SYNCHRONIZATION METHOD AND DATA TRANSFER SYSTEM
FIELD
[0001] The invention relates to a data transfer system and a synchronization method in a data transfer system employing multicarrier transmission.
BACKGROUND
[0002] Wireless data transfer systems have recently been the object of intense development. Several new services have been developed, and there is a need to develop solutions enabling a high data transfer capacity. Alongside the generally used FDMA (Frequency Division Multiple Access) and CDMA (Code Division Multiple Access), data transfer methods utilizing multi- carrier transmission have been developed. In these methods, a signal is transmitted simultaneously by using several different carriers.
[0003] Multicarrier signals are used particularly in wideband data transfer over a fading multipath radio channel. Multipath propagation means that a signal propagates along several different paths between a transmitter and a receiver. The division of a signal into several parallel carriers prolongs the duration of the symbols transferred in the radio channel and enables frequency-domain encoding, thus allowing interference caused by multipath propagation to be eliminated without substantially compromising the data transfer capacity.
[0004] In a digital multicarrier system, information is divided into a plurality of parallel carriers at the frequency domain, after which conversion into the time domain is performed, after which the signal is transmitted. Timing, i.e. synchronization is essential in data transfer and particularly in multicarrier transmissions. In prior art solutions, a few known symbols are created for the synchronization and they are transmitted, modulated at the frequency domain, prior to the actual data. The receiver is able to identify the synchronization symbol and initiate timely reception. Accurate timing is important, since the time-domain signal is again converted in the receiver into the frequency domain for expression.
[0005] In prior art solutions; synchronization typically takes place by the use of different correlation methods. Known methods include for instance the use of a differential correlator and a matched filter.
[0006] In differential correlation, a signal is divided into two parts,
one of which is delayed and correlated with the non-delayed part. If the signal received and sampled by the receiver is composed of two successive identical sequences, a peak is obtained from the output of the correlator, and when the peak exceeds a predetermined threshold level, the transmission is assumed to arrive, and reception is initiated.
[0007] When a matched filter is used, correlation is calculated between the incoming signal, modulated at the frequency domain, and coefficients pre-programmed in the filter. When the signals are sufficiently similar, a peak is obtained from the output of the matched filter, the peak being compared with the threshold level, and once it is exceeded, reception is initiated.
[0008] The weaknesses in prior art methods are for instance poor timing accuracy, complexity of implementation, slowness and reliability of synchronization.
[0009] In addition to time synchronization, frequency synchronization constitutes a problem, since the orthogonality of a multicarrier signal weakens significantly because of the frequency error, thus impairing performance. In addition, the functioning of phase-accurate, i.e. coherent signal detection requires that the transmitter and receiver function very accurately at the same frequency. In practice, this is not possible, but generally the receiver has to measure the frequency difference between the transmitter and the receiver from the received signal, and to amend its frequency or the incoming signal such that the frequency difference is compensated for.
[0010] In adaptive systems, wherein for instance the modulation method or other transmission parameters employed in signal transmission vary, the shortcoming in the above methods is that part of the transmitted signal has to be allocated to information transfer required for adaptation, which impairs the efficiency of the usage of the band.
BRIEF DESCRIPTION
[0011] The object of the invention is to implement an efficient synchronization solution.
[0012] This is achieved with a synchronization method in a data transfer system, wherein bursty data modulated at the frequency domain is transferred to several carriers. A pseudorandom bit string is added to the beginning of each burst in a transmitter, the bit string being transmitted before other data by using one carrier, and, after the pseudorandom bit string, at least one frequency-domain modulated synchronization pattern, which is transmitted
by using a plurality of carriers, and a receiver employs the bit string when time and/or frequency synchronizing with a received signal.
[0013] The invention also relates to a data transfer system comprising a transmitter and a receiver, the transmitter being arranged to transmit bursty data modulated at the frequency domain to a plurality of carriers. The transmitter comprises means for adding a time-domain modulated pseudorandom bit string to the beginning of each burst, and the transmitter is arranged to transmit the bit string before other data by using one carrier, and that the transmitter is arranged to add at least one frequency-domain modulated synchronization pattern to the beginning of each burst after the bit string, and that the transmitter is arranged to transmit the synchronization pattern by using a plurality of carriers and that the receiver comprises means for employing the pseudorandom bit string when time and/or frequency synchronizing with a received signal.
[0014] The invention also relates to a transmitter arranged to transmit bursty data modulated at the frequency domain to a plurality of carriers. The transmitter comprises means for adding a time-domain modulated pseudorandom bit string to the beginning of each burst and at least one frequency- domain modulated synchronization pattern after the pseudorandom bit string, and the transmitter is arranged to transmit the bit string before other data by using one carrier and the synchronization pattern by using a plurality of carriers.
[0015] The invention also relates to a receiver arranged to receive a bursty signal transmitted frequency-domain modulated to a plurality of carriers. The receiver comprises means for identifying, at the beginning of each burst, a bit string added thereto and transmitted on one carrier, and a synchronization pattern transmitted on a plurality of carriers, and means for employing the pseudorandom bit string when time and/or frequency synchronizing with a received signal.
[0016] Preferred embodiments of the invention are described in the dependent claims.
[0017] The solution of the invention provides a plurality of advantages. At the time domain, the pseudorandom PN sequence can be implemented shorter in duration than a synchronization symbol modulated at the frequency domain, synchronization thus being faster than with known methods. In addition, synchronization accuracy is better than the accuracy of synchroni-
zation implemented with a differential correlator, for example. The use of binary bit strings makes the implementation of the matched filter used in the receiver easy and it consumes little calculation power compared with a situation where a frequency-domain modulated signal is correlated with a matched filter. [0018] The proportion of the peak power of a multicarrier signal to the mean power is extremely high; i.e. the transmission includes strong momentary peaks even if the average signal level were low. The linearity of the transmitter and receiver parts has to be dimensioned according to the requirements of the peak power. The envelope curve of a pseudorandom PN signal is almost constant, i.e. the proportion of the peak power to the mean power is low. This being so, the PN sequence can be transmitted at a significantly higher mean power than the actual data, whereby synchronization becomes extremely reliable.
LIST OF THE FIGURES
[0019] In the following, preferred embodiments of the invention will be described in detail with reference to the accompanying drawings, in which
Figures 1A, 1 B and 1 C show examples of the structure of a burst and a multicarrier transmission,
Figure 2 illustrates an example of a transmitter,
Figure 3 illustrates an example of a receiver,
Figure 4 illustrates a second example of a receiver, and
Figure 5 illustrates a third example of a receiver.
DESCRIPTION OF THE EMBODIMENTS
[0020] Preferred embodiments of the invention can be implemented in data transfer systems employing multicarrier transmission, wherein bursty data is transmitted. With reference to Figure 1A, an example of the structure of a burst and a multicarrier transmission will be studied. Figure 1A shows a burst 100. The burst comprises actual data 102 to be transferred and composed of one or more successive symbols, for example. Before the data, the burst comprises at least one synchronization symbol 104 used in a receiver for channel estimation. The synchronization symbol may be for instance such a symbol whose phase and amplitude are previously known. In preferred embodiments, a pseudorandom bit string 106, which is a binary bit string, for example, is added to the beginning of the burst before the synchronization symbols.
[0021] Accordingly, a signal transmitted in a multicarrier system is
divided into a plurality of parallel carriers at the frequency domain, followed by conversion to the time domain, after which the signal is transmitted, Figure 1A illustrates carriers 108 to 118 used in the transmission. The horizontal axis shows time and the vertical axis frequency. The number of carriers and the frequency difference between them are system-dependent parameters, as is evident to a person skilled in the art.
[0022] In the transmission of the burst 100, the bit string 106 at the beginning of the burst is modulated for instance by the BPSK method and transmitted using one carrier 112. The rest of the burst, i.e. one or more synchronization symbols 104 and the actual data are then modulated into carriers 108 to 118 reserved for the purpose, and transmitted to a receiver. The bit string 106 may be composed of one or more BPSK-encoded binary pseudorandom sequences.
[0023] Figure 1 B illustrates another embodiment wherein the bit string 106 of the burst 100 is transmitted on what is called a 0 carrier 120, located at the medium frequency of the transmission. The number of frequencies 122 and 124 on both sides of the medium frequency is thus equal. In an embodiment, no actual data is transmitted on the carrier 120 employed in the transmission of the bit string.
[0024] Figure 1 C illustrates an example of a signal burst at the time domain. In the figure, the horizontal axis is time and the vertical axis amplitude. The burst comprises a bit string 106, synchronization symbols 104 and actual data 102. The figure illustrates the high proportion of the peak value of the data signal 102 to the average signal. Accordingly, the data signal comprises strong momentary peaks. This being so, the transmitter and receiver parts have to be dimensioned according to the highest signal to be sufficiently linear. The amplitude of the bit string 106, in turn, does not vary, and it can therefore be transmitted at a higher mean power than the data signal 102. This improves the reliability of the synchronization.
[0025] Figure 2 illustrates an example of a transmitter. The transmitter comprises a modulator 200, the data 202 to be transmitted being at its input. In the modulator 200, the data is processed as blocks of a selected length (e.g. 400 bits, but the block length can also have any other value). Each block can be encoded e.g. by convolutional encoding and interleaved, i.e. the order of the bits in the block is mixed with the desired sequence. In this context, the data is divided into different carriers. The blocks are then modulated and pre-
distorted. The operation of the modulator 200 is known to a person skilled in the art. Modulated complex symbols 204 are at the output of the modulator, and they are applied in the desired order to a converter 206, wherein the symbols are converted from the frequency domain into the time domain by means of IFFT (Inverse Fast Fourier Transform), for example. Before the actual data symbols, the desired number of known synchronization symbols that the receiver uses in signal detection is applied to the converter 206.
[0026] A complex output signal 208 of the converter is further applied to a signal processing block 210 implemented by means of a processor and signal processing software or programmable logics, for example. Other alternatives are also feasible, as is evident to a person skilled in the art. In the signal-processing block 210, the complex signal can be modified in the desired manner, compressed, limited, filtered or buffered, for example.
[0027] The transmitter also comprises a control unit 212 for controlling the operation of all blocks required for signal processing. The control unit also controls the operation of a bit block generator 214. Controlled by the control unit, the bit block generator 214 generates a selected pseudorandom bit string, which is added to the signal before the actual data and the frequency- domain modulated synchronization symbols.
[0028] A complex mixer 216 transfers the complex signal to the desired intermediate frequency. This can also be carried out analogically in radio frequency parts. The intermediate-frequency signal is converted into analogical form in a D/A converter 218 and applied to radio frequency parts 220 for amplification and mixing to the actual transmit frequency. Finally, the signal is transmitted by means of an antenna 222.
[0029] In an embodiment, the control unit selects the bit string to be transmitted at the beginning of each burst from several alternatives. On the basis of the selection, the control unit transmits a corresponding control to the bit block generator 214, which generates the desired bit string. This way, the bit string can be used to transfer some desired information to the receiver. For example, in adaptive systems, wherein the modulation method employed in transmission varies, information about the modulation method employed can be transferred to the received by using a given bit string.
[0030] Figure 3 illustrates an example of a receiver. The structure presented is particularly suitable for embodiments where the bit string employed does not vary. The receiver receives a signal with an antenna 300. A
received signal 302 is applied to radio frequency parts 304 wherein the signal is typically amplified, filtered and converted into the intermediate frequency. The intermediate-frequency signal is applied to an analogue/digital converter 306, wherein the signal is converted into digital form. From the output of the converter, the signal is applied to a mixer 308, wherein the signal is mixed complexly to the base frequency, the signal then being composed of its 1 and Q parts that have a 90-degree phase shift. This can be also carried out analogically in the radio frequency parts. The base-frequency signal is applied in parallel to a signal processing block 310 and a matched filter 312. The signal- processing block 310 is implemented by means of a processor and signal processing software or programmable logics, for example. Other alternatives are also feasible, as is evident to a person skilled in the art.
[0031] The receiver uses the matched filter 312 to synchronize with the bit string at the beginning of the burst. In the filter, correlation is calculated between the pre-programmed PN sequence and the received I and Q signal parts. The output of the filter 312 is coupled to a control unit 314 wherein squared sum power is calculated from the correlation result of the I and Q parts. The calculated sum power is compared with a preset threshold value, and once the value of the sum power exceeds the threshold value, the control unit detects the starting time of the reception of the burst.
[0032] In the signal processing block 310, the complex signal can be processed in the desired manner, determine the frequency error in the signal, for example. Once the frequency error is known, it can be used as the basis for frequency correction in the mixer 308. Thus, synchronization can be performed on a frequency-corrected signal, which improves the synchronization.
[0033] From the output of the control unit 314, a synchronizing signal 316 is applied to a converter 318 and a demodulator 320. The received in the converter 318 is converted from the time domain to the frequency domain by means of FFT (Fast Fourier Transform), for example. An output signal of the signal-processing block 310 is at the input of the converter 318. When the converter 318 receives the synchronizing signal 316 from the control unit 314 about the start of the burst, the converter is able to immediately synchronize with the following part of the burst, i.e. one or more synchronization symbols. The synchronization symbols are for instance symbols whose phase and amplitude are previously known. The receiver uses the synchronization symbols
for instance for channel estimation, i.e. to measure the signal distortion caused by the radio path. This allows the phase and amplitude changes occurred on the signal path to be detected and compensated for in the receiver. After the synchronization symbols, the converter receives the actual data. From the output of the converter 318, the signal, converted into the frequency domain, is applied to a detector 320, wherein prior art channel decoding, deinterleaving and data unloading from carriers are performed. Each burst comprised at least one frequency-domain modulated synchronization pattern.
[0034] Figure 4 illustrates a second example of a receiver. The structure presented is particularly suitable for embodiments wherein the bit string employed is varied. In this example, the front end of the receiver is similar to that in the previous example. The operation of the antenna 300, radio frequency parts 304, analogue/digital converter 306 and mixer 308 are as described above. From the output of the mixer 308, the base-frequency signal is applied to the signal-processing block 310 and to a set of matched filters 400 to 406.
[0035] The output of each filter 400 to 406 is coupled to the control unit 314. Parameters corresponding to the bit strings employed in the transmission are pre-programmed in the filters. Once a received bit string is applied to the matched filters 400 to 406, a peak is obtained from the output of the matched filter that contained the parameters of said bit string. In the control unit 314, the output signal of each filter is compared with a preset threshold value, and when the output of a filter exceeds the threshold value, the control unit detects the starting time of the reception of the burst and, on the basis of the filter that detected the bit string, the control unit is able to conclude which bit string was concerned. The received signal can now be processed in the signal-processing block 310, the converter 318 and the demodulator 320 and in other blocks according to the information obtained from the control unit.
[0036] In an embodiment, the receiver employs an antenna array composed of a plurality of antennas or antenna elements for receiving signals. This allows the directivity of the antenna array to be adjusted by phasing a signal received with different antenna elements and summing up the signals. The known bit string, added to the beginning of each burst can be employed in the receiver for determining the phase differences of the signals coming from the different antenna elements. The control unit can utilize this information in determining a suitable phasing for obtaining the desired directional pattern.
[0037] Figure 5 illustrates an embodiment employing an antenna array. A signal is received with an antenna array 500 composed of a plurality of antenna elements. The signal received with the antenna array 500 is applied to a mixer 506 via radio frequency parts 502 and an A/D converter 504. The operation of these parts is as described above. The receiver also comprises a summer 508, wherein the signals received with the different antenna elements are phased and summed up on the basis of a control coming from the control unit 314. In the matched filter 312, the original phase of the signal received with each antenna element is determined, and this information is transferred to the control unit 314, which utilizes it in determining the phasing required for obtaining the desired directional pattern. From the summer, the signal is further applied to the signal-processing block 310.
[0038] Although the invention is described above with reference to the example according to the attached drawings, it is apparent that the invention is not limited thereto, but can be modified in a plurality of ways within the scope of the appended claims.