GB2260049A - Receivers - Google Patents

Abstract

A receiver for detecting low signal to noise transmissions (eg LPI radar pulses) comprises a sampling circuit (4) for digitally sampling a received signal, a divider circuit (6) for dividing the sampled signals into a plurality of successive sample blocks, a fast Fourier transformation circuit (10) for providing, for each of the sample blocks, transformation signals indicative of the frequency spectrum of each sample block, a tracking circuit (12) for tracking a peak amplitude of the transformation signals for the successive sample blocks, and a filter circuit having a passband characteristic controlled to match the frequency of the tracked peak amplitude of the transformation signals. <IMAGE>

Description

IMPROVEMENTS IN OR RELATING TO RECEIVERS The present invention relates to receivers, and in particular, to receivers for the detection of low power radar signals having a low signal to noise ratio.

In modern radar systems, the radar transmitter can be arranged to transmit radar pulses of relatively long duration and at very low power levels. The transmitted pulses have, therefore a very low signal to noise ratio and, hence, the transmitted pulses are coded, usually using chirp or biphase chip coding, to assist detection of the radar return by a receiver of the radar system. These low power pulses and the coding also render the radar transmissions extremely difficult to detect by third parties. Such transmissions are commonly known, therefore, as low probability of intercept (LPI) radar pulses. The receivers used by LPI radars have filter circuits matched to the transmission frequencies which provide sufficient process gain to incoming return pulses that have the correct code, thereby enabling detection of the low power return signals in the noisy signal environment.Hitherto, such LPI transmissions have therefore, to a large extent, gone undetected by third parties.

The present invention seeks to provide an improved form of receiver which enables low signal to noise transmissions, such as LPI radar pulses, to be readily detected.

Accordingly, there is provided a receiver comprising means for sampling a received signal and for producing therefrom a plurality of blocks of samples, first transformation means for deriving from the sample blocks transformation signals indicative of the frequency spectra thereof, tracking means for tracking a peak amplitude of the transformation signals and filter means having a passband characteristic controlled in dependence upon the tracked peak amplitude.

Thus a receiver according to the invention may comprise sampling means for digitally sampling a received signal, divider means for dividing the sampled signals into a plurality of successive sample blocks, first transformation means for providing, for each of the sample blocks, transformation signals indicative of the frequency spectrum of each sample block, tracking means for tracking a peak amplitude of the transformation signals for the successive sample blocks, and filter means having a passband characteristic controlled in dependence upon the tracked peak amplitude of the transformation signals.

The tracking means may be arranged to track the peak amplitude of the transformation signal in dependence upon the frequency of a previously detected peak amplitude.

Preferably the receiver comprises a further transformation means, responsive to output signals from the filter means, for providing further signals related to the received signals and having a signal to noise ratio higher than that of the received signals.

The sampling means may comprise a digital radio frequency memory for providing in-phase and quadrature samples from the received signals. The receiver further may comprise means for combining the further signals with the sampled signals thereby to effect a phase correction to the sampled signals.

Advantageously, the first and further transformation means comprise fast fourier transform circuits.

The receiver may also comprise a memory for storing the sampled signals.

The present invention will now be described, by way of example only, with reference to the accompanying drawings, in which Figure 1 shows a schematic block diagram of a receiver circuit in accordance with a preferred embodiment of the present invention.

Figure 2 shows the frequency spectra of two successive sample blocks, and Figure 3 shows the frequency spectra of Figure 2 after filtering.

Referring to Figure 1, a receiver 2 comprises a sampling circuit 4, such as a digital radio frequency memory (DRFM), for digitally sampling a pulsed analogue incoming or received signal having a low signal to noise ratio. The sampling circuit 4 is arranged to produce in-phase (I) and quadrature (Q) digital samples which are fed to a memory circuit 6. In the embodiment shown 2048 samples are provided from each pulse of the received signal but it should be realised that other numbers of samples may equally be provided.

A divider circuit 8 is coupled to the memory circuit 6 for dividing the 2048 I and Q digital samples into successive sample blocks, which in the embodiment shown are 32 samples in length.

The sample blocks are afforded to a first transformation circuit, such as a fast Fourier Transform (FFT) circuit 10, so as to provide transformation signals which are indicative of the frequency spectrum of the I and Q digital samples in each block.

The digital samples are, therefore, converted from the time domain into the frequency domain by the FFT circuit 10.

Chirp and biphase chip coding, the two commonest forms of LPI pulse coding, both exhibit readily identifiable characteristics in the frequency domain, and it is this characteristic pattern which the receiver attempts to detect.

The digital samples, converted to the frequency domain are therefore fed from the FFT circuit 10 to a tracking filter circuit 12. The filter circuit 12 searches for the peak amplitude of the frequency domain samples and, once the peak is located, an appropriate bandwidth os passband characteristic is selected and applied to the spectral data received from the FFT circuit 10. To avoid any spurious peaks in the frequency spectrum due to noise components, the search for any peak amplitude is, preferably, limited to a small band of frequencies around the previously detected peak. The bandwidth chosen for the filter will be determined by the chirp rates used for the LPI signals, as will the sampling rate.Provided the incoming signal is sampled at a high enough rate the frequency change across successive blocks of samples will be small, allowing the bandwidth of the tracking filter 12 to be minimised.

The filter cicuit 12 has the effect of reducing the noise component in the signal. After filtering, the sample blocks in the frequency domain are transformed back into the time domain by a further transformation circuit, in the form of FFT circuit 14.

The 'cleaned up' samples, which have an improved signal to noise ratio compared to the original data, are then normalised and conjugated in circuit 16 to provide phase inverted samples. The above process is repeated until all 2048 samples have been adaptively filtered in the sample blocks of 32 sample length.

The 'clean' samples are then combined with the original samples in multiplier 18, thereby to provide a phase correction to the original data samples. Combining the 'cleaned' samples with the original samples in multiplier 18 has the effect of rotating the phase of any data within any pulse towards zero, but the phase of the remainder of the signal is unaffected. The resultant output signal from the multiplier 18 has a signal to noise ratio substantially higher than the received signal, and can be fed to a coherent detector.

It will be appreciated that the tracking filter 12 performs two functions, searching for the peak amplitude of the transformed sample block, and thereafter filtering dependent on the frequency of the peak amplitude. It acts as a filter with a fixed bandwidth (determined by the specific application) but with variable centre frequency. The centre frequency of the filter is determined by the frequency spectrum of each block of data using a maximum search algorithm. In the preferred technique, a maximum initially is found by sweeping the full spectrum of the first sample block, and thereafter the search for the maximum in a block of data is confined to a predefined range either side of the position of the maximum in the previous block of data. The width of this range is determined by the parameters of the signals which are expected.Once the new maximum has been found, a filter is applied to reduce the high and low frequency components. In a digital system, such as the one described, this filtering will consist of a simple multiplication.

Figure 2 shows the frequency spectra of two successive blocks of samples from the incoming signal. In this case, the sample blocks are for simplicity shown as 16 samples long. The peak has already been found in block m and is in the 4th frequency bin.

In order to find the peak in block m+l, a maximum search is conducted starting 3 frequency bins below the maximum in block m, and extending to 3 frequncy bins above the previous maximum (i.e.

from bin 1 to bin 7). The maximum in this region is in bin 6.

Note that the spurious peak (assumed to be due to noise) in bin 10 is not captured by this process.

Now the filter function is applied. In this case it is a simple rectangular function of width 5 bins. This is centred on the maximum in each block of spectral data. The filter coefficients are shown as zeros and ones underneath the spectra in Figure 1.

After multiplication the filtered spectra are as shown in Figure 2, and are then ready to be transformed back into the time domain to form a processed copy of the incoming signal, but with an improved signal to noise ratio.

Although the present invention has been described with to a specific embodiment, it is to be realised that modifications may be effected within the scope of the invention. For example, sampling circuits other than a digital radio frequency memory may be utilised and the number of samples taken for each pulse of the received signal, as well as the data length of the sample blocks can be chosen to split the characteristics of the LPI pulses to be detected.

Claims (10)

1. A receiver comprising means for sampling a received signal and for producing therefrom a plurality of blocks of samples, first transformation means for deriving from the sample blocks transformation signals indicative of the frequency spectra thereof, tracking means for tracking a peak amplitude of the transformation signals and filter means having a passband characteristic controlled in dependence upon the tracked peak amplitude.

2. A receiver comprising sampling means for digitally sampling a received signal, divider means for dividing the sampled signals into a plurality of successive sample blocks, first transformation means for providing, for each of the sample blocks, transformation signals indicative of the frequency spectrum of each sample block, tracking means for tracking a peak amplitude of the transformation signals for the successive sample blocks, and filter means having a passband characteristic controlled in dependence upon the tracked peak amplitude of the transformation signals.

3. A receiver as claimed in claim 1 or claim 2 wherein the tracking means is arranged to track the peak amplitude of a transformation signal in dependence upon the frequency of a previously detected peak amplitude.

4. A receiver as claimed in any preceding claim comprising further transformation means, responsive to output signals from the filter means, for providing further signals, related to the received signals and having a signal to noise ratio higher than that of the received signals.

5. A receiver as claimed in any preceding claim wherein the sampling means comprises a digital radio frequency memory for providing in-phase and quadrature samples from the received signals.

6. A receiver as claimed in any preceding claim comprising means for cbmbining the further signals with the sampled signals thereby to effect a phase correction to the sampled signals.

7. A receiver as claimed in any preceding claim wherein the first and further transformation means comprise fast fourier transform circuits.

8. A receiver as climed in any preceding claim comprising a memory for storing the sampled signals.

9. A receiver substantially as herein described with reference to the accompanying drawings.

10. Any novel feature or combination of features herein disclosed and not otherwise claimed.