WO1994007315A1

WO1994007315A1 - Digital common wave radio transmission system with a means for avoiding stationary cancellation areas

Info

Publication number: WO1994007315A1
Application number: PCT/GB1993/001955
Authority: WO
Inventors: John Sykes; Simon Parnall
Original assignee: British Broadcasting Corporation
Priority date: 1992-09-15
Filing date: 1993-09-15
Publication date: 1994-03-31
Also published as: AU4825093A; GB9219466D0

Abstract

A radio transmitter receives digital samples of a signal to be transmitted. The bits of each sample are each assigned to one of a plurality of carriers. These carriers are then transmitted as a burst signal and the burst signals making up the transmission are perturbated for avoiding stationary cancellation areas in the reception zone.

Description

DIGITAL COMMON WAVE RADIO TRANSMISSION SYSTEM WITH A MEANS FOR AVOIDING STATIONARY CANCELLATION AREAS

This invention relates to radio transmission systems and in particular to digital transmission systems.

In a conventional FM transmission system a plurality of transmitters are spaced throughout the area to which a radio signal is to be distributed. If all transmitters were to operate at the same frequency then there would be interference in intermediate or "mush" areas between transmitters. This would result in there being no signal available in particular locations. For this reason the system is arranged that adjacent transmitters transmit the same radio programme on different frequencies so that interference in the mush areas is avoided.

A European consortium known as Eureka 147 has proposed a digital audio broadcast (DAB) transmission system. This operates on what is known as a coded orthogonal frequency division multiplexed (COFDM) method. This is a broad band transmission system and its operation is illustrated with reference to Figure I. In Figure 1 a single 16 bit sample taken at an instant in the radio programme which is to be transmitted is shown at a) . This will be produced from the radio programme by an analogue to digital converter and a large number of such samples will be generated every second. From this 16 bit sample a 4 bit error code is generated and this is shown at b) . Each bit of the error code is generated by performing a logical operation on a plurality of the bits in the 16 bit sample. Thus the error code can be used, after reception of the 16 bit sample and the error code, to check whether the sample has been received correctly. This type of error correction code is well known.

In the COFDM transmission method the bits of the sample and its associated error code are distributed over a plurality of carriers by quadrature phase shift keying or some other form of quadrature amplitude modulation. The Fast Fourier Transform of all these cariers is then taken and the resultant spectrum modulated by a further carrier and transmitted as a burst signal and the reverse FFT taken to give the original carriers and their respective amplitudes and phases. The bits cf each sample may also be distributed between transmission bursts. An example of the distribution of a single sample between carriers and transmission bursts (prior to FFT) is shown below the sample in Figure 1. This comprises transmission bursts at times to, tl, t2, t3 , and t4. Each transmission burst occupies a bandwidth F the limits of which are illustrated by the dotted lines. Using QAM more than one bit of the 16 bit sample and the associated error code is assigned to each carrier in one of the bursts. These are identified as F01 to F44. Each transmission burst will also include a plurality of other carriers carrying bits from other samples and their associated error codes. Adjacent carriers in each burst are mathematically orthogonal and do not interfere with each other.

Because of the broad band nature of the COFDM system each transmitter uses the same transmission frequency. Because of this there will be interference between signals from adjacent transmitters at particular frequencies within the bandwidth F. Because of the error correction code it should, in most instances, be possible to reconstruct the whole of a sample even if some bits are lost due to interference. This will not, however, always be the case. In particular in built-up areas the interference patterns will be made worse by reflections of radio waves from buildings etc. This will result in more pieces of data being lost due to interference at particular locations.

As the radio transmitters are transmitting at the same frequencies the interference patterns set up will be stationary. Thus there will be some locations where reception is good and others where it is unacceptably bad.

The same type of problem can occur with individual FM transmitters where there are reflections from buildings etc. For example, it is often found that an FM receiver will work in certain locations in a house but if moved a few feet will receive no signal whatsoever. This is due to an interference node set up by reflections of radio waves. All the radio waves are produced by the same transmitter and are thus at the same frequency and so the node is stationary. The present invention seeks to overcome the problem of standing interference patterns being set up with and is particularly applicable to DAB COFDM transmitters operating as part of a network. The invention may also be applied to television signals transmitted using some form of COFDM.

The invention is defined in appended claims to which reference should now be made.

The invention will now be described in detail by way of example with reference to the further accompanying drawings in which:

Figure 2 shows a transmission pattern between two adjacent receivers; and

Figures 3a and 3b show the amplitudes of received signals at a destructive interference node in the system of Figure 2 in accordance with an embodiment of the invention.

In Figure 2 the transmission and interference pattern which is generated by a pair of transmitters A and B is shown. In the shaded region there will be both constructive and destructive interference. Thus some locations in the shaded region will receive no signal whilst those at other locations will receive an enhanced signal.

A receiver, such as a car radio receiver, moving on trajectory X will move from a location where it receives predominantly the signals from transmitter A through constructive and destructive interference in the shaded region to a location where it will predominantly receive the signal from transmitter B. Thus if transmitters A and B are transmitting a COFDM DAB signal the receiver moving on trajectory X will at some times be receiving a signal in which some of the data is lost due to interference nodes at particular locations and frequencies within the broad band COFDM signal and at other times it will be receiving the whole signal. Thus the data it loses varies with time and is less than would be lost if the receiver were positioned at a destructive interference node. For this reason the chances of the error correction routine breaking down when applying the error code to a received sample is considerably reduced.

We have appreciated that it is possible to simulate the type of reception obtained by a receiver moving on trajectory X at a fixed location receiver. This is achieved by placing a perturbation on the signal produced by one of the transmitters. This may be, for example, an amplitude perturbation or a phase perturbation.

In the case of an amplitude perturbation the amplitude of alternate transmission bursts is amplified by a factor of, for example, 1.5. The effect of this on a destructive interference node will be to produce a signal of 50% of the maximum peak to peak amplitude on alternate frequency bursts at that node. Thus on a first frequency burst without the amplification the signals from transmitters A and B will cancel each other out but on the following transmission burst there will be either a peak or a trough of 50% of the transmission strength which would be received from a single transmitter at that point.

At nodes where there is constructive interference there will be a further enhanced signal by addition of the amplified signal from one of the transmitters.

Figure 3a) shows how the signals from transmitters A and B add destructively on a first transmission burst and Figure 3b) shows how they add on a second transmission burst if the amplitude of the signal from transmitter A is not amplified by a factor of 1.5. This gives a resultant amplitude of 50% of the usual maximum.

An alternative form of perturbation can be put upon the signal from one of the transmitters by varying the lengths of the synchronising intervals between transmission bursts. ' This has the effect of introducing a phase shift between the transmission bursts from one transmitter and the corresponding transmission bursts from the other transmitter. The effect of this at a fixed location in the shaded region of Figure 2 will be to move interference nodes between transmission bursts. Thus a half wave length difference in transmission bursts produced by transmission A on alternate bursts would result in a destructive interference node moving half a wave length between transmission bursts. A wavelength is typically of the order of a couple of feet and this would be sufficient to move a destructive interference node away from a fixed radio receiver for alternate transmission bursts.

It will be appreciated that putting a perturbation on the signals transmitted by one receiver results in some of the data which would have been lost at a destructive interference node being recovered at that node. Thus the chances of actually losing so much data that the error correction routines employed cannot reconstruct a sample are considerably reduced and reception for fixed receivers in the shaded region of Figure 2' is improved. Perturbation is similar to the effect of moving the receiver along trajectory X in Figure 2.

Claims

CLAIMS :

1. A radio transmitter comprising means for supplying digital samples of a signal to be transmitted, means for distributing the bits of each digital sample over a plurality of carriers, means for transmitting the carriers as a burst signal, and means for perturbating the burst signals making up the transmission.

2. A radio transmitter according to claim 1 in which bits of each digital sample are distributed over carriers in more than one burst signal.

3. A radio transmitter according to claim 1 in which the perturbating means comprising an amplitude perturbating means.

4. A radio transmitter according to claim 1 in which the perturbating means comprises a phase perturbating means.

5. A radio transmitter according to claim 1, 2, 3 or 4 in which the signal to be transmitted is an audio broadcast signal.

6. A radio transmitter according to any preceding claim in which adjacent carriers in signal are orthogonal.

7. A radio transmission system comprising at least one radio transmitter, the transmitter comprising means for supplying digital samples of a signal to be transitted, means for distributing the bits of each digital sample over a plurality of carriers, means fo transmitting the carriers as a burst signal, and means for perturbating the signal to be transmitted.

8. A method for transmitting a radio signal comprising the steps of providing digital samples of the signal to be transmitted, distributing the bits ot each digital sample over a plurality of carriers, transmitting the carriers as a burst signal, and perturbating burst signals making up the transmission.

9. A method for transmitting a radio signal according to claim 8 in which the distributing step comprises assigning bits of each digital sample to carriers in more than one burst signal.

10. A method for transmitting a radio signal according to claim 8 or 9 in which the perturbating step comprises perturbating the amplitude of the burst signals.

11. A method according to claim 8 or 9 in which the perturbating step comprises perturbating the pahse of the burst signals.