preferred embodiment, each ~/PATH correlator is realized : by an IMS A100 Cascadable Signal Proces~or (available : 15 from Inmos Corp., Colorado Springs, Colorado) conventionally connected as a correlator. During the acquisition sequence at the beginning of each desired timeslot, CI(t) i~ shifted into the TAP control entry :~ o~ ~/PATH correlators 312 and 317, and CQ(t) is shifted : 20 into the TAP con~rol entry of ~/PATH correlator~ 313 and 315. Peak detector 318 is shown in Fig. 4 and comprises a root 8um 0~ squares approximator 401 and a threshold detector 403 having an output which ~ignal~ the sequence controller 311 of the first signi~icant ray of multipath.
Th~ GequenCe controller 311 then provide~ just enough : additional reference port clocks to shift this peak all but through the M~PATH correlators, thereby capturiny CI(t) and CQ(t~ in their respective M/PATH
correlators. In the preferred embodiment, the root sum of squares approximator 401 is realized e~ploying a magnitude adder 405 which adds lCI(t) I and (1/2)lCQ(t)l and magnitude adder 407 which adds ICQ(t) I and (1/2) ICI(t) I . The outputs of magnitude adder 405 and magnitude adder 407 are input to 35 conventional magnitude comparators 409 and 411, respectively, where the root sum of squares approximation ~ z9~ 3 is compared to a predetermined threshold to generate an output to the ~equence controller 311 (via OR gate 413).
Thi~ and other approximations to the square root of the ~um of the squares may be found in, eg., A.E. Filip, "A
05 Baker ~ 8 Dozen ~agnitude Approximations and Their Detection Stati~tics," IEEE Transaction~ on Aerospace and Electronic Syste~s, vol. AES-12, pp.86-89, January 1976.
Thi~ output to the ~eguence controller 311 i~ ~hown as td in the example of Fig. 5. Thus, the peak detector 318 reports the first eignificant peak to the ~equence controller 311 which, in turn, ~tarts the loading at ~/S
~top, to thus capture the channel profile in each of the M/PATH correlator~.
The four M/PATH correlators (312, 313, 315, and 317 in Fig. 3) thus have the information available to perform equations (14)-(17~, below, whose results (A, B, C, and D) appear at the outputs of M/PATH correlator~ 312, 313, 315, and 317 respectively.
20m A-CI (t)UI ' (t)~ [ (l/4)a(0)cri2cos2 (~+~cri) -(l/4)b(o)~i2cos(l~+~cri)sin(~+~cri) ] 114 m B CQ(t)UI'(t)~[(l/4)a(0)c~i2cos(~+~ocri)sin(~+c)cri)-i-0 (l/4)b(0)c~i2sin2(~+~cri)] (15) m C-cI(t)uQ~ (t)-~ [ (l/4)a(o)~yi C05(~ c i) (1/4)b(0)c~ 2cos2(~+~cr )] (16) m D--CQ(t)UQ' (t)~ [ (1/4)a(O)~i2sin2 (y+~-)cri)+
(l/4)b(0)~i2cos( r+~cri)sin(~+~cri) ] (17) ~Z~41~73 Properly combining the quantities A khrough D, one obtains the following expressions for the ln-phase and 05 quadrature outputs o~ the receivar at time t=0:
m I A+D~(1/4)a(0)~ 2 (Co52 { y+~ocr 3+sin2 {'y+l~Cr, ) ) i~O
m -~(1/4)a(0)~12 ~ In-phase data (18) i~O
m Q-C - B-~ ( 1/4 ) b ( O) ~i 2 ( cos 2 { ~+~cri ) +s in2 ( ~tocr m ~(l/4)b(0)~i2 ~ Quadrature-phase data. (19 i=O
Conventional adder 331 implements equation (18~ to produce the recovered in-phase signal I and adder 335 implement~ equatlon (19) to produce the recovered quadrature signal Q, which are replicas of the : transmitted I and Q channel data, respectively. The outputs I and Q are actually four sequential numbers per chip interval. It is pos~ible to integrate them and apply a s~mpla threshold ~or a binary data stream, or to simply integrate them to provide relative weighting, both at the original rate, or to preserve their discrete ,~
~9~3 sample form for use in somewhat more elaborate symbol or character correlation.
The outputs I and Q from the adders 331 and 335 may subsequently be applied to a data signal recovery circuit such as the conventional baseband synchronous feedback equalizer described in the aforementioned PCT
International Publication No. W0 88/05981.
It can be seen by following the general input equation (3) through to equations (18) and (19) that the information contained in each of the paths of the multipath signal is coherently combined in the receiver, thereby permitting an ef~ective time diversity gain in the receiver. Furthermore/ since the sin and cos terms of equations (14) through (17) are cancelled algebraically and trigonometrically, the received signal phase offset / is cancelled.
In the preferred embodiment four M/PATH correlators 312, 313, 315, and 317 operate on 128 samples, or 32 chip intervals so as to accommodate as much as an 8 microseconds variation in the path delays, any one with respect to the others. This also imposes the requirement that the acquisition sequence be of no less than 9 microseconds duration , preferably two to four times that long.
Referring to Fig. 5, a representative graph of the outputs CI(t) and Co(t) is shown on one axis with time on the other axis. The outputs of the synchronization correlators 303 and 305 have signed responses at each clock pulse but none of the responses exceed the established threshold magnitude until a correlation with the predetermined synchronization sequence aT(t) is realized. As shown, a correlation is found at time td.
~294~73 Referring now to Fig. 6, there is shown a block diagram of a QPSK tran~mitter which may be employed as a fixed ~it~ transmitter in a TDMA system. A ai~ilar transmitter may be employed as a tran~itter in a nobile 05 or portable transceiver. At th~ fixed site, digital speech vr data are input ~rom a number of user~ and formatted by ~ormatter 601 into individual TDMA message plu8 synchronization str~am A predetermined number of data bits from one user are interleaved with the ynchronlzation seguence by means o~ ~oftware control in a microproce~or. A repre~ntativ~ ~or~atter is ~urther : described in PCT publication No. wo 88/05981 published on 11 August 1988.
The TDMA controll~r 603 perform~ the function of time-multiplexing each user message with the other us~r messages to for~ a TDMA signal. The time-multiplexing fu~ction may, preferably, be performed in a software-mediated proces~ of a microproces~or but ~ay al~o be r~alized via a time-controlled switch (as d~scrlbed in the data ~heet of a ~otorola ~C14416 Time Slot A ~igner).
The output of the TDMA controller 603 i~ a aignal consi~ting of N user message~ (as shown in Fig. 7) and is ~: applied, a~ quadrature signal~ a(t) and b(t), to 25 conventional modulators 605 and 607. Modulator 605 modulates the output of ~ignal o~cillator 609 (co (~ct}) with the a(t) signal; modulator 607 modulat~s the 90 phase ~hifted (from phase shlfter : 611) output of signal oscillator 609 with b(t) to create ths b(t)sin(~ct) quadrature ~ignal. The two signals are then summed in conventional summer 613 and applied to the kransmitter ampli~ication stage 615 for amplification prior to transmission ~as x{t~).
; 35 ,....
.1 ~L2~4~
During the period the synchronization sequence is being transmitted for each timeslot, the present invention transmits the synchronization sequence only on one vector of the QPSg modulated signal. In the 05 preferred embodiment, this i8 accomplished by preventing the b(t)sin(~ct) from being summed in summer 613. (~he transmission of b(t)sin(~ct) and the suppresion of a(t)cos(~ct) would work equally well and is the choice of the system designer. Furthermore, it may be desirable to transmit the acquisition sequence at some other angle relatiYe to I and Q, for example, simultaneously and identically in both I and Q ~or a 45 shift. Any angle can be accommodated by operating on the multipath profile estimates CI(t)and CQ(t) when applying them to M/PATH
correlators)-A co~ventional sig~al switch 617 open-circuits the ; coupling between the modulator 607 and the summer 613 upon a com~and ~rom the ~DMA controller 603 indicating a synchronization ~equence period. This signal switch 617 reconnects the modulator 607 and the summer 613 during periods of message data transmission.
Since the turning of~ of b(t)sint~ct) during the acquisition synchr~nization sequence e~fectively reduces the transmitter output power by 3dB during the synchronization sequence, the transmitter may optionally be equipped with the capability of increasing the power gain of the trans~itter ampli~ication tage 615. The amplification stage 615, which may be a conventional variable output power amplifier, is switched during the synchronization sequence to an output power 3dB greater than the output power during the transmission of the message data of each timeslot. In this way, the system gain is maintained during both synchronization sequence and message data transmission.
In summary, then, the present invention describes a unique phase coherent method for transmitting and receiving a QPSK radio signal that ha~ been subject to a multipath ~ading radio channel. In order that the 05 equalization for reception of a radio ~ignal subject to Rayleigh and multipath fading be adapted for the channel, a training or synchronization signal is transmitted on only one of the vectors of a quadrature phase modulated signal. The random amplitudes and phases of copies of the modulated signal added to the signal by channel multipath are correlated and combined in accordance with a multipath profile ignal developed from the : synchronization signal. Therefore, while a particular embodiment of the invention has been shown and described, it should be understood that the invention i8 not limited thereto since modification~ unrelated to the true s~irit and scope of the invention may be made by those skilled in the art. It is therefore contemplated to cover the present invention and any and all such modifications by ~he claims of the present invention.
We Claim:
:
,~ . ...
4Q~3 TDMA RADIO SYSTEM EMPLOYING BPSK SYNCHRONIZATION
FOR QPSK SI&NALS SUBJECT TO RANDOM PHASE
VARIATION AND MULTIPATH FADING
5Backqround of the Invention This invention relates generally to digital radio systems and more specifically to a system transmitting, receiving, and demodulating TDMA QPSK modulation in a multipath fading environment such as an environment where 10the transceivers may be in motion and are served from fixed site transmitters and receivers such as in a radiotelephone system. This invention is related to instant assignee's Canadian Patent Applications 576,174 "Rapid Reference Acquisition and Phase Error Compensation ~15for Radio Transmission of Data", filed August 31, 1988 ; (the inventors being Labedz et al) and 576,640 "Phase-Coherent TDMA Quadrature Receiver for Multipath Fading Channels", filed on September 7, 1988 (the inventors being Borth et al) and containing related subject matterO
20In a typical environment, a UHF or microwave radio channel exhibits a multipath structure in addition to ~:; Rayleigh fading. Thus, a radio transceiver for a mobile or portable TDMA system operating at high data rates must accommodate reception of multiple replicas of the 25transmitted signal, each with a random magnitude, phase, '' ~
~' ~.~
, ~
: . ,,f ~I
:' ~IL2~40 73 and time delay with respect to the transmitted signal.
Without corrective measures, the data message can be obliterated by the multipath signals. As early as 1958, a receiver capable o~ accommodating these impairments was 05 described for the use o~ either Differential Binary Phase-Shift Keying (DBPSK) or noncoherent Fre~uency-Shift Keying (FSK). It employed a channel sounding method to estimate the channel impulse response or channel profile, and a transversal equalizer having taps which were adjusted in response to the estimated channel profile.
By 1960 the multipath channel had been exhaustively studied and simulated, and optimum reception had been de~ined, but largely avoiding phase coherent techni~ues.
Such demodulation techniques do not permit the use of higher spectral e~ficiency modulation methods which employ two-dimensional signaling techniques such as shaped Quadratuxe Phase-Shift Keying (QPSK) and its variations.
By 1983, TDMA (Time Division Multiple Access) receivers for digital telephony using Binary Phase-Shift Keying (BPSK) phase coherent detection had been described in U.S. Patent No. 4,587,662. In 1985 this was extended to include QPSK, but the receiver was only described in general terms.
In 1986 an MSK receiver, with possible application to QPSK, was reported which could accommodate two rays of multipath and which used an adaptive equalizer employing both ~eedforward and feedback filtering. (See Krister Raith et al., "Multi-Path Equalization for Digital 30 Cellular Radio Operating at 300 kbits/s", 36th IEEE
Vehicular ~onference, pp.268-272, May 1986). Although this adaptive e~ualizer apparently has never been thoroughly described in the literature, it is different ; than the multipath correlation employed in the present invention since it requires decisions to be made on the output in order to adjust the equalizer.
1294~)~3 Adaptive egualization generally operating continuously on the data being received has been utilized in digital microwave receivers receiving continuous data streams. Such continuous receivers can equallze over a 05 relatively long period o~ time. TD~A, due to its burst-like characteristics, demands rapid determination of the channel pr~file including the significant multipath contributions. Even though the data transceiver may be moving, the channel profile can be a~sumed to undergo negligible changes in a given TDMA timeslot if the timeslot is sufficiently short in duration.
Summary of the Invention Therefore, it is one ob~ect of the present invention to synchronize a TDMA receiver to a synchronization sequence transmitted during a timeslot to compensate for multipath signals and random phase introduced by a radio channel to transmitted digital information.
It is another object of the pre~ent invention to transmit the synchronization saquence on a single vector of the QPSK modulated signal.
It is a ~urther ob~ect of the present invention to create a radio channel multipath profile from the synchronization sequence and utilize that model during only on~ time~lot period.
Accordingly, these and other objects are achieved in the present invention which encompasses a time division multiple access (TDMA) radio system which conveys a data message using quadrature phase shift keying (QPSK) ~ modulation and a synchronization sequence in a timeslot ; over a radio channel. The transmitter transmits the synchronization seguence in only one vector of the QPSK
modulation and the receiver receives and employs the one vector synchronization to compensate the data message for multipath and phase vaxiation during one TDMA timeslot.
~2~0~73 Brief_Descri~ion of the Drawin~s Figure 1 is a block diagram of a data transmission system employing quadrature digital transmission and reception.
Figures 2A and 2B are, together, a block diagram of a TDMA receiver which may receive QPSK signals.
Figure 3 is a block diagram of a TDNA receiver signal processor which may advantageously employ the present invention.
Figure 4 is a block diagram of the peak detector circuit of the receiver of Fig. 3.
Figure 5~ is a graph of the synchronization correlator outputs CI(t) and CQ(t) plotted against time and showing a possible set of outputs including a correlati~n detection.
Figure 6 is a block diagram of a QPSK transmitter which transmits a synchronization sequence during a portion of a TDMA time slot in accordance with the present invention.
Figure 7 is an illustration of the TDMA signal format which may be transmitted from the transmitter of Fig. 6.
Description of the Preferred Embodiment A radio frequency system conveying a data signal from a transmitter 101 to a receiver 103 is shown in Fig.
1. In a radiotelephone system, transmitter 101 would be a fixed site transmitter serving a radio coverage area which would be populated by mobile or portable transceivers, the receiver of which is shown as receiver 103. Additionally, more than one radio coverage area can be linked in such a way that continuous coverage may be provided over a wide area, i.e., a cellular radio telephone system or an integrated area such as an office building or shopping mall. (See, for example, instant ... .
. .
1~4(~73 assignee ' s Canadian Patent Application No. 560, 416 "Microcellular Communication System Using Macrodiversity", filed March 3, 19~8, the inventor being Gerald P . Labedz ) .
In the preferred embodiment, quadrature phase shift keying (QPSK) is employed to increase the through put of the channel although other multi-dimensional signaling may equivalently be employed. Further, the well-known time division multiple access (TDMA) technique of sharing a limited channel resource among a large number of users is employed in the present invention. Each of the users is assigned a brief period of time (a timaslot) during which a message may be transmitted to or received from the user. The advantages of such a TDMA technique over other techniques (such as frequency division multiple access l'DMA) are: a) no duplexer is required for full duplex communications, b) variable data rate transmission may be accommodated through the use of multiple adj acent time slots, c) a common radio frequency power amplifier may be used to amplify multiple channels at any power level without the combining losses or intermodulation distortion present with FDMA, and d) a capability of scanning other "channels (timeslots) without requiring separate receivers may be providPd.
The high data rate employed in the present invention (200Kbps to 2Mbps) exceeds the channel coherence bandwidth of the mobile radio channel for many urban and suburban environments. As a result, the channel exhibits a multipath structure in addition to the expected Rayleigh fading. The receiver of the present invention enables TDMA quadrature signals to be coherently received over a multipath fading channel. This embodiment will demodulate a 2-megachip/sec QPSK radio signal, the only 3LZ~ 3 constraint being that the acquisition sequence be transmitted as a binary phase shi~t keying (BPSK) signal with a predetermined phase relative to the QPSK data.
Figs. 2A and 2B are a block diagram of a TDMA
quadrature phase shift keying data and is described in instant assignee's PCT International Publication W0 88/05981 "TDMA Communications System with Adaptive Equalization" published on August 11, 1988 and filed on behalf of David E. Borth.
The digital signal outputs of the A/D converters 209 and 211, respectively, are applied to in-phase (I) time slot correlator 213 and quadrature (Q) correlator 215, respectively, as well as to their respective signal buffers 217 and 219. I correlator 213 performs a correlation function between all received bits of the input signal and a pre-loaded synchronization word (I
sync word) corresponding to the in-phase time slot sync word.
The output of I correlator 213 is a digital bit stream representing the sample-by-sample correlation of the received data with the stored synchronization word replica for the timeslot. The correlation function exhibits a peak when the I sync word is located in the received sample data. In the same way, Q correlator 215 performs a correlation function between the pre-stored guadrature Q sync word from memory 221 and the sampled quadrature (Q) input.
The outputs of correlators 213 and 215 are applied to squaring blocks 223 and 225, respectively. I~he squaring blocX output signals represent the squared values of the separate I and Q correlation operations respectively. The squaring block outputs are then applied to summing block 227. The I and Q correlation .
- 6a -signals are summed together to form a squared envelope signal which represents the sum of squares of the correlation signal. The squared envelope of the correlation signal makes an explicit determination of the phase ambiguity unnecessary. Thus, without resolving any ~29~73 ambiguity, a large amplitude signal output from summing block 227 repres~nts a possible start location for a particular timeslot.
The output of summing block 227 is then routed to 05 time slot detector 229, wherein the summed correlation signal is compared with a predetermined threshold value.
This threshold value represents the minimum allowable correlation value which would represent a detected time-slot. If the summed output is greater than the threshold value, a time slot detect signal is generated and applied to system timing controller 231.
Timing controller 231 functions as a phase-locked loop (PLL), using a stable timing reference to validate the timeslot detect signal and provide a validated detect output signal. The validated timeslot detect signal is applied to AND gate 233 along with a bit clock output.
The combined timeslot detect/bit clock signal i8 then routed to the I and Q signal buffers 217 and 219, re~pectively. Data ~ignals are clocked lnto 6ignal buffers 217 and ~19 using the combined detect/bit clock signal.
In the implementation shown in Figs. 2A and 2B, a conventional baseband synchronous decision feedback equalizer (D~E) 234 is employed for data signal recovery.
The DFE 234 basically consists of two parts: a forward linear transversal filter 235 and a feedback linear transversal filter 237. The forward filter 235 attempts to minimize the mean-square-error (MSE~ due to intersymbol interference (ISI), while the feedback filter 237 attempts to remove the ISI due to previously detected symbols.
The decision feedback equalizer 234 structure is adapted at least once each time slot in order to compensate for the effects of the time-varying multipath profile. The equalized and quantized complex data output from quantizer 238 is applied to multiplexer 239 for 2:1 ~Z~ 3 multiplexing together with the data clock and output as an output data word.
Returning to Fig. 1, in a QPSK communication system, a transmitted signal x(t) may be expre sed as:
x(t)=a(t)cos~ct+b(t)sin~ct (1) ~ where a(t) and b(t) are the in-phase and quadrature : information signals and ~c is the carrier frequency of the QPSK signal in radians/sec.
~ A frequency-selective (or delay-spread) channel that : is, a radio channel subject ~o multipath inter~erence, may be ch~racterized by an equivalent channel impulse ; response given by:
h(t)~:xo6(t-ro)+~l6(t-r )+~ 6(t-r )~ . . .
m =~ Ixi6(t-ri)~ (2) i-0 : where ~i is the amplitude o~ the i-th resolvable path, ri is the (excess) path delay associated with the th resolvable path, and m+l is the total number of resolvable paths.
For a channel input given by eguation (1), the output of the equivalent delay-spread channel having the impulse response of equation (2) is essentially constant during any given timeslot, and given by.
m y(t)-x(t)*h(t)=~rx(r)h(t-r)dr=J`x(r)~ 6(t-r-ri)dr i=O
m , =~i J;~(r)~(t-r-ri)dr i~O - a~
~, m y~t)=~iX(t Ti) O
=~ i [a(t-r; )Cos~c(t-r~ b(t-ri )Sin~c(t~ri )] (3) It is this signal, ylt), which is input to rsceiver ;103. When the local oscillator reference 105 in the receiver has a phase of~set of y with respect to the (direct-path) received QPSX transmiscion, the receiver local oscillator reference may b2 given by cos (~ct+~) and is es~entially constant during a TDMA timeslot.
(Although the antenna is shown connected to the mixers 107 and 111, it is likely t~at additional signal processing will be required for higher frequ~ncy radio ignals. If down-conversion to an intermediate frequency is used, the output frequency of local oscillator may be different~. Let UI(t) denote the output of the mixer 107 2~ in the uncorrected in-phase branch of the receiver and lat UI'(t) denote the low-pa~ ~iltered version of UI(t)output from low pas~ filter 109. Similarly, let UQ(t) denote the output of the mixer 111 in the uncorrected quadratur~ phase branch of the receiver and ~25 let UQ'(t) denote the low-pas~ filtered version of UQ(t) -~from filter 113. UI'(t) and UQ'(t) ar~ subsequently input to signal processor 115 for resolution into I and Q
data and then cou'pled to data signal recovery 117.
UI(t) i9 give~ by:
c ~i~0 i [ ( i ) c ( i ) ( i ) in~c ( t ri ) ]
:~ m ~ i [(l/2)a(t-ri)(cos{~+ocr~)+Cost2~)ct~ cr ))+
3 5 i 0 1 i (1/2)b(t-ri ) (sint21~ct+~-~cri ) -sint~+~cri ))] . (4) .
The low-pass filtered ver~ion UI'(t) of UI(t) is given by:
m 05 UI (~ /2)~i[a(t-ri)CS(y+~cri)-b(t-ri)sin(~*~)cri)] (5) Similarly UQ(t) is given by:
m c Yi~ oi [ a ( t - ri ) Cos~c ( t - ri ) +b ( t - ri ~ s in~ ( t - r ) ]
m ='~ i [ (1/2)a(t-ri ) (sin~20ct+y-ocri ~+sin~y+~)cri })+
i--O
(1/2)b(t-ri ) (Cos~y+~ocri ) -Cos~2~ct+y-~cri ~) ] (6) and UQ'(t) i8 given by:
m i~ ( / ) i la(t-ri )9in(Y+~cri )+b(t-ri )Cos(y+locr ) ] (7) :~ 20 Con~idering the operation of the present invention `~ in mathematical form, it is an important feature that the : transmitted slgnal xT(t) duriny the synchronization (or ~: training~ phase of the equalizer 115 i9 a BPSK signal.
;~ ~
~: 30 ; 35 When kransmitted in the I phase it is given by:
XT(t)-aT(t)COs~Ct (a) 05 where signal aT(t) (not shown) is a predetermined synchronization sequence with good aperiodic autocorrelation properties, such as one of the Barker sequencss.
The uncorrected in-phase and guadrature receiver branch outputs corresponding to the synchronizing transmitted signal xT(t) may be found by substituting the signal of equation (8) in the received and low pass filtered signals UII(t) and UQ'(t) of equations (5) and (7) respectively, yielding:
1~
m ,~ UI'T(t)~(l/2)~i[aT(t-ri)CS(~+~c~i)] (9) i=0 .~
and, : 20 UQ~T(t) ~(l/2)c~i[aT(t-ri)sin(~+~cri)](10) Thus UI'(t) and UQ'(t) are defined during the : 25 training phase as IlT" as shown in eguations (9) and (10).
: Referring now to Figuxe 3 which illustrates the pre~erred embodiment of the present invention in block diagram ~orm, the signals UIIT(t) and UQ'T(t) are ~IL2~ 73 applied to synchronization correlators (303 and 305, respectively) via conventional fast A/D converters 307 and 309. In the preferred embodiment, synchronization correlators 303, 305 are 4 by 32 bit digital finite 05 impulse response (FIR) filters programmed to provide signed weighted correlation outputs. Synchronization correlator~ 303 and 305 are realized by an IMS A100 Cascadable Signal Processor available from Inmos Corp., Colorado Springs, Colorado. The outputs of aorrelators CI(t) and CQ(t) which are, in simple terms, weighting factors for each i-th resolvable path, generated during reception of the acquisition ~equence, may have the appearance as shown in Fig. 5 and are given by:
m CI(t)~ (l/2)'~iCOs( Y~>cri)~(t-ri) (11) and, Q(t)~ /2)~iSin(~+~Cri)~(t-ri)- (12) The ~ function in equations (11) and (12) determine when to sample the in-phase and quadrature receiver ; branch outputs and the ~ factor provides a weighting for each i-th resolvable pas~ contribution. In the preferred e~bodiment, a sequen¢e controller 311 is realized using a : conventional microprocessor (such as an ~C68020 microprocessor available from Motorola, Inc.3 and associated memory and timing dividers. The sequence 3L~94~73 controller 311 loads a predetermined normalized replica of the acquisition sequence (32 each 4-bit words) stored in the memory of sequence controller 311 into synchronization correlators 303 and 305 prior to the 05 desired TDMA timeslot to be demodulated. TDMA frame ; timing is determined by the sequence controller 311 employing a conventional framing al~orithm to confirm and maintain timeslot acquisitionO
Synchronization correlators 303 and 305 2ach correlate the stored acquisition ~eguence against the last 32 received A/D ~amples, and for each new sample perform another complete correlation. While receiving noise or random data, the outputs CI(t) and CQ(t) of synchronization correlators 303 and 305 ars small numbers of either polarity, emerging at the same rate as the A/D
sampling rate (4 per chip interval). If the radio channel were free of noise and not degraded by multipath, then when an acquisition sequence has been received and digitized and entered into the correlator~ 303 and 305, CI(t) and CQ(t) would simultaneously manifest a pair (or sometimes two ad~acent pairs) of signed numbers signi~icantly larger than those produced by noise or random data, ~uch that the root sum o~ squares of these numbers would be proportional to the ~agnitude of the received signal, and the phase angle ~ relative to the local re~erence oscillator is:
~ arc tan [CQ(t)/CI(t)]- (13) ::
,~
~L2~4C~3 - 14 - CE00027~
In the presence of multipath, each ~igni~icant path will result in the presence of ~uch a peak pair appearing on CI(t) and CQ(t), the signs and magnitudes of each pair of outputs at each peak defining the delay, phase 05 angle, and amplitude contribution of that path to the : total, fulfilling the equations (11) and (12). Thus, each sequence of number~ CI(t) and CQ(t) are bipolar multipath channel profile e~timates, which re~emble a cla~sic multipath channel profile, except that they are bipolar.
Each of the M/PATH correlators 312, 313, 315, and 317 are FIR filters of at least 32 taps. In the preferred embodiment, each ~/PATH correlator is realized : by an IMS A100 Cascadable Signal Proces~or (available : 15 from Inmos Corp., Colorado Springs, Colorado) conventionally connected as a correlator. During the acquisition sequence at the beginning of each desired timeslot, CI(t) i~ shifted into the TAP control entry :~ o~ ~/PATH correlators 312 and 317, and CQ(t) is shifted : 20 into the TAP con~rol entry of ~/PATH correlator~ 313 and 315. Peak detector 318 is shown in Fig. 4 and comprises a root 8um 0~ squares approximator 401 and a threshold detector 403 having an output which ~ignal~ the sequence controller 311 of the first signi~icant ray of multipath.
Th~ GequenCe controller 311 then provide~ just enough : additional reference port clocks to shift this peak all but through the M~PATH correlators, thereby capturiny CI(t) and CQ(t~ in their respective M/PATH
correlators. In the preferred embodiment, the root sum of squares approximator 401 is realized e~ploying a magnitude adder 405 which adds lCI(t) I and (1/2)lCQ(t)l and magnitude adder 407 which adds ICQ(t) I and (1/2) ICI(t) I . The outputs of magnitude adder 405 and magnitude adder 407 are input to 35 conventional magnitude comparators 409 and 411, respectively, where the root sum of squares approximation ~ z9~ 3 is compared to a predetermined threshold to generate an output to the ~equence controller 311 (via OR gate 413).
Thi~ and other approximations to the square root of the ~um of the squares may be found in, eg., A.E. Filip, "A
05 Baker ~ 8 Dozen ~agnitude Approximations and Their Detection Stati~tics," IEEE Transaction~ on Aerospace and Electronic Syste~s, vol. AES-12, pp.86-89, January 1976.
Thi~ output to the ~eguence controller 311 i~ ~hown as td in the example of Fig. 5. Thus, the peak detector 318 reports the first eignificant peak to the ~equence controller 311 which, in turn, ~tarts the loading at ~/S
~top, to thus capture the channel profile in each of the M/PATH correlator~.
The four M/PATH correlators (312, 313, 315, and 317 in Fig. 3) thus have the information available to perform equations (14)-(17~, below, whose results (A, B, C, and D) appear at the outputs of M/PATH correlator~ 312, 313, 315, and 317 respectively.
20m A-CI (t)UI ' (t)~ [ (l/4)a(0)cri2cos2 (~+~cri) -(l/4)b(o)~i2cos(l~+~cri)sin(~+~cri) ] 114 m B CQ(t)UI'(t)~[(l/4)a(0)c~i2cos(~+~ocri)sin(~+c)cri)-i-0 (l/4)b(0)c~i2sin2(~+~cri)] (15) m C-cI(t)uQ~ (t)-~ [ (l/4)a(o)~yi C05(~ c i) (1/4)b(0)c~ 2cos2(~+~cr )] (16) m D--CQ(t)UQ' (t)~ [ (1/4)a(O)~i2sin2 (y+~-)cri)+
(l/4)b(0)~i2cos( r+~cri)sin(~+~cri) ] (17) ~Z~41~73 Properly combining the quantities A khrough D, one obtains the following expressions for the ln-phase and 05 quadrature outputs o~ the receivar at time t=0:
m I A+D~(1/4)a(0)~ 2 (Co52 { y+~ocr 3+sin2 {'y+l~Cr, ) ) i~O
m -~(1/4)a(0)~12 ~ In-phase data (18) i~O
m Q-C - B-~ ( 1/4 ) b ( O) ~i 2 ( cos 2 { ~+~cri ) +s in2 ( ~tocr m ~(l/4)b(0)~i2 ~ Quadrature-phase data. (19 i=O
Conventional adder 331 implements equation (18~ to produce the recovered in-phase signal I and adder 335 implement~ equatlon (19) to produce the recovered quadrature signal Q, which are replicas of the : transmitted I and Q channel data, respectively. The outputs I and Q are actually four sequential numbers per chip interval. It is pos~ible to integrate them and apply a s~mpla threshold ~or a binary data stream, or to simply integrate them to provide relative weighting, both at the original rate, or to preserve their discrete ,~
~9~3 sample form for use in somewhat more elaborate symbol or character correlation.
The outputs I and Q from the adders 331 and 335 may subsequently be applied to a data signal recovery circuit such as the conventional baseband synchronous feedback equalizer described in the aforementioned PCT
International Publication No. W0 88/05981.
It can be seen by following the general input equation (3) through to equations (18) and (19) that the information contained in each of the paths of the multipath signal is coherently combined in the receiver, thereby permitting an ef~ective time diversity gain in the receiver. Furthermore/ since the sin and cos terms of equations (14) through (17) are cancelled algebraically and trigonometrically, the received signal phase offset / is cancelled.
In the preferred embodiment four M/PATH correlators 312, 313, 315, and 317 operate on 128 samples, or 32 chip intervals so as to accommodate as much as an 8 microseconds variation in the path delays, any one with respect to the others. This also imposes the requirement that the acquisition sequence be of no less than 9 microseconds duration , preferably two to four times that long.
Referring to Fig. 5, a representative graph of the outputs CI(t) and Co(t) is shown on one axis with time on the other axis. The outputs of the synchronization correlators 303 and 305 have signed responses at each clock pulse but none of the responses exceed the established threshold magnitude until a correlation with the predetermined synchronization sequence aT(t) is realized. As shown, a correlation is found at time td.
~294~73 Referring now to Fig. 6, there is shown a block diagram of a QPSK tran~mitter which may be employed as a fixed ~it~ transmitter in a TDMA system. A ai~ilar transmitter may be employed as a tran~itter in a nobile 05 or portable transceiver. At th~ fixed site, digital speech vr data are input ~rom a number of user~ and formatted by ~ormatter 601 into individual TDMA message plu8 synchronization str~am A predetermined number of data bits from one user are interleaved with the ynchronlzation seguence by means o~ ~oftware control in a microproce~or. A repre~ntativ~ ~or~atter is ~urther : described in PCT publication No. wo 88/05981 published on 11 August 1988.
The TDMA controll~r 603 perform~ the function of time-multiplexing each user message with the other us~r messages to for~ a TDMA signal. The time-multiplexing fu~ction may, preferably, be performed in a software-mediated proces~ of a microproces~or but ~ay al~o be r~alized via a time-controlled switch (as d~scrlbed in the data ~heet of a ~otorola ~C14416 Time Slot A ~igner).
The output of the TDMA controller 603 i~ a aignal consi~ting of N user message~ (as shown in Fig. 7) and is ~: applied, a~ quadrature signal~ a(t) and b(t), to 25 conventional modulators 605 and 607. Modulator 605 modulates the output of ~ignal o~cillator 609 (co (~ct}) with the a(t) signal; modulator 607 modulat~s the 90 phase ~hifted (from phase shlfter : 611) output of signal oscillator 609 with b(t) to create ths b(t)sin(~ct) quadrature ~ignal. The two signals are then summed in conventional summer 613 and applied to the kransmitter ampli~ication stage 615 for amplification prior to transmission ~as x{t~).
; 35 ,....
.1 ~L2~4~
During the period the synchronization sequence is being transmitted for each timeslot, the present invention transmits the synchronization sequence only on one vector of the QPSg modulated signal. In the 05 preferred embodiment, this i8 accomplished by preventing the b(t)sin(~ct) from being summed in summer 613. (~he transmission of b(t)sin(~ct) and the suppresion of a(t)cos(~ct) would work equally well and is the choice of the system designer. Furthermore, it may be desirable to transmit the acquisition sequence at some other angle relatiYe to I and Q, for example, simultaneously and identically in both I and Q ~or a 45 shift. Any angle can be accommodated by operating on the multipath profile estimates CI(t)and CQ(t) when applying them to M/PATH
correlators)-A co~ventional sig~al switch 617 open-circuits the ; coupling between the modulator 607 and the summer 613 upon a com~and ~rom the ~DMA controller 603 indicating a synchronization ~equence period. This signal switch 617 reconnects the modulator 607 and the summer 613 during periods of message data transmission.
Since the turning of~ of b(t)sint~ct) during the acquisition synchr~nization sequence e~fectively reduces the transmitter output power by 3dB during the synchronization sequence, the transmitter may optionally be equipped with the capability of increasing the power gain of the trans~itter ampli~ication tage 615. The amplification stage 615, which may be a conventional variable output power amplifier, is switched during the synchronization sequence to an output power 3dB greater than the output power during the transmission of the message data of each timeslot. In this way, the system gain is maintained during both synchronization sequence and message data transmission.
In summary, then, the present invention describes a unique phase coherent method for transmitting and receiving a QPSK radio signal that ha~ been subject to a multipath ~ading radio channel. In order that the 05 equalization for reception of a radio ~ignal subject to Rayleigh and multipath fading be adapted for the channel, a training or synchronization signal is transmitted on only one of the vectors of a quadrature phase modulated signal. The random amplitudes and phases of copies of the modulated signal added to the signal by channel multipath are correlated and combined in accordance with a multipath profile ignal developed from the : synchronization signal. Therefore, while a particular embodiment of the invention has been shown and described, it should be understood that the invention i8 not limited thereto since modification~ unrelated to the true s~irit and scope of the invention may be made by those skilled in the art. It is therefore contemplated to cover the present invention and any and all such modifications by ~he claims of the present invention.
We Claim:
:
,~ . ...
Claims (13)
1. A quadrature phase shift keying (QPSK) time division multiple access (TDMA) radio system conveying a data signal comprising a synchronizing portion and a message portion in a timeslot over a radio channel which introduces multipath distortion and phase variation in the data signal, the system comprising:
means for transmitting the synchronizing portion consisting of a plurality of data bits in a predetermined pattern in only one vector of the data signal;
means for receiving the data signal subjected to multipath distortion and phase a variation; and means responsive to said one vector synchronizing portion of a received data siganl, for compensating the multipath distortion and phase variation of the message portion of said received data signal for the duration of one TDMA timeslot.
means for transmitting the synchronizing portion consisting of a plurality of data bits in a predetermined pattern in only one vector of the data signal;
means for receiving the data signal subjected to multipath distortion and phase a variation; and means responsive to said one vector synchronizing portion of a received data siganl, for compensating the multipath distortion and phase variation of the message portion of said received data signal for the duration of one TDMA timeslot.
2. A quadrature phase shift keying (QPSK) time division multiple access (TDMA) radio system in accordance with claim 1 wherein said mean for transmitting further comprises means for amplifying the data signal to an output level.
3. A quadrature phase shift keying (QPSK) time division multiple access (TDMA) radio system in accordance with claim 2 wherein said means for transmitting further comprises means for increasing said output level during the synchronizing portion of the data signal.
4. A method of conveying, by way of a quadrature phase shift keying (QPSK) time division multiple access (TDMA) radio system, a data signal comprising a synchronizing portion and a message portion in a timeslot over a radio channel which introduces multipath distortion and phase variation in the data signal, comprising the steps of:
means for transmitting a synchronizing portion consisting of a plurality of data bits in a predetermined pattern in only one vector of the data signal;
receiving the data signal subjected to multipath distortion and phase variation; and compensating, in response to said one vector synchronizing portion of a received data signal, the multipath distortion and phase variation of the message portion of said received data signal for the duration of one TDMA timeslot.
means for transmitting a synchronizing portion consisting of a plurality of data bits in a predetermined pattern in only one vector of the data signal;
receiving the data signal subjected to multipath distortion and phase variation; and compensating, in response to said one vector synchronizing portion of a received data signal, the multipath distortion and phase variation of the message portion of said received data signal for the duration of one TDMA timeslot.
5. A method in accordance with the method of claim 4 wherein said transmitting step further comprises the step of amplifying the data signal to an output level.
6. A method in accordance with the method of claim 5 wherein said transmitting step further comprises the step of increasing said output level during the synchronizing portion of the data signal.
7. A time division multiple access (TDMA) radio system having at least one radio coverage area served by fixed site transmitter conveying a quadrature phase shift keying (QPSK) modulated data signal comprising a synchronizing portion and a message portion in at least one TDMA timeslot over a radio channel which introduces multipath distortion and phase variation in the data signal to a receiver, the system comprising:
means for transmitting a synchronizing portion consisting of a plurality of data bits in a predetermined pattern in only one vector of the data signal;
means for receiving the data signal subjected to multipath distortion and phase variation; and means receiving to said one vector synchronizing portion of a received data signal, or compensating the multipath distortion and phase variation of the message portion of said received data signal for the duration of one TDMA timeslot.
means for transmitting a synchronizing portion consisting of a plurality of data bits in a predetermined pattern in only one vector of the data signal;
means for receiving the data signal subjected to multipath distortion and phase variation; and means receiving to said one vector synchronizing portion of a received data signal, or compensating the multipath distortion and phase variation of the message portion of said received data signal for the duration of one TDMA timeslot.
8. A quadrature phase shift keying (QPSK) time division multiple access (TDMA) radio system in accordance with claim 1 wherein said means for compensating the multipath distortion and phase variation further comprises means for coherently combining at least two rays of multipath signal.
9. A quadrature phase shift keying (QPSK) time division multiple access (TDMA) radio system in accordance with claim 1 wherein said means for compensating the multipath distortion and phase variation further comprises means for algebraically and trigonometrically cancelling the phase variation.
10. A method in accordance with the method of claim 4 wherein said step of compensating the multipath distortion and phase variation further comprises the step of coherently combining at least two rays of multipath signal.
11. A method in accordance with the method of claim 4 wherein said step of compensating the multipath distortion and phase variation further comprises the step of algebraically and trigonometrically cancelling the phase variation.
12. A time division multiple access (TDMA) radio system in accordance with claim 7 wherein said means for compensating the multipath distortion and phase variation further comprises means for coherently combining at least two rays of multipath signal.
13. A time division multipath access (TDMA) radio system in accordance with claim 7 wherein said means for compensating the multipath distortion and phase variation further comprises means for algebraically and trigonometrically cancelling the phase variation.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US128,677 | 1987-12-04 | ||
| US07/128,677 US4873683A (en) | 1987-12-04 | 1987-12-04 | TDMA radio system employing BPSK synchronization for QPSK signals subject to random phase variation and multipath fading |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1294073C true CA1294073C (en) | 1992-01-07 |
Family
ID=22436451
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA000575635A Expired - Lifetime CA1294073C (en) | 1987-12-04 | 1988-08-25 | Tdma radio system employing bpsk synchronization for qpsk signals subject to random phase variation and multipath fading |
Country Status (9)
| Country | Link |
|---|---|
| US (1) | US4873683A (en) |
| EP (1) | EP0318686B1 (en) |
| JP (1) | JP2990694B2 (en) |
| AT (1) | ATE99475T1 (en) |
| CA (1) | CA1294073C (en) |
| DE (1) | DE3886712T2 (en) |
| ES (1) | ES2047529T3 (en) |
| FI (1) | FI87712C (en) |
| HK (1) | HK19497A (en) |
Families Citing this family (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5245611A (en) * | 1991-05-31 | 1993-09-14 | Motorola, Inc. | Method and apparatus for providing carrier frequency offset compensation in a tdma communication system |
| CA2076099A1 (en) * | 1991-09-03 | 1993-03-04 | Howard Leroy Lester | Automatic simulcast alignment |
| SE469678B (en) * | 1992-01-13 | 1993-08-16 | Ericsson Telefon Ab L M | SET FOR SYNCHRONIZATION AND CHANNEL TESTING IN TDMA RADIO SYSTEM |
| DE69332139T2 (en) * | 1992-04-24 | 2003-03-27 | Oki Electric Ind Co Ltd | Receiver for digital messaging systems |
| FI92357C (en) * | 1992-10-12 | 1994-10-25 | Nokia Mobile Phones Ltd | Channel equalizer for a bidirectional telecommunication system |
| GB2272133B (en) * | 1992-11-02 | 1996-06-12 | Motorola Inc | Radio system |
| US5677934A (en) * | 1992-12-30 | 1997-10-14 | Nokia Mobile Phones Limited | Multipath propagation compensation in a TDMA system |
| FI94818C (en) * | 1993-06-02 | 1995-10-25 | Nokia Telecommunications Oy | A method for demodulating a digitally modulated signal and a demodulator |
| JP3160453B2 (en) * | 1993-11-29 | 2001-04-25 | 松下電器産業株式会社 | Data receiving device |
| US5479453A (en) * | 1993-12-29 | 1995-12-26 | Tsci Corporation | Wireless telecommunication digital receiver |
| US5461646A (en) * | 1993-12-29 | 1995-10-24 | Tcsi Corporation | Synchronization apparatus for a diversity receiver |
| DE69433267T2 (en) * | 1993-12-29 | 2004-07-22 | Atmel Corp., San Jose | DIGITALLY SYNCHRONIZED DIVERSITY RADIO RECEIVER |
| US5533063A (en) * | 1994-01-31 | 1996-07-02 | The Regents Of The University Of California | Method and apparatus for multipath channel shaping |
| US6195399B1 (en) * | 1994-03-28 | 2001-02-27 | Ericsson Inc. | Method and apparatus for converting a wideband if signal to a complex (quadrature) baseband signal |
| WO1995034140A1 (en) * | 1994-06-03 | 1995-12-14 | Motorola, Inc. | Method and apparatus for cancelling interference in signals having undergone multipath |
| US5881100A (en) * | 1994-09-09 | 1999-03-09 | Omnipoint Corporation | Method and apparatus for coherent correlation of a spread spectrum signal |
| US5754584A (en) * | 1994-09-09 | 1998-05-19 | Omnipoint Corporation | Non-coherent spread-spectrum continuous-phase modulation communication system |
| US5963586A (en) * | 1994-09-09 | 1999-10-05 | Omnipoint Corporation | Method and apparatus for parallel noncoherent correlation of a spread spectrum signal |
| US5757847A (en) * | 1994-09-09 | 1998-05-26 | Omnipoint Corporation | Method and apparatus for decoding a phase encoded signal |
| US5754585A (en) * | 1994-09-09 | 1998-05-19 | Omnipoint Corporation | Method and apparatus for serial noncoherent correlation of a spread spectrum signal |
| JP3581448B2 (en) * | 1994-10-21 | 2004-10-27 | キヤノン株式会社 | Spread spectrum communication equipment |
| US5621766A (en) * | 1994-10-31 | 1997-04-15 | Motorola, Inc. | Method and apparatus for burst detecting |
| US6128331A (en) | 1994-11-07 | 2000-10-03 | Cisco Systems, Inc. | Correlation system for use in wireless direct sequence spread spectrum systems |
| US5742907A (en) * | 1995-07-19 | 1998-04-21 | Ericsson Inc. | Automatic clear voice and land-line backup alignment for simulcast system |
| US6055280A (en) * | 1997-06-27 | 2000-04-25 | Raytheon Company | High data rate digital demodulator and bit synchronizer |
| DE69838759T2 (en) * | 1997-12-29 | 2008-10-30 | Kabushiki Kaisha Kenwood | CIRCUIT FOR CATCHING A FRAME SYNCHRONIZER SIGNAL IN A RECEIVER |
| US6560652B1 (en) * | 1998-11-20 | 2003-05-06 | Legerity, Inc. | Method and apparatus for accessing variable sized blocks of data |
| FR2808951B1 (en) * | 2000-05-10 | 2002-09-27 | Jean Pierre Dehaene | 4-PHASE MODULATION-DEMODULATION METHOD USING THE PUNCHING OF CERTAIN SYMBOLS TO IMPROVE ACQUISITION OF LOW SIGNAL-TO-NOISE PHASE-FREQUENCY SYNCHRONIZATION |
| US6912249B2 (en) * | 2001-03-15 | 2005-06-28 | Telefonaktiebolaget L M Ericsson (Publ) | Method and system for training a radio receiver |
| US7603081B2 (en) * | 2001-09-14 | 2009-10-13 | Atc Technologies, Llc | Radiotelephones and operating methods that use a single radio frequency chain and a single baseband processor for space-based and terrestrial communications |
| US6973296B2 (en) * | 2001-12-04 | 2005-12-06 | Intersil Americas Inc. | Soft decision gain compensation for receive filter attenuation |
| FR2855684B1 (en) * | 2003-05-26 | 2005-07-01 | Commissariat Energie Atomique | ULTRA LARGE BAND SIGNAL RECEIVER AND ASSOCIATED RECEIVING METHOD. |
| KR100594146B1 (en) * | 2004-02-11 | 2006-06-28 | 삼성전자주식회사 | Apparatus and method for initial frequency offset estimation in asynchronous mobile communication system |
| US7627004B2 (en) * | 2005-05-26 | 2009-12-01 | Cisco Technology, Inc. | Transporting synchronization channel information across a packet network |
| FI20065534A0 (en) * | 2006-08-28 | 2006-08-28 | Nokia Corp | synchronization |
| JP5217999B2 (en) * | 2008-12-12 | 2013-06-19 | 沖電気工業株式会社 | Propagation delay time adjustment method, propagation delay time adjustment system, propagation delay time adjustment device, propagation delay time adjustment program, and node device |
Family Cites Families (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3537008A (en) * | 1967-05-09 | 1970-10-27 | Trw Inc | Communications system incorporating means for combatting multipath interference |
| US3617889A (en) * | 1969-08-13 | 1971-11-02 | Rca Corp | Time-frequency-phase in-band coded communications system |
| JPS5855709B2 (en) * | 1977-05-07 | 1983-12-10 | 三菱電機株式会社 | phase demodulator |
| FR2428946A1 (en) * | 1978-06-13 | 1980-01-11 | Ibm France | METHOD AND DEVICE FOR INITIALIZING AN ADAPTIVE EQUALIZER FROM AN UNKNOWN DATA SIGNAL IN A TRANSMISSION SYSTEM USING QUADRATURE AMPLITUDE MODULATION |
| US4267591A (en) * | 1979-04-17 | 1981-05-12 | Cincinnati Electronics Corporation | QPSK Suppressed carrier with rotating reference phase |
| US4349915A (en) * | 1981-02-02 | 1982-09-14 | General Electric Company | Minimization of multipath and doppler effects in radiant energy communication systems |
| DE3121146A1 (en) * | 1981-05-27 | 1983-01-05 | Siemens AG, 1000 Berlin und 8000 München | DIGITAL RADIO SYSTEM |
| DE3302828A1 (en) * | 1983-01-28 | 1984-08-02 | Standard Elektrik Lorenz Ag, 7000 Stuttgart | RECEIVER |
| US4597089A (en) * | 1983-07-11 | 1986-06-24 | Hycom Incorporated | Single pulse fast learn modem |
| US4672638A (en) * | 1983-10-24 | 1987-06-09 | Nec Corporation | Multipath canceller for cancelling a distortion caused to a radio frequency pulse by multipath transmission |
| US4606051A (en) * | 1983-11-10 | 1986-08-12 | Universal Data Systems, Inc. | QPSK demodulator with I and Q post-detection data correction |
| US4599732A (en) * | 1984-04-17 | 1986-07-08 | Harris Corporation | Technique for acquiring timing and frequency synchronization for modem utilizing known (non-data) symbols as part of their normal transmitted data format |
| US4707841A (en) * | 1984-08-21 | 1987-11-17 | University Of Toronto, Innovations Foundation | Digital data receiver for preamble free data transmission |
| FR2571566B1 (en) * | 1984-10-09 | 1987-01-23 | Labo Electronique Physique | DIGITAL DATA RECEIVING DEVICE COMPRISING AN ADAPTIVE RHYTHM RECOVERY DEVICE |
| DE3444449A1 (en) * | 1984-12-06 | 1986-06-12 | Robert Bosch Gmbh, 7000 Stuttgart | DEMODULATOR FOR DIGITAL RECEIVERS |
| US4669091A (en) * | 1986-02-10 | 1987-05-26 | Rca Corporation | Adaptive multipath distortion equalizer |
| SE457399B (en) * | 1987-04-23 | 1988-12-19 | Ericsson Telefon Ab L M | DEVICE IN A COHERENT MOBILE PHONE RECEIVER FOR REDUCING BIT ERRORS |
-
1987
- 1987-12-04 US US07/128,677 patent/US4873683A/en not_active Expired - Lifetime
-
1988
- 1988-08-25 CA CA000575635A patent/CA1294073C/en not_active Expired - Lifetime
- 1988-10-17 ES ES88117271T patent/ES2047529T3/en not_active Expired - Lifetime
- 1988-10-17 EP EP88117271A patent/EP0318686B1/en not_active Expired - Lifetime
- 1988-10-17 AT AT88117271T patent/ATE99475T1/en not_active IP Right Cessation
- 1988-10-17 DE DE88117271T patent/DE3886712T2/en not_active Expired - Lifetime
- 1988-10-26 FI FI884939A patent/FI87712C/en not_active IP Right Cessation
- 1988-12-02 JP JP63305900A patent/JP2990694B2/en not_active Expired - Lifetime
-
1997
- 1997-02-20 HK HK19497A patent/HK19497A/en not_active IP Right Cessation
Also Published As
| Publication number | Publication date |
|---|---|
| DE3886712T2 (en) | 1994-04-28 |
| US4873683A (en) | 1989-10-10 |
| JP2990694B2 (en) | 1999-12-13 |
| JPH01190152A (en) | 1989-07-31 |
| ATE99475T1 (en) | 1994-01-15 |
| DE3886712D1 (en) | 1994-02-10 |
| EP0318686A3 (en) | 1990-08-29 |
| FI884939A (en) | 1989-06-05 |
| EP0318686A2 (en) | 1989-06-07 |
| HK19497A (en) | 1997-02-20 |
| ES2047529T3 (en) | 1994-03-01 |
| FI87712B (en) | 1992-10-30 |
| FI87712C (en) | 1993-02-10 |
| EP0318686B1 (en) | 1993-12-29 |
| FI884939A0 (en) | 1988-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA1294073C (en) | Tdma radio system employing bpsk synchronization for qpsk signals subject to random phase variation and multipath fading | |
| EP0318685B1 (en) | Phase coherent TDMA quadrature receiver for multipath fading channels | |
| EP0318684B1 (en) | Rapid reference acquisition and phase error compensation for radio transmission of data | |
| USRE42538E1 (en) | System and method for signal synchronization in a communication system | |
| US5428647A (en) | Method and apparatus for synchronizing a received signal in a digital radio communication system | |
| US5408504A (en) | Symbol and frame synchronization in a TDMA system | |
| US5297162A (en) | System and method for bit timing synchronization in an adaptive direct sequence CDMA communication system | |
| CA2071869C (en) | Carrier recovery method and apparatus having an adjustable response time determined by carrier signal parameters | |
| US5732111A (en) | Frequency error compensation for direct sequence spread spectrum systems | |
| KR0140361B1 (en) | Apparatus & method for recovering a time-varying signal using multiple sampling points | |
| US7486748B2 (en) | Method and system for signal quality measurement based on mean phase error magnitude of a signal | |
| US5259005A (en) | Apparatus for and method of synchronizing a clock signal | |
| US6222874B1 (en) | Frequency acquisition method for direct sequence spread spectrum systems | |
| JP3467165B2 (en) | Time division multiple access (TDMA) communication system with adaptive equalizer control function | |
| JP3251432B2 (en) | Apparatus and method for determining frequency and time slot position in digital wireless communication | |
| WO1999021323A2 (en) | Wireless multimedia carrier system | |
| US5787128A (en) | Nonlinear signal correlator and method therefor | |
| US6917657B2 (en) | Reduced MIPS pulse shaping filter | |
| EP0737380B1 (en) | A wireless digital synchronized diversity receiver | |
| JP2000307493A (en) | Carrier control system | |
| WO2003084080A2 (en) | System and method for signal time alignment using programmable matched filter coefficients |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MKEX | Expiry |