Pulse or digital communications – Synchronizers – Phase displacement – slip or jitter correction
Reexamination Certificate
2000-01-27
2003-12-02
Chin, Stephen (Department: 2734)
Pulse or digital communications
Synchronizers
Phase displacement, slip or jitter correction
C375S354000, C375S373000
Reexamination Certificate
active
06658075
ABSTRACT:
TECHNICAL FIELD
The invention relates to the field of mobile radio communications.
BACKGROUND
Radio communication systems employing digital signalling are becoming increasingly common. The information carried by any digital communication system can be recovered only after the receiver has first synchronised to the received signal. Thus there is a need in a mobile digital radio system for synchronisation in the radio receiver.
Synchronisation, in the sense used here, means that the radio receiver has to know:
a) When is the best time to sample the received signal to recover the bits carried by each symbol. This is ‘symbol synchronisation’.
b) Where each block of symbols starts and ends in time. This is ‘frame synchronisation’.
c) What the carrier frequency of the received signal is, if the signalling employed in the communication system is based on using a carrier signal. This is ‘frequency control’.
A state of synchronism is often simply referred to as ‘lock’.
Several Time Division Multiple Access (TDMA) type digital radio communications systems are known, such as ‘TETRA’, ‘GSM’ or ‘IS54’. In TDMA systems the received signal carries sequences of known, predetermined symbols, which are termed ‘training sequences’. The training sequences are used by the receiver to estimate the symbol & frame timing phase and the carrier frequency offset. These training sequences therefore allow the receiver to achieve synchronism.
The algorithms used to achieve synchronism are typically based on ‘maximum likelihood’ theory. For symbol and frame timing recovery, maximum likelihood theory is applied by correlating the symbol wave-form or training sequence waveform with the actual received signal over a block of data. Clearly, the actual signal received may have been corrupted by additive ‘noise’ or distorted by the communication channel. Furthermore, the receiver clearly must apply the maximum likelihood technique at a time when it is not yet synchronised to the training sequence, although the values of the training sequence wave-form are known to the receiver. The assumed correct timing phase is taken as being the phase for which there is a maximum value of the correlation.
An automatic frequency control (AFC) circuit in a radio receiver adjusts the reference frequency used within the receiver to match the carrier frequency of the received signal. An automatic frequency control circuit can use the derivative of a maximum likelihood function to drive the reference frequency used within the receiver to lock. The derivative of the maximum likelihood function is zero at the lock point.
The arrangements detailed above relate to how synchronisation is achieved within a radio receiver.
Besides actually achieving synchronisation however, it is extremely useful for the receiver to know accurately at any time whether or not a state of synchronisation has yet been reached. In particular, the receiver should have some indication of the reliability of the timing phase found, and to know when the frequency loop has locked. This is because a receiver may approach a state of synchronisation in steps. Also, it is sometimes possible for a receiver's synchronisation circuitry to adopt the mode of operation appropriate to when synchronisation has been achieved, even though synchronisation has not in fact been achieved.
For example, the synchronisation circuitry will find a maximum of the correlation when the received signal is entirely absent. This anomalous value of correlation is due to the nature of the maximum likelihood approach to symbol and frame timing recovery. If the received signal is absent but the receiver then mistakenly assumes from the finding of maximum correlation that synchronisation has been attained, then clearly the receiver would continue decoding the data without knowing if the signal is there, or whether it had found the correct phase. To operate effectively therefore, the receiver requires some form of lock detector, both for timing recovery and frequency synchronisation.
The performance of the lock detector is in fact critical to the performance of the synchroniser. The lock detector must reliably reject a synchronisation indication when no signal is present or when a false synchronisation position is found. Otherwise, the receiver will waste time and battery power trying to decode non-existent signals. The action of mistakenly accepting a false state of synchronisation is termed ‘falsing’. A good synchronisation lock detector must therefore have a low rate of falsing. A lock detector must also reliably recognise and accept a correct synchronisation, for the receiver to reliably set up calls. These two requirements are generally conflicting. Therefore a good lock detector is one which finds a good compromise between having a low falsing rate and low rejection of good locks.
The radio units in a public or private mobile radio communications system are often referred to as ‘mobile stations’. Normally, a mobile station is within communication range of a base station of the communication system's infrastructure. In this case, the mobile station will communicate through the base station, this mode of operation typically being termed ‘trunked mode’. However, some mobile radio communication systems allow an individual mobile station to set up a direct radio link to another mobile station, without the communication link passing through the infrastructure, for example a base station, of the communication system. This form of communication between two mobile stations is referred to as ‘Direct Mode’ operation.
Mobile stations operating in direct mode have to receive radio signals which typically show greater variation in their parameters than radio signals received from the infrastructure of the communications system. This is because of differences in the signals transmitted from a mobile station in comparison to those from a base station of the infrastructure, and signal path differences. For example,
a) A mobile station may only transmit with low or variable power;
b) A mobile will typically have a lower quality internal clock than a base station;
c) A mobile may be moving, but a base station is stationary, movement of the mobile introducing characteristics such as Rayleigh fading to the signal;
d) A mobile may suffer interruption of its transmission due to passing by an obstruction, low power or poor battery contacts. As these factors don't affect a base station, the base station will suffer interruption of its broadcast signal less often.
The above factors make the need for accurate synchronisation, and recognition of synchronisation, particularly important for direct mode operation of mobile stations.
Several circuit arrangements that enable a radio receiver to establish synchronisation are known in the prior art. Examples of these circuit arrangements are explained briefly below, under the headings ‘frame timing recovery’ and ‘frequency lock detection’.
Frame Timing Recovery
Two techniques are known for frame timing recovery:
1) Motorola sells TETRA radios under the system name ‘Dimetra’. The Dimetra trunked mode algorithm is an example of a prior art technique for frame timing recovery. This algorithm compares the correlation value with the received signal energy. This algorithm works well when the mobile radio is communicating in trunked mode, because the mobile radio is more likely to be receiving a signal.
The algorithm must be capable of rejecting noise-like signals. These occur more in direct mode operation than in trunked mode operation. In fact, the algorithm must discriminate very well against noise-like signals for direct mode operation.
2) The second technique for frame timing recovery involves always accepting the lock, attempting to decode the data, and relying on the error coding in the received data to reject false locks. The received data has ‘CRC’ coding (cyclic redundancy coding) to enable the receiver to recognise errors in the data. The problem with this technique is that the receiver does not reject a false lock early enough in this process. The receiver th
Ahn Sam K.
Davis Valerie M.
Santema Steven R.
LandOfFree
Synchronization lock detector and method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Synchronization lock detector and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Synchronization lock detector and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3152498