EP1066707A1

EP1066707A1 - Frame synchronisation in multi-carrier transmission systems

Info

Publication number: EP1066707A1
Application number: EP99922038A
Authority: EP
Inventors: Markus Radimirsch; Karsten Brueninghaus
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 1998-04-01
Filing date: 1999-03-18
Publication date: 2001-01-10
Also published as: DE19814530A1; JP4298918B2; US6879649B1; WO1999052252A1; JP2002510939A

Abstract

The invention relates to a method for digitally transmitting data in a wireless communications network. A controlling communication ("downlink") is executed from a master station to all other subscribers within a fixed bandwidth of a channel with the help of a fixed signaling frame (1). The start of the signaling frame (1) is characterised by a special frame synchronisation signal (5) which can be detected by the subscribers. The frame synchronisation signal used consists of at least one sub-range of the bandwidth being occupied by transmitting signals and at least one complementary sub-range of the bandwidth not being occupied by transmitting signals.

Description

FRAME SYNCHRONIZATION IN MULTI-CARRIER TRANSMISSION SYSTEMS

The invention relates to a method for the digital transmission of data in a wireless communication network in which controlling communication from a master station to all subscribers ("downlink") takes place within a defined bandwidth of a channel with the aid of a defined signal frame and the beginning of the signal frame is identified by a special frame synchronization signal that can be detected by the participants.

The invention further relates to a receiving device for receiving data transmitted according to the aforementioned method.

Already implemented digital data transmission systems such as DAB (Digital Audio Broadcast) or DVB (Digital Video Broadcast) are based on OFDM modulation (Wächter: "The transmission method of the future digital radio broadcasting", the telecommunications engineer 11 and

12/92, pages 1 to 43; Engels, Rohling, Breide "OFDM transmission method for digital television broadcasting", broadcasting technology announcements 1993, pages 260 to 270). In order to enable the detection of the transmission frame, a so-called

The zero symbol is used, ie there is no or only one 2 very low power sent out. In such pure distribution services, the zero symbol is unique and therefore the detection in the receiver is simple and unambiguous.

It is also known to transmit symbols for synchronization purposes which can be detected in the receiver using correlation techniques due to their correlation properties, in particular autocorrelation properties (e.g. DE 43 19 216 AI). The correlation technique has the disadvantage that it is relatively complex to implement in the receiver because of the many complex multiplications and therefore consumes a lot of power.

In a communication system in which the transmission channel can be occupied alternately by a base station and by mobile users, as is the case for mobile radio systems, the zero symbol suitable for the pure distribution services for the frame synchronization is not clear, because before the start of the transmission - transmission from the base station into the transmission channel

("Downlink") and also at the start of the transmission from a mobile radio subscriber to the transmission channel ("Uplink") there is a transmission-free transceiver turnaround interval and, moreover, the time slots for the uplink signaling are not mandatory, since the subscribers partially access the transmission channel in random access.

Since the detection of the beginning of the frame, in particular in a wireless TDMA (time division multiple access) -

TDD (time division duplex) multi-carrier transmission system is of particular importance, since the duration between two downlink phases (the "frame duration") can vary due to a flexible organization of multiple access of the participants via the downlink, so far the disadvantages of the correlation technique to be accepted. 3

In contrast, the invention is based on the problem of enabling simple recognition of the beginning of the frame, that is to say in particular the beginning of a downlink phase.

Based on this problem, the method of the type mentioned at the outset is characterized in that an occupancy of at least a portion of the bandwidth with transmit signals and non-occupancy of at least a complementary portion of the bandwidth with transmit signals are used as the frame synchronization signal.

The method according to the invention thus provides for a frame transmission in which a special frequency pattern is transmitted at the beginning of each transmitter frame and is detected in the receiving device. The frequency pattern consists of a defined occupancy of at least a partial area of the bandwidth with transmit signals and non-occupancy of at least one complementary partial area of the bandwidth with transmit signals.

In the simplest and preferred embodiment of the invention, one half of the bandwidth is occupied by transmission signals and the other half of the bandwidth is free of transmitter signals. Such a pattern can be generated very easily in an OFDM system if the upper or lower N / 2 subcarrier amplitudes of the N available subcarriers are set to zero.

The frame synchronization signal according to the invention can be detected according to the invention in a receiving device with a detection device which has a filter arrangement for dividing the intermediate frequency band into partial areas and a comparison device for comparing the received transmission energy in the partial areas. Such training of the reception 4 devices is possible with little hardware and can be easily optimized with regard to the required electrical power. In particular, the receiving device does not have to carry out complex multiplication functions continuously, as in correlation technology, in order to be able to recognize a frame synchronization signal. Rather, it is possible in the receiving device according to the invention to operate a power supply of the receiving device in a power-saving mode and to switch to a full operating state depending on the detection of the frame synchronization signal by the detection device if the full electrical energy for the evaluation of the downlink signal is within of the signal frame is required.

In the context of the present invention, it is expedient if a uniform amplitude distribution is implemented in the sub-area occupied by transmission signals. When using N subcarriers in the OFDM system, the phase and amplitude of the N / 2 subcarriers used for the transmission signals are selected so that, on the one hand, the energy is distributed as evenly as possible over all occupied subcarriers, in order to increase sensitivity to frequency-selective interference in radio - to minimize channel, and on the other hand the resulting

Time signal has an envelope that is as constant as possible in order to avoid problems with non-linear transmission amplifiers.

The frame synchronization signal designed in accordance with the invention is unambiguous for the systems to be considered, since all other symbols have a uniform power distribution over the subcarriers used. This also applies to the zero symbol. Accordingly, by evaluating the power difference in the

Subareas that are occupied or not occupied, the 5

Frame synchronization signal can be detected. The performance difference can be evaluated analog or digital.

The master station emitting the frame synchronization signal according to the invention can regularly be a base station of a mobile radio network. However, it is also possible to permanently or temporarily assign a master function to a user in a network in which the users communicate directly with one another, ie not via a base station, so that this user then represents the master station in the sense of the invention.

6

The invention will be explained in more detail below with reference to exemplary embodiments shown in the drawing. Show it:

Figure 1 - an example of a data structure in a method according to the invention

Figure 2 - an example of a signal structure of a downlink signal

Figure 3 - a schematic representation of a frame synchronization signal according to the invention for an OFDM transmission

Figure 4 - a schematic diagram for the construction of a receiver for OFDM signals

FIG. 5 shows a block diagram of a detection circuit for recognizing the frame synchronization signal according to the invention

Figure 6 - a variation of the detection circuit according to Figure 5 for the additional detection of a zero signal

Figure 7 - a schematic representation of a first embodiment of an evaluation circuit for the frame synchronization signal

Figure 8 - a second embodiment of an evaluation circuit for the frame synchronization signal

Figure 9 - a third embodiment of an evaluation circuit for the frame synchronization signal 7

Figure 10 - another embodiment of a frame synchronization signal according to the invention

Figure 1 shows that for communication between a base station and a large number of communication participants, for example in a mobile radio network, a signal frame 1 from a downlink phase 2, in the downlink signals DS and DC from the base station via the wireless transmission channel to the participants are sent, and there is an uplink phase 3 in which uplink signals UC and US are transmitted from subscribers to the base station via the transmission channel.

Figure 1 shows before the start of the downlink phase 2 and the uplink phase 3 short transmission-free intervals 4, which result from the switching of the devices between transmission and reception mode (transceiver turn-around interval). These intervals, which have no transmission energy, result in an ambiguity of zero signals which are otherwise used as a frame synchronization signal.

The organization of the communication on the transmission channel is carried out by the base station

Transmission of the downlink signals DC in the downlink phase 2, through which the individual subscribers are allocated time slots for the transmission of data during the uplink phase 3. In uplink phase 3, the participants also share any information with the base station

Send requests that will be taken into account in the allocation during the next downlink phases 2.

FIG. 2 shows details of the downlink signals in the downlink phase 2, which follows the switchover interval 4. The downlink signal in the downlink period 2 begins with a frame synchronization signal 5 that 8 can be designed according to the invention in accordance with the exemplary embodiments explained in more detail below. A preamble 6 and two OFDM symbols 7 for fine synchronization then follow. This is followed by data packets 8 in the number required in each case, which results in the variable length of the downlink phase 2.

An embodiment of a frame synchronization signal 5 is shown in Figure 3.

The abscissa shows N frequencies for subcarriers of an OFDM signal in discrete equal intervals, of which there are N / 2 subcarriers on the positive and on the negative side of a main carrier frequency.

In the illustrated embodiment, the N / 2 subcarriers on the negative side are switched off, i.e. without signal energy. The N / 2 subcarriers on the positive (p) side, on the other hand, are transmitted with an amplitude that is equal to one another to form the frame synchronization signal.

FIG. 4 shows a basic structure of a receiving device for an OFDM received signal. The

Received signal is amplified in an amplifier 10 and pre-filtered with a bandpass filter 11. In a mixer stage 12, to which a first reference frequency f _ref 1 is supplied, the received signal is mixed down to an intermediate frequency, then filtered again with a bandpass filter 13 and then divided into a branch stage 14. An output branch of branch stage 14 is connected to the input of a further mixer stage 15, to which a second reference signal f _{ref2 is} fed. The signal mixed into the baseband in this way reaches an OFDM demodulator 17 on the one hand via an analog-digital converter 16 and on the other hand 9 to a synchronization unit 18. The synchronization unit 18 is responsible for the block, clock and frequency synchronization. The functional sequences of the OFDM demodulator 17 and the synchronization unit 18 are controlled by a control unit 19.

The other output of the branching stage 14 reaches a detection device 20 for recognizing the frame synchronization signal. The detection device 20 generates an output signal with which the control unit 19 is informed of the fact and the time of the occurrence of a frame symbol. The control unit 19 then issues a command to the synchronization unit 18 to carry out the exact synchronization. When this is done, pass the

Synchronization unit 18 sends the determined data about frequency offset and block start to the OFDM demodulator 17, which then demodulates the signal and provides a received data sequence on the output side.

The analog implementation of the detection of the frame detection signal shown here has the advantage that it can be carried out completely independently of the remaining digital signal processing. The start of the frame is not recognized by active observation of the channel (as in the correlation). Rather, an event is triggered when the frame symbol occurs, i.e. the receiving device is passive and is notified by the frame synchronization signal.

Therefore, the method according to the invention can also be used to wake up the receiving device from a power saving mode, as a result of which energy-efficient mobile subscriber terminals can be implemented.

Figure 5 illustrates the structure of a detection device 20, the branched intermediate frequency signal 10 is fed to the output of branch stage 14. This input signal is divided into two branches, each of which has a bandpass 21, 22, a downstream squarer 23, 24 and a downstream low pass 25, 26. The bandpass filter 21 filters out the upper frequency band (p) and the bandpass filter 22 filters out the lower frequency band (n). Both filtered signal components are mixed linearly into the baseband by the squarers 23, 24 and filtered by a low-pass filter 25, 26. The resulting signal in each

Zweig is proportional to the received energy within a time window t ₀ , which can be variably set by the bandwidth of the low pass 25, 26. The signals s _p (t) and s _n (t) thus formed are compared in a comparison device 27. If the difference is sufficiently large, the reception of the frame synchronization signal is concluded. The detection device 20 'shown in FIG. 6 has the identical components 21 to 26 for generating the signals s _p (t) and s _n (t) and an identical comparison device 27 for the frame synchronization signal according to FIG. 3.

In addition, an addition stage 28 is also provided, in which the two signals s _p (t) and s _n (t) are added and are detected in an evaluation device 29 for recognizing a zero signal. If the sum at the output of the addition stage 28 is less than a threshold value slightly above the noise intensity, a conclusion can be drawn about a zero signal. When a frame detection signal is detected, the comparison stage 27 outputs a positive output signal, and when a zero signal is detected, the evaluation device 29 outputs a positive output signal. On the basis of a suitable delay in one of the output signals, an AND stage 30 can emit a frame detection signal d (t) to the control device 19. 11

By using second detection criteria for the beginning of a signal frame 1, the probability of incorrect detections is significantly reduced. The prerequisite is, of course, that the corresponding transmitter transmits a zero signal at the start of the signal frame 1 immediately before or after the frame detection signal according to FIG. 3.

The comparison device 27 compares the signals ^s _P (t) and s _n (t) and generates a corresponding output signal d (t).

According to Figure 7, the output signal d (t) with the condition

1 if _p (t) - s _n (t)> threshold d (t) = {0 otherwise

For this purpose, the inverted signal s _n (t) is fed to an addition stage 30 and the difference formed in this way is compared in a threshold value detector 31 with a set threshold value.

If the difference exceeds the set threshold of the threshold value detector 31, a potential characterizing the detection of the frame synchronization signal is generated at the output of the threshold value detector 31.

This arrangement has the least complexity, but has the disadvantage that the optimal threshold depends on the attenuation of the transmission signal.

This disadvantage can be avoided by the embodiment according to FIG. 8, in which the two energy signals s _p (t) and s _n (t) are divided by one another. The condition for the output signal d (t) is therefore 12

if .102ι ^'s " ^(t) > threshold

A _* M) Otherwise, the hardware is implemented in that the input signals s _p (t) and s _n (t) are each fed to a logarithmizer 32, 33 and then the difference between the logarithmic signals is formed in the addition stage 30. Mathematically, this corresponds to the formation of the logarithm of the quotient of the signals s _p (t) and s _n (t).

FIG. 9 shows a comparison stage 27 that matches the condition

^ s _p (t) -s _n (t)

1 if log d {t)> threshold s _p W + s _n (t), 0 otherwise

is working. For this purpose, the logarithmers 32, 33 are each preceded by an addition stage 34, 35, the addition stage 34 receiving the input signal s _p (t) and the inverted input signal s _n (t) and the addition stage 35 the input signals s _p (t) us s _n (t) are supplied without inversion. This realization of the comparison device 27 allows the variance of the detection time to be reduced compared to the method according to FIG. 8.

FIG. 10 shows a variant of FIG. 3 for the formation of the frame synchronization signal according to the invention. Here, too, the amplitudes of the subcarriers of the negative side (s) are zero. On the positive (p) side, however, only every second subcarrier is occupied with transmission energy, while the subcarriers in between likewise have an amplitude of zero.

This results in a periodicity in the time domain 13 over the frame symbol. This periodicity can be evaluated by a correlation and used for fine synchronization in the synchronization unit 18.

The frame synchronization signal according to FIG. 10 is also useful for performing a digital frame detection. The received signal is sampled and processed by a Fourier transform (FFT), with an FFT window of half the symbol length

(= N / 2) is used. This guarantees that at least one symbol section per transmitted OFDM symbol is free of interblock interference. By occupying only every second subcarrier within the positive (p) or negative (n) frequency range, the magnitude spectrum of the Rahra symbol becomes independent of the position of the time window.

A frame synchronization signal is detected precisely at block k, in which the variable H

Pt ^{• n} k <= ι <= - '

exceeds a predetermined threshold. R _i> denotes the complex output signal of the N / 2 FFT at the frequency iΔF (ΔF = subcarrier spacing) at the time kT / 2 (T = useful symbol duration of an OFDM symbol).

The frame synchronization method according to the invention is particularly suitable for OFDM transmission, which is also shown in the example, since the generation of the signal is much easier in OFDM than in single-carrier methods. In principle, however, it is also possible to use the method according to the invention with single-carrier transmission methods. In this case it is a good idea 14 to store samples of the time signal and read them out if necessary.

The intermediate frequency used in the receiving device should be as small as possible. This ensures that low-complexity band passes can be realized.

Although the frame synchronization signals shown in FIGS. 3 and 10 only have an occupied sub-area (p) and an unoccupied sub-area (s), it is fundamentally possible, for example, to divide the spectrum into four and to occupy the sub-bands differently. A further subdivision of the spectrum is theoretically conceivable, but should not be practical in general.

Even if the frame synchronization signal only contains power components in a subband, it can still be used as a reference signal for setting an amplitude gain in the receiver.

For this purpose, the frame synchronization signal should be selected so that the envelope has a course that is as constant as possible in the time domain. On the one hand, this is important in order to use the frame synchronization signal for setting the gain control, on the other hand, this prevents the transmitter amplifier from being overdriven.

The assignment of the individual subcarriers for the frame synchronization signal is independent of the type of modulation, since the frame synchronization signal is not demodulated. Any points in the signal space (complex plane) can therefore be selected.

In certain cases, it may prove expedient to also ensure the detection of uplink periods 3. One possible solution is to 15 Exchange of positive and negative sideband. For example, only the positive sideband (p) could be occupied for the detection of the start of a downlink phase 2, while only the negative sideband (n) would be occupied for the identification of the start of the uplink phase 3. For this purpose, one of the comparisons made by the comparison device 27 with logarithmic modules can take place. Only the sign of the signal present at the threshold value detector 31 changes.

To reduce the probability of incorrect detection, a combination of frame synchronization signal 1 - frame synchronization signal 2 - can be used as a variation on the arrangement in FIG. 6 instead of detecting a zero signal and a frame synchronization signal, in order to reduce the probability of incorrect detection. In this case, for example, all positive subcarrier frequencies can be occupied for the first frame detection signal and all negative subcarrier frequencies for the second frame detection signal.

The method according to the invention has been described for use in time division duplex (TDD) systems. However, it is also possible to apply the method to frequency division duplex (FDD) systems with corresponding modifications. This can be particularly advantageous for the implementation of current-blocking subscriber terminals.

Claims

16 Claims

1. Method for the digital transmission of data in a wireless communication network in which there is a controlling communication from a master station to all participants ("downlink") within a defined bandwidth of a channel ("downlink") with the aid of a defined signal frame (1) and the beginning of the signal frame ( 1) is characterized by a special frame synchronization signal (5) which can be detected by the subscribers, characterized in that the frame synchronization signal is occupied by at least a sub-area of the bandwidth with transmission signals and one

Not occupying at least one complementary portion of the bandwidth with transmission signals is used.

2. The method according to claim 1, characterized in that at least one further synchronization signal is used in addition to the frame synchronization signal.

3. The method according to claim 2, characterized in that the further synchronization signal is a zero signal.

4. The method according to claim 2, characterized in that the further synchronization signal from the 17

Frame synchronization signal is different, but is formed in the same way.

5. The method according to any one of claims 1 to 4, characterized in that the frame synchronization signal (5) has a maximum of two sub-areas occupied with transmission signals and a maximum of two sub-areas free of transmission signals.

6. The method according to claim 5, characterized in that the frame synchronization signal (5) has a sub-area occupied with transmit signals (p) and a sub-area free of transmit signals (n).

7. The method according to any one of claims 1 to 6, characterized by a uniform amplitude distribution in the sub-areas occupied by transmission signals (p).

8. The method according to any one of claims 1 to 7, in which the signal transmission takes place on a plurality of subcarriers evenly distributed over the bandwidth, characterized in that subcarriers lying in the occupied sub-region (p) are used with approximately the same amplitude.

9. The method according to claim 8, characterized in that in the occupied partial area (p) every second subcarrier is used.

10. Receiving device for receiving data transmitted according to the method of one of claims 1 to 9, characterized by a detection device (20, 20 ') with a filter arrangement (21, 22) for dividing the intermediate frequency band into partial areas and a comparison device (27) for Comparison of the received transmission energy in the 18th

Sub-areas.

11. Receiving device according to claim 10, characterized by an additional evaluation device (28, 29) for detecting a zero signal.

12. Receiving device according to claim 10 or 11, characterized by a power supply control of the receiving device, which in dependence on the detection of the frame synchronization signal by the

Detection device (27; 27, 29, 30) can be switched from a power saving mode to a full operating state.