Frequency and Timebase Acquisition in a Radio Communications System
The present invention relates to a method of, and an apparatus for, frequency and timebase acquisition in a radio communications system, and in particular to such a method and apparatus suitable for use in a TDMA (Time Division Multiple Access) radio system, such as the TETRA (TErrestrial Trunked RAdio) system.
In mobile radio systems, the mobile radio stations must be able to synchronise their own frequency and timing to the frequency and timing of the signal transmitted by the base station to allow them to decode the received signals correctly and/or make transmissions correctly to the base station.
In practice there may initially be frequency and timing offsets or errors between the mobile stations and the base station e.g due to differences in their local frequency references and other offsets due to motion (e.g. Doppler Shift) . The mobile station must therefore be able to determine the frequency and timing of the base station signals, and adjust or tune its own local frequency and timing references accordingly. When the mobile station's frequency and timing are tuned sufficiently accurately for it to receive and transmit signals, it is often said to be "in lock".
Another scenario where a mobile station must tune to a received reference frequency and timing is when mobile stations are in direct radio communication without the presence of a base station (e.g. Direct Mode in TETRA) . In this case one mobile station defines the frequency and timing reference and the other mobile station tunes to that reference.
In a TDMA radio system, the mobile station usually receives signals in discrete bursts (e.g. timeslots or
groups of timeslots, usually spaced in time), and each individual burst presents an opportunity for the mobile station to determine and then tune to the received signal frequency and timing. Typically the mobile station will attempt to tune its local frequency and timing references by estimating its frequency and timing offsets or errors relative to the received signal and use those estimates to adjust its local frequency and timing references accordingly. This means that at the next received burst the mobile station should be better tuned to the received signal and therefore ideally able to decode it. However, the mobile station is not typically able to decode the first received burst, since its local frequency and timing references have not been adjusted at that stage. It must therefore wait for the next appropriate burst before it can possibly decode useful information from the transmitter, which may be several tens or hundreds of milliseconds later. According to a first aspect of the present invention, there is provided a method of receiving a radio signal in a radio receiver having local frequency and timebase references comprising: receiving a transmitted radio signal burst; estimating the frequency and/or timebase offset between the received signal and the radio receiver's local frequency and timebase references; and adjusting the received signal burst in frequency and/or timing position on the basis of the estimated frequency and/or timebase offset, respectively. According to a second aspect of the present invention, there is provided a radio receiver, comprising: a local frequency reference; a local timebase reference; means for receiving a transmitted radio signal burst ;
means for estimating the frequency and/or timebase offset between the received signal and the radio receiver's local frequency and timebase references; and means for adjusting the received signal burst in frequency and/or timing position on the basis of the estimated frequency and/or timebase offset, respectively.
In the present invention the determined timing and/or frequency offsets for a received signal burst, are used to adjust the frequency and/or timing of the received signal burst from which the offsets are determined, rather than simply being used to apply a correction to the receiver's (e.g. mobile station's) local frequency and timebase references. By making such an adjustment to the received signal burst, subsequent signal processing steps in the receiver can be carried out using a "corrected" version of the received signal burst and so the chances of the receiver being able to correctly decode the received signal burst are increased, such that the receiver may not have to wait for the next burst before it can decode -the received signal correctly. Since each received burst will usually contain useful information, it is advantageous to be able to decode bursts straight away, rather than having to wait for the next burst. This can therefore permit faster and improved readiness for operation and faster call set-up for a radio receiver in a mobile radio communications system.
The frequency and timebase offsets estimated are preferably also used to adjust the local frequency and timebase references of the radio receiver, as is known in the art, so as to tune and lock those references to the received signal.
The frequency and/or timing of at least the initial received burst or bursts should be adjusted until the receiver's local references are in lock (whereafter the adjustment may not be needed) . Preferably, however,
estimates of the offsets for each received burst are made, and each received burst is adjusted appropriately, even after the receiver is in lock.
The frequency and timebase offsets can be estimated and adjusted for in any way known in the art.
In a preferred embodiment coarse frequency and timing offsets are estimated from a predetermined portion of the received signal burst from which appropriate frequency and timing information can be derived, such as the frequency correction field in
TETRA. These determined offsets are then used to adjust the received signal burst in frequency and timing to remove the estimated offsets.
This coarse offset estimation and correction is preferably no longer performed once the receiver's local frequency and timebase references are sufficiently close to the received signal timing and frequency (e.g. are within a predetermined margin of these parameters), i.e. the receiver is "in lock", as it is then not entirely necessary.
Other, finer estimates of the frequency and timing offsets are preferably also made after the above coarse estimation, to further refine the adjustment and tuning process. An advantage of correcting the received signal burst as in this embodiment of the present invention is that the coarse offset is removed first, therefore allowing a further, finer estimate of the residual or remaining offset to be made. This means that more accurate adjustments to the receiver's local frequency and timebase references can be made over fewer received signal bursts, such that faster frequency and timebase lock to the received signal can be achieved.
Thus preferably a further timing offset estimation is made by estimating the position of symbols, or bits, etc in the received signal burst relative to their expected positions. This can be done by, for example, as is known in the art, searching the received signal
burst for, or correlating the received signal burst with, a predetermined sequence, such as a training sequence, over a window of time at plural predetermined time delays relative to the start of the window. This yields a finer timing offset that can be used to advance or retard the received signal burst by an amount appropriate to remove the offset .
A further finer timing offset can also be derived by looking at the phase of the received signal averaged over the received signal burst.
A finer frequency offset can be estimated (preferably after the coarse offset has been removed) by considering the ideal (i.e. transmitted) phases of the received symbols and the actual phases the symbols in the signal burst are received at, e.g. by comparing the phase that each symbol is actually received at and the phase that the symbol would have been received at under ideal conditions (i.e. the phase at which it was or should have been transmitted) , preferably over a group of plural symbols.
These finer offset estimations are preferably performed and used to adjust each received signal burst, even after the receiver has achieved 'lock in'.
All of the above estimations are preferably also used to adjust the receiver's local timebase and frequency references. Thus preferably the coarser offsets estimated and any finer offsets estimated are used together to adjust the receiver's local references appropriately. Once the receiver's local references are sufficiently close to the received signal frequency and timing, the coarser estimates need no longer be used, but preferably the finer estimate, if any, is always used to make any appropriate fine adjustments to the receiver's local frequency and timing references. It is believed that using an appropriate combination of coarser and finer offset estimates to tune a radio receiver's local frequency and/or timing
references is advantageous in its own right .
Thus, according to a third aspect of the present invention, there is provided a radio receiver comprising : a local frequency reference; means for receiving a transmitted radio signal; means for detecting a predetermined portion of the received signal from which information relating to the frequency of the received signal can be derived and for estimating therefrom a first frequency offset of the receiver's local frequency reference relative to the received signal; means for estimating a further frequency offset in the received signal from the expected transmitted phase and the actual received phase of the symbols in the received signal; means for using the first estimated frequency offset and the further estimated frequency offset to determine an overall frequency offset and for using the overall frequency offset to adjust the local frequency reference of the receiver.
According to a fourth aspect of the present invention, there is provided a method of operating a radio receiver having a local frequency reference, comprising: receiving a transmitted radio signal; detecting a predetermined portion of the received signal from which information relating to the frequency of the received signal can be derived and estimating therefrom a first frequency offset of the receiver's local frequency reference relative to the received signal; estimating a further frequency offset in the received signal on the basis of the ideal and actual phases of the symbols in the received signal; using the first estimated frequency offset and the further estimated frequency offset to determine an
overall frequency offset and using the overall frequency offset to adjust the local frequency reference of the receiver .
The above third and fourth aspects of the invention estimate the frequency of a received signal more accurately and therefore can be used to tune the local frequency reference of the receiver more accurately. This permits faster frequency lock to the received signal, and therefore improved readiness for operation of the receiver and faster call setup.
In a particularly preferred embodiment the receiver or method of the third or fourth aspect of the invention, respectively, further includes a local timebase reference and appropriate timing offset estimation processes or means, such as any one or more of the timing offset estimation processes or stages discussed above.
According to a fifth aspect of the present invention, there is provided a radio receiver comprising: a local timebase reference; means for receiving a transmitted radio signal; means for detecting a predetermined portion of the received signal from which information relating to the timing of the received signal can be derived and for estimating therefrom a first timing offset of the receiver's local timebase reference relative to the received signal; means for searching the received signal for a predetermined signal sequence over a window of time at a plurality of predetermined time delays relative to the start of the time window to thereby estimate a further timing offset; means for using the first timing offset and the further timing offset to determine an overall timing offset and for using that overall timing offset to adjust the local timebase reference of the receiver.
According to a sixth aspect of the present invention, there is provided a method of operating a radio receiver having a local timebase reference, comprising : receiving a transmitted radio signal; detecting a predetermined portion of the received signal from which information relating to the timing of the received signal can be derived and estimating therefrom a first timing offset of the receiver's local timebase reference relative to the received signal; searching the received signal for a predetermined signal sequence over a window of time at a plurality of predetermined time delays relative to the start of the time window to thereby estimate a further timing offset; using the first timing offset and the further timing offset to determine an overall timing offset and using that overall timing offset to adjust the local timebase reference of the receiver.
The fifth and sixth aspects of the invention estimate the timing of a received signal more accurately, and therefore can be used to tune the local timebase reference of the receiver more accurately. This permits faster timebase lock to the received signal, and therefore improved readiness for operation of the receiver and faster call set-up.
In a particularly preferred embodiment the receiver or method of the fifth or sixth aspect of the invention, respectively, further includes a local frequency reference and appropriate frequency offset estimation processes or means, such as any one or more of the frequency offset estimation processes or stages discussed above .
Thus, according to a seventh aspect of the present invention, there is provided a radio receiver comprising: a local frequency reference; a local timebase reference;
means for receiving a transmitted radio signal; means for detecting a predetermined portion of the received signal from which information relating to the frequency and timing of the received signal can be derived and for estimating therefrom a first frequency offset and a first timing offset of the receiver's local frequency and timebase references relative to the received signal; means for estimating a further frequency offset in the received signal from the expected transmitted phase and the actual received phase of the symbols in the received signal; means for searching the received signal for a predetermined signal sequence over a window of time at a plurality of predetermined time delays relative to the start of the time window to thereby estimate a further timing offset; means for using the first timing offset and the further timing offset to determine an overall timing offset and for using that overall timing offset to adjust the local timebase reference of the receiver; means for using the first estimated frequency offset and the further estimated frequency offset to determine an overall frequency offset and for using the overall frequency offset to adjust the local frequency reference of the receiver.
According to an eighth aspect of the present invention, there is provided a method of operating a radio receiver having a local frequency reference and a local timebase reference, comprising: receiving a transmitted radio signal; detecting a predetermined portion of the received signal from which information relating to the frequency and timing of the received signal can be derived and determining therefrom a first frequency offset and a first timing offset of the receiver's local frequency and timebase references relative to the received signal;
estimating a further frequency offset in the received signal on the basis of the ideal and actual phases of the symbols in the received signal; searching the received signal for a predetermined signal sequence over a window of time at a plurality of predetermined time delays relative to the start of the time window to thereby estimate a further timing offset; using the first timing offset and the further timing offset to determine an overall timing offset and using that overall timing offset to adjust the local timebase reference of the receiver; using the first estimated frequency offset and the further estimated frequency offset to determine an overall frequency offset and using the overall frequency offset to adjust the local frequency reference of the receiver.
The above third to eighth aspects of the invention preferably further comprise using each estimated frequency or timing offset to adjust the received signal in time or frequency appropriately, most preferably before that signal is passed to the next offset estimation stage.
The methods in accordance with the present invention may be implemented at least partially using software e.g. computer programs. It will thus be seen that when viewed from further aspects the present invention provides computer software specifically adapted to carry out the methods hereinabove described when installed on data processing means, and a computer program element comprising computer software code portions for performing the methods hereinabove described when the program element is run on a computer. The invention also extends to a computer software carrier comprising such software which when used to operate a radio system or receiver comprising a digital computer causes in conjunction with said computer said system or receiver to carry out the steps of the method
of the present invention. Such a computer software carrier could be a physical storage medium such as a ROM chip, CD ROM or disk, or could be a signal such as an electronic signal over wires, an optical signal or a radio signal such as to a satellite or the like.
It will further be appreciated that not all steps of the method of the invention need be carried out by computer software and thus from a further broad aspect the present invention provides computer software and such software installed on a computer software carrier for carrying out at least one of the steps of the methods set out hereinabove.
A preferred embodiment of the present invention will now be described by way of example only and with reference to the accompanying drawings, in which:
Figure 1 shows a radio receiver in accordance with the present invention; and
Figure 2 is a drawing of the complex plane showing ideal and exemplary received TETRA symbols in each quadrant.
The present embodiment is described with reference to the TETRA system, which is a TDMA (time division multiple access) radio system. It should be noted that the present invention is, of course, applicable to other radio systems (and not just TDMA systems) as well.
Figure 1 shows a radio receiver which receives at RF receiver subsystem 1 a radio frequency (RF) input signal . That signal is captured and down converted into a digital complex baseband representation of the signal by the RF receiver subsystem 1. This received complex baseband signal is then passed through a series of stages in the receiver which are used to estimate accurately the frequency and timebase offsets from the received signal, to adjust the received signal for these offsets internally, and to tune the receiver's local internal frequency and timebase references for subsequent transmission or reception.
The receiver operates in one of two modes . When the receiver is not tuned to the transmitted signal frequency and timing, i.e. it is out of lock, it must first synchronise itself to the relevant base station. It therefore searches incoming radio signals for base station synchronisation signals which it can attempt to lock to. In this mode the receiver is initially configured by a control subsystem 2, and its local timing reference Tref, local frequency reference Fref and DC null which are needed to control the RF receiver subsystem 1 are set at default levels which are approximately in the middle of their tunable ranges .
When the receiver identifies base station synchronisation signals, the appropriate complex baseband signal portion or burst is captured and is passed through the various stages of the receiver so that corrections to the frequency and timing of the received signal and of the receiver's local frequency and timebase references can be made to tune the receiver to the received signal.
In this arrangement the received RF input signal burst is down converted by the RF receiver subsystem 1 into a digital complex baseband representation of that signal . The baseband representation is then input to a DC null stage 3 which is responsible for measuring, tracking and removing DC offsets which may be present in the complex baseband signal from the receiver subsystem 1. The DC null stage 3 estimates DC offsets from the complex baseband signal and subtracts them digitally from the baseband signal before passing it to the next stage. Alternatively, the DC offset value may be applied to the receiver subsystem hardware where the offset may be removed. The output from the DC null stage in this synchronisation mode is then passed to a frequency correction field detection stage 4, which is responsible
for searching the incoming signal for a predetermined signal portion from which information regarding the frequency and timing of the transmitted signal can be derived. In the TETRA system this signal portion is the frequency correction field which is a portion of one type of burst transmitted by a base station (or another mobile station when operating in direct mode) in TETRA. Detection of this field in the signal can be used to yield coarse frequency and timing offset estimates (ΔF and ΔT) as follows.
In the TETRA system, the 'frequency correction field' (FCF) is defined by a sequence of bits •
(f
9,fιo, ,f
72) = (0, 0, ,0)
(f73 f7 ' - - - - f8θ) = ' 1 / 1 , .. - . # 1 )
which, when modulated, generate a long signal 2.25 kHz above the nominal carrier frequency, preceded and followed by a short signal at 6.75 kHz below the carrier frequency.
This field s detected by carrying out an integration over the signal 2.25 kHz above the carrier frequency:
where N=32 (the duration of the signal portion 2.25 kHz above the carrier frequency of the frequency correction field), y1 is the current complex symbol, sampled at the local symbol rate, z is a complex decision variable which reaches a maximum at the position of the frequency correction field, and * denotes the complex conjugate.
This integration is carried out for every received symbol .
Alternative methods could be used to detect the frequency correction field, such as computation of a fast Fourier Transform, discrete Fourier Transform, or matched filtering over an appropriate interval.
If the receiver is locked to the frequency of the transmitted signal, then the phase of the decision variable z± should be π/4 as this corresponds to 2.25 kHz at the local sample rate used. If there is a frequency offset between the received signal and the local reference, the frequency error or offset, ΔF, between the receiver's local reference and the transmitted signal : a rg { z ) -π/4 ΔF=
2πr
where T is the symbol duration.
The timing error or offset ΔT is estimated as follows. The ideal position or time, tfcf, of the frequency correction field in a synchronisation burst is fixed and predetermined and thus known to the receiver. This time, t£cf, will be the position or time, tpk, of the peak or maximum of the above decision variable Z (which can be derived from the above analysis) if the receiver is locked to the timing of the transmitted signal. Thus the timing offset :
The effects of noise fading on the signal and the finite duration of the frequency correction field means that the frequency and timing estimates derived are only coarse estimates.
The coarse offsets estimated are used by the frequency correction field detection stage 4 to shift
the received signal up or down in frequency and adjust its timing position by an appropriate amount to remove the estimated frequency and timing offsets before the signal is passed on to the next stage of the receiver, i.e. before any subsequent processing of the received signal burst.
The estimated frequency offset ΔF is removed from the received samples by correcting them with a rotating complex phasor pk as follows:
with pk recursively calculated on the real and imaginary part :
and Pi = cos(πΔFT/2)+jsin(IlΔFT/2) .
The estimated timing offset ΔT can be removed from the captured signal burst by advancing or retarding the start position of that signal burst in the receiver's memory by an appropriate amount .
The coarse frequency and timing offsets derived are also supplied to the timebase estimator 13 and automatic frequency control stage 15, which will be discussed further below.
The next stage of the receiver is a square timing recovery stage 5. This stage estimates a finer timing error, φ, which it derives from the phase of the received signal averaged over a burst. It also estimates the optimum sampling point of a received symbol in the signal (i.e. the positions of the symbols in the signal) , and decimates the signal from four complex samples per symbol to one complex sample per symbol.
These steps can be carried out using the method proposed by Oerder & Meyr {M Oerder & H Meyr, "Digital
Filter & Square Timing Recovery", IEEE Trans Comm, 36 , No. 5, 1988.}. That method comprises estimating the timing error, φ, by estimating, over a burst, the phase of the square magnitude of the received signal at the symbol rate using a discrete Fourier Transform. The received samples are also interpolated to an optimum sampling point (which is defined to be that sampling point which maximises the signal to noise ratio) and decimated from four samples per symbol to one sample per symbol during the interpolation process.
The derived error value φ is provided to the timebase estimator 13, and also used to interpolate the received signal digitally to the optimum sampling point before it is passed onto the next stage of the receiver. If desired, a less complex square timing recovery arrangement could be used which estimates the sampling instant, but does not necessarily interpolate the received samples to the optimum sampling time on the current burst. (This arrangement would have to iterate timebase corrections over successive bursts to achieve timing lock (frequency lock would not be affected) , thereby taking longer to achieve timing lock, and is therefore not preferred.)
The next stage of the receiver is the digital frequency correction stage 6. This stage estimates and removes any residual frequency offset, δF, in the signal on the basis of the estimated ideal phases and the actual received phases of the received symbols in the signal. It preferably considers the whole signal burst to estimate the residual frequency offset, δF. (There may be a residual frequency offset at this stage for a number of reasons, such an inaccurate previous estimate, Doppler Shift or noise.)
This digital frequency correction can be carried out in the TETRA system as follows. In the TETRA system, information bits are differentially modulated into symbols at the transmitter such that the phase
change between consecutive symbols can have one of 4 values only, φk = (π/4, 3π/4, -π/4 and -3π/4) . At the receiver, differential demodulation is carried out:
The demodulated symbols, zi f are complex numbers whose phase (under ideal conditions) will correspond to the transmitted phase. However, in practice the phase of the received signal will include errors due to noise, fading and frequency offset . The digital frequency correction stage subtracts the ideal phase from the received phase (i.e. the phase that the symbol is actually received at) for each symbol and integrates over the burst of N symbols:
N-l m y -jΦk e i=0
The ideal phase of each symbol is estimated by assuming that the received symbol lies in the same quadrant as the ideal transmitted signal. Thus since TETRA modulation uses only one symbol phase in each quadrant, knowing the quadrant gives the ideal phase of the received symbol . Figure 2 which is a drawing of the complex plane showing ideal and received symbols in each quadrant illustrates this. (In figure 2, '*' represents an ideal symbol, '+' represents a received symbol with a frequency error and noise, 'φ' is the phase angle of the ideal symbol, and 'ε' is the phase error of the received symbol due to the noise and frequency error.)
The frequency offset, δf, can then be calculated from the equation:
1 δ f= a r g (m)
2UT
where T is the symbol duration.
The received signal is shifted up or down in frequency by the digital frequency correction stage 6 by the appropriate amount to remove the estimated frequency offset before it is passed onto the next stage. The correction can be performed with a rotating complex phasor pk as described above for the frequency correction field detection stage 4, but using δf for the frequency offset . The fine frequency offset δF is also provided to the automatic frequency control stage 15.
The digital frequency correction stage is not essential, although it is strongly preferred, since it is a very useful step in achieving fast frequency-lock. An alternative would be to for successive signal bursts estimate the frequency offset (e.g. using frequency correction field detection on synchronisation bursts) , and apply the offset to the local frequency reference iteratively until frequency-lock is achieved. This would naturally increase the time it takes to achieve frequency-lock, although timing-lock would not be affected.
The next stage in the receiver is the slot synchronisation stage 7. This stage is responsible for locating the precise position (timing) of the slots
(i.e. particularly groups of symbols) in the received burst in terms of the slot offset from the expected slot position. It does this, as is known in the art, by searching the received signal for a training sequence over a window of time, and at a plurality of different time delays, to find the best correlation between the received signal and a reference training sequence. This yields a timing offset estimate δT and a training sequence identity, TS . The received signal is advanced or retarded by the appropriate amount to remove the estimated timing offset before it is passed onto the next stage. The estimated timing offset δT is also
provided to the timebase estimator 13.
This completes the processing of the signal as regards frequency and timebase adjustments. However, for completeness, the subsequent stages of the receiver will briefly be described. These stages comprise a differential demodulator 8 which converts the complex signal information it receives from the signal into soft decision values, each value representing one bit of the signal . This differential demodulation can be carried out in a conventional way by calculating the following for each symbol (i=l, 2...,N) in a burst of N symbols:
where z is a complex number whose real and imaginary parts represent (soft) information bits, γl t γ1_1 are the current and previous received complex symbol , and * denotes the complex conjugate.
The soft decision value is then passed to the burst demultiplexer 9 which splits the received soft decision values into appropriate streams for the logical channel decoders 10, 11, 12, based on the training sequence identified by the symbol synchronisation unit 7 and the initial system configuration. The logical channel decoders 10, 11, 12 decode the signal and provide the appropriate output bits.
As well as being used to adjust the received signal as it is processed by the receiver, the estimated frequency and timing offsets are also used to adjust the local internal frequency and timebase references of the receiver, to try to tune them to the received signal.
The local timebase reference or system timebase 14 of the receiver is adjusted by the timebase estimator 13. The timebase estimator 13 combines the coarse timing estimate ΔT from the frequency correction field detection stage 4, the symbol timing estimate δT from the symbol synchronisation stage 7, and the fine timing
phase φ from the square timing recovery stage 5 into an overall timing offset. This overall timing offset is applied to the system timebase 14 at the appropriate time to synchronise the receiver's timebase to the base station's timebase such that the timebases are in lock. If desired, the applied timing offset may be smoothed to remove jitter. It should be noted that the adjustment to the system timebase 14 only affects signals which are received after the adjustment is made. Thus prior to this the receiver must rely on its ability to make timing adjustments within the processing stages as described above .
The receiver's local frequency reference or system frequency reference 16 is adjusted by the automatic frequency control stage 15 on the basis of the coarse frequency estimate ΔF and the fine frequency estimate δF. The automatic frequency control stage 15 combines the coarse frequency estimate ΔF and the fine frequency estimate δF into an overall frequency offset . This is applied to the system frequency reference oscillator to remove the offset. This process tunes the radio receiver's frequency reference to the base station. If desired, the applied frequency offset may be smoothed to remove jitter. (Again it should be noted that the adjustment to the system frequency reference 16 only affects signals which are received after the adjustment is made. Prior to this the receiver must rely on the frequency adjustments made within the signal processing stages as described above . ) The automatic frequency control stage 15 also estimates if the frequency is in lock (at which point transmissions are allowed to be made) . It provides a signal Flock to the control subsystem 2 to indicate lock status . Once the receiver's local frequency and timebase references are sufficiently close to the received signal frequency and timing (e.g. within a predetermined margin
of those parameters), i.e. the receiver is in lock, it is no longer necessary to estimate and apply the coarse frequency and timing offsets ΔT, ΔF performed by the frequency correction field detection stage 4. Thus the receiver switches to normal or lock mode in which incoming signals pass through the DC null stage 3, but by-pass the frequency correction field detection stage 4, before passing onto the subsequent stages. This is because in this mode it is necessary to make only small corrections to frequency and timebase digitally to the received signal since the receiver is already in lock. Small corrections are also made in the analogue domain to the system time base.
It can be seen that the above embodiment provides a baseband receiver architecture for a mobile radio system in which the frequency and timing of a received signal can be measured accurately and used to tune local frequency and timebase references to enable the mobile station to be ready for subsequent transmission or reception rapidly. The applicants have found that typically the described baseband receiver architecture nearly always enables the mobile station to receive and decode the initial burst received, and, having adjusted its frequency and timebase reference, to receive further signals or make transmissions at the first opportunity (i.e. within one burst) .
In its preferred embodiment, the receiver uses two different frequent estimates and three different timebase estimates to configure the mobile ' s frequency and timing references. By virtue of this architecture, faster frequency and timebase lock to base station (or mobile station in direct mode) are achieved, thereby providing improved readiness for operation and faster call setup. The present invention is particularly applicable to mobile radio stations operating in direct mode (i.e. in which mobile stations communicate directly over the air
interface without the presence of a base station) , since in that mode of operation the frequency and timing reference is generally not as stable as that originating from a mobile station from a fixed base station. In this case the frequency and timebase agility of the receiving mobile station must be higher. The present invention provides this .