FR2853169A1 - Coded signal demodulating device for use in a vehicle alarm, has phase comparator that compares phases of output signals of filters to deduce coded signal by frequency shift keying and provide demodulated signal - Google Patents

Abstract

The device has two filters for receiving a demodulation signal and providing respective output signals, where the output signals are in out of phase. A phase comparator compares the phases to deduce a coded signal by frequency shift keying (FSK) and provide a demodulated signal. An extraction module extracts a FSK message and provides a demodulated FSK message.

Description

IMPROVEMENTS RELATING TO REDUCING INTERFERENCE FOR

  WIRELESS AND DEMODULATED RECEPTION OF A FREQUENCY HOPPING SIGNAL

  The present invention relates to techniques for use in processing the reception of a communications signal. The invention may be particularly suitable for processing a received wireless signal (e.g. a remote lock / unlock signal for a vehicle or a vehicle alarm), but the invention is not limited to a wireless environment thread. More specifically, the invention relates to the reduction of interference effects in a received signal and to a technique for processing and demodulating the received signal encoded by frequency shift coding modulation (FSK).

  In FSK modulation, bits or sequences of bits are represented by signals of at least two different frequencies. Due to technical constraints, the modulation properties are not always stable over time. For example, modulation properties can be affected by temperature variations, power variations, component tolerances, and relative movement between the transmitter and the receiver. In particular, in the case for example of a vehicle security system, the transmitter is typically mounted in a small keychain or a handful of keys, and constitutes a miniature circuit at low cost (poor tolerances), powered by a miniature battery. With such a transmitter, it is difficult to accurately predict the frequencies that will be generated by the transmitter. Frequencies may also fluctuate during transmission. Likewise, the times at which FSK modulated signals will be transmitted may be unknown at the receiver.

  Non-coherent techniques are known for demodulating such unpredictable signals with FSK modulation.

  However, just as they present a high computational overload, conventional non-coherent techniques are extremely vulnerable to the effects of interference from other signals in the same frequency range as the signals of interest with FSK modulation. Non-coherent techniques can be particularly affected by coherent interference in the form of one or more relatively stable interference frequencies.

  Such coherent interference can, for example, be generated by electrical lighting or by electronic equipment, for example digital circuits having a stable clock frequency. In an automotive environment, there are many independent circuits which can produce such interference at a frequency close to the modulation considered.

  In addition, interference may be caused by frequency sweeping interference. Indeed, industrial pollution such as the starting of the motors generate disturbances in the frequency band of the FSK signal which make it more difficult to demodulate the signal.

  Techniques are known to reduce the effects of noise and interference. However, the most effective of these techniques relies on a reference source of the noise or interference to be suppressed. For example, the reference source may be a second receiver at a location different from a first receiver to provide a different received signal in space. However, such a second receiver 25 adds significant cost and it is often highly inconvenient to implement. The reference source can alternatively take the form of a preprogrammed signal generator or a mathematical modeling. However, the use of a preprogrammed reference assumes prior knowledge of the interference. Similarly, a preprogrammable reference is not adaptable to the modification of conditions or even to a range of different conditions.

  In summary, there remain many significant problems in providing a technique which can work with conventional high dispersion transmitters, and which can provide efficient processing and demodulation at low cost and yet robust FSK modulation signals, and also provide relatively high immunity to interference in the frequency band of the signal of interest.

  A first aspect of the invention relates to the reduction of interference interference by frequency sweeping in a received communications signal. The first aspect generally includes filtering of the received signal using an adaptive filter.

  Such a technique can allow extremely efficient and adaptive suppression of interference due to frequency sweeping interference, which has a coherence time of duration longer than the signal of interest.

  The filter can be a subtraction filter for removing one or more components obtained from a reference signal from the received signal. The reference signal can be obtained from the received signal. The reference signal can be a time-delayed signal obtained from the received signal. The filter can be of the Wiener type.

  The first aspect of the invention above may be particularly suitable for use, but not limited to, with an FSK modulated signal. However, the first aspect can be used with any type of information modulation or coding, to eliminate interference due to existing frequency sweep interference.

  Another aspect of the invention relates to a technique for processing and / or demodulating in an inconsistent manner the signal with FSK modulation. This aspect generally includes the demodulation of the signal of interest, in which non-coherent interference has been suppressed as described above, by double filtering the signal autocorrelation and using phase dynamics related. The group delays of the two filters are different and chosen as a function of the general characteristics of the expected signal of interest (sampling frequency and bit period). These filtered signals are then compared to deduce the FSK message.

  The invention therefore relates to a device intended for demodulating frequency shift coded components (FSK) of a received communication signal, the device comprising an estimator of the correlation of the received signal providing a demodulation signal, a first filter intended receiving said demodulation signal and outputting a first phase signal, a second filter causing a different signal integration time through the filter than said first filter and intended to receive said demodulation signal and to output a second phase signal, said second phase signal being out of phase with respect to said first phase signal, a phase comparator for comparing the phases of said first and second phase signals to deduce therefrom the components coded by frequency displacement (FSK ) of said communication signal received and providing a demodulated signal at output, and an extraction module d u FSK message which receives the demodulated signal and which provides the demodulated FSK message.

  Other characteristics and advantages of the invention will emerge during the description which follows. A preferred and nonlimiting embodiment of the invention is now described, by way of example only, with reference to the accompanying drawings in which FIG. 1 is a simplified diagram showing the principles of an interference suppressor provided with 'a filter 25 from Wiener.

  Figure 2 is a simplified block diagram showing a signal processor of a communications receiver.

  FIG. 3 is a simplified block diagram representing stages of information processing in a first conditioning section of the signal processor of FIG. 2.

  FIG. 4 is a simplified block diagram representing the stages of information processing.

  FIG. 5 is an example of the properties of the low pass filters (power spectral density and group delay) allowing demodulation.

  Figure 6 is an example of demodulation results using the filters of Figure 5.

  Before describing in detail the entire preferred embodiment, the principles of an interference suppressor are first briefly described with reference to Figure 1. The interference suppressor includes an adaptive filter, for example a filter Wiener. This Figure 1 illustrates the principles of using the suppressor including a Wiener filter to cancel interference in a received signal, based on a noise reference. The noise reference can be obtained from the received signal, delayed by an appropriate delay, preferably Z-1. The principle of such a Wiener suppressor consists in estimating a linear filter which allows the generation of a phased signal which can be subtracted from the signal received by a subtractor. After the subtraction, only the components of the signal that are not consistent with the noise reference remain. This signal is used as the error signal of the adaptive filter, while the signal leaving the filter is used as an output signal which is suitable for demodulation. The principle of this filter is to identify and select a linearly predictable copy of the input signal. The non-linearly predictable components are therefore removed by the filter.

  It is thus understood that if the input signal does not include non-linearly predictable components, it is not necessary to use such an adaptive filter. The signal received without nonlinearly predictable components can in fact be directly demodulated, as will be explained in more detail below.

  Thus, the interference suppressor generally comprises an adaptive filter 22 receiving an input signal 24 and an error signal 26, and generating an output signal 28. By way of example, the adaptive filter 22 can be of the subtraction type, for example a Wiener filter. The filter 22 can be realized digitally and can operate after sampling to process a sequence of digitized samples of the signal. The filter 22 comprises a first signal line 30, a second signal line 32 comprising a delay (Z'1), and a subtractor 34, in order to subtractively combine the components of the signal coming from the first and from the second line signal 30 and 32. The filter 22 further comprises a coefficient estimator 36 sensitive to the error signal 26 in order to dynamically update the coefficients of the filter. The dynamic updating of the coefficients is dimensioned as a function of the characteristics of the modulation of interest. Thus one can tend towards the suppression of all the non-linearly predictable components of the signal at the output 28. It will be appreciated that the principles described can be applied to any adaptive filter.

  The error signal 26 controls the adaptation of the filter 22 in which the estimator 36 operates to update the coefficients. In general, the estimator 36 continuously updates the coefficients of the filter to compensate for any variation, creation or dispersion of the components of the signal. The filter is therefore effective in continuing to subtract all the non-linearly predictable components while retaining the signal of interest to be demodulated which is linearly predictable.

  The principle of operation of this filter is to allow the generation of a signal comprising only the predictable components on the output line 28. In fact, it is assumed that the interference due to interference by frequency sweeping changes quickly and is not -linear, and that the signal FSK, that is to say the signal of interest as well as other components are more stable and linear.

  By subtracting the outgoing signal from the adaptive filter from the input signal, rapid non-linear interference can thus be eliminated.

  The suppressor may be continuously functional, or it may be functional only at periodic intervals, for example if the circuit containing the suppressor is only active periodically to decrease the consumption of electrical power. For example, a periodically activated state can be used for applications in vehicles where the circuit is powered from a battery. In such a state, it is preferred that the filter 22 has an adaptation response time less than the time during which the circuit is activated at each activation cycle. Such a response time can allow the filter 22 to adaptively eliminate new interference detected at the start of an activation cycle. For example, the filter may have an adaptation response time less than about one-tenth of the time that the circuit is active.

  The above principles make it possible to provide an extremely effective technique for reducing interference due to frequency sweeping interference, such as, for example, sinusoidal interference signals.

  The FSK modulation frequencies of an FSK modulation input signal can be generally defined by F, + / Ft, where F, is a center or carrier frequency, and 2Ft is the frequency difference between the frequencies.

  The central or carrier frequency F can be 314 MHz for Japan or 433 MHz for Europe. Ft can be around 30 kHz, so the difference between the two FSK modulation frequencies is around 60 kHz.

  Referring to Figure 2, a signal processor 60 is illustrated for processing a received communications signal with FSK modulation 62 in a receiver. The receiver can be a wireless receiver, for example a radio, microwave or infrared receiver. An application of this embodiment resides in the field of remote control of security systems, for example for vehicle security systems (for example an alarm, a immobilizer and / or door locking) or building security systems (eg an alarm and / or door locks). The signal processor 60 can be produced in hardware form, or in the form of software executed on a processor, or else in a mixture of hardware and software. The signal processor may include a mixture of analog and digital processing circuits or may be fully analog.

  The signal processor 60 can generally comprise a first preconditioning section of the signal A intended to filter by bandpass filtering the received signal 62 and to convert the frequency of the received signal into a complex baseband, a second adaptive filtering section C for eliminating signals from scrambling interference by frequency sweeping, and a third demodulation section B for demodulating the resulting signal to obtain information relating to a signal of interesting FSK modulation.

  The first section A will be explained in more detail with reference to FIG. 3. In the first section A, the received signal 62 can first of all be filtered by a bandpass filter by means of a first analog bandpass filter 70. D typically, the width of the passband of the filter 70 is about 10% of the carrier frequency F i. After bandpass filtering, the signal is supplied to a frequency mixer 72 for frequency change to an intermediate frequency for subsequent filtering by a second analog bandpass filter 74. The intermediate frequency can be typically 10.7 MHz, since a wide range of filter circuits 74 is currently already available based on the 10.7 MHz intermediate frequency standard. The signal filtered by a resulting bandpass filter can typically have a bandwidth of about 600 kHz. It may be difficult to obtain a narrower bandwidth using other analog filters. For this reason, in the present embodiment, the signal is digitized by a digitizing stage 78, for further digital processing. For example, the signal can be sampled at a sampling frequency, for example of about 1.3 MHz (about twice the bandwidth of the signal).

  The digitized signal is then changed in frequency to the baseband by the stage 79, and filtered by a digital filter stage 80, which can further reduce the bandwidth, for example to around 120 to 130 kHz. Finally, the signal is decimated by a subsampler stage 84, for example, by a factor of around 5 or even 10.

  Such a narrow bandwidth makes it possible to provide a high degree of noise rejection, so that subsequent interference suppression techniques can be used more effectively to reduce interference from interference by frequency sweeping, and relatively inexpensive techniques. in terms of computing power can be used for demodulation of the FSK modulation frequencies. In general, the cost and complexity of digital processing circuits depend on the sampling frequency used.

  The use of a baseband signal provides significant advantages in reducing the cost and complexity of the signal processor 60.

  It will be appreciated that the above described embodiment of signal conditioning is merely an example, and that many other bandwidth conditioning and / or filtering and / or conversion techniques frequency can be used.

  Referring to Figure 2, the second section C processes the output signal from the first section A in the complex baseband to eliminate non-linear interference components that are not recognized as an interesting signal. The second section C is based on a Wiener filter, having a signal input 90 and a reference input 92 generated by means of a delay 94. The filter can be similar to that already described with reference to the figure 1. Therefore, the adaptive filter 25 of the second section C receives as an error signal the output signal from the subtractor, and the suppressor 2d actively adapts to cancel the components of a preexisting interference, as described below. -above.

  The third section B includes a demodulation section 100 (see Figure 4). In the demodulation section 100, the demodulation is based on the dynamics of the apparent phase of the first correlation coefficient.

  As shown in FIG. 4, the third demodulation section B comprises a first filter 42, and a second filter 44, and receives the signal 28 as an input. This input signal first passes through a module of estimation of the correlation function 41, and is then filtered by a low-pass filter 43 and decimated by a decimation module 45 to prepare the signal for demodulation. The signal leaving the decimator 45 is therefore the signal containing only the signal of interest (or interesting signal), and linear components, but not the non-linear components.

  As already mentioned above, if the received signal does not contain non-linearly predictable components, it is not necessary to use the adaptive filter. The signal received without nonlinearly predictable components is therefore considered the signal of interest and this signal can pass directly into the demodulation section.

  Each filter 42, 44 has a different group delay, as shown by an example in FIG. 5. The signal to be demodulated passes through each filter 42, 44 with a different integration time. The first filter 42 provides an output signal S1 and the second filter 44 provides an output signal S2. Then, these signals Si and S2 are supplied to a phase comparator 47 to deduce the message FSK by a simple comparison. For example, a decision rule could be as follows: If the phase of S1> the phase of S2, the FSK message is +1, if not, the FSK message is -1.

  Therefore, the FSK message can be extracted in a message extraction module 48 to obtain the demodulated FSK message. Figure 6 shows an example of demodulation according to this rule.

  Of course, the filters used 42, 44 can be low-pass or high-pass filters. Depending on the filter used, an additional phase shift of IF may be necessary, as will be understood by a person skilled in the art.

  To further detail, the baseband signal after filtering and decimation can be modeled by S (, "(1) = FSK (t) + iJammer (t) + Noise (t) The modeling of the FSK signal can be expressed by FSK (t) = AAX (t - kT) exp (27 (.f - fk> - 29k) k = - l1 OA: the amplitude of the signal.

  Xl (t): a niche function which corresponds to: X1 (t) = 1 for t e] O, T] and Xl (t) = O for t] O, T]; fc: the carrier frequency; {fk}: the FSK binary message using a two-tone signal with the offset frequencies: f + 1 and f-1; f + 1 -f-1 represents the bandwidth of the FSK signal T: the duration of transmission for each bit; (Pk: a set of variables randomly distributed between O and 2n.

  The component of interference at stable frequencies (commonly called by the English term "jammer") is represented by: Janinie (t) = J, exp (2tf + 0,) The noise component (called in the formula "Noise (t)" from the English term "noise" commonly used to indicate noise) is assumed to correspond to Gaussian noise. 25 Thus, we can determine the first correlation coefficient: r, (t1) E {S (.,. "(T) S ..,) (t - I)} E {[FSK (t) + Jammer (t ) + Noise (t)] [FSK (t- 1) + Jammer '(t- 1) + Noise (t -1)]} 30 E {FSK (t) FSK * (t - I)} + E {Jammer (t) Janiner (t - I)} + E {Noise (t) Noise (t I)} The correlation coefficient has 3 separate parts, because each component does not share any information with another component and these components can therefore be 35 considered orthogonal components.

  E {FSK (t) FSK * (t - 1)} A2 XI, (t - kT) E {exp (2. (f + k) t - fk) exp (- 2. (f + k Xt - 1) + k)} AZX (t - kT) exp (2 (f0 + f)) For each FSK modulation component, the correlation function is a constant complex number during a Tchip period (the standard period or cycle of the FSK modulation signal) .

  For the sake of simplification, Tchip is assumed to be greater than 1. As there is no interference between the two components of FSK modulation, it is possible to factorize: E {. Jamnier (t) Jamnier (t - I)} = J, exp (2ift. + 0,) J, exp (-2.rf (tl) -,) = E J2E {exp (2qt + O,) exp (- 2rf (t 1) -, )} = E J2exp (2, f) =,. ,, Since the different components of the interference component at stable frequencies are orthogonal, only the first correlation coefficients of each interference can be retained, normalized in amplitude. The summation does not depend on time, but is a constant complex value. Depending on the phase, the components of interference at stable frequencies are not necessarily added.

  E {Noise (t), Noise (t - I)} = O The noise contribution can be ignored because this component is supposed to correspond to the Gaussian noise.

  Thus: F (t) =. ,,,, I ,, +. A2 EZ (t-kT) exp (2 (f .. -f)) k = - The first correlation coefficient is a complex signal which can change each Tchip. Its bandwidth is very reduced.

  Therefore, the FSK message to be demodulated can be obtained by a relative measurement of the coefficient Fr (t, 1) as described previously using the two filter application.

  Advantageously, in the Wiener filter described in FIG. 1, filters with finite impulse response (FIR) or filters with infinite impulse response (IIR) can be used for the linear filter. FIR type filters have an advantage in terms of stability. However, certain limitations of FIR response filters can result in additional processing complexity and cost. In order to be able to cancel the effect of interference close to an interesting signal, the FIR response filter must recover the phase of each interference component. To do this, the number of coefficients of the FIR response filter should be increased to obtain the appropriate degree of freedom for the filter. For example, in order to cancel up to 5 interference components, the minimum number of theoretical filter coefficients is 10. In practice, due to noise estimation, this number should be increased, usually up to to about 30. The number of coefficients can be directly associated with the complexity and cost of processing, and should preferably be as small as possible to reduce the computational load.

  In general, an IIR response filter can provide a more efficient embodiment with greater flexibility. In order to avoid stability problems, it may be preferred that the IIR response filter is made in a series-parallel manner.

  When the filter arrangement is used in an intermittently activated power saving mode, the coefficients from the previous activation cycle are kept as a history of the interference components and are used as coefficients initials for the next activation cycle.

Claims (9)

  1. Device (100) for demodulating frequency shift coded components (FSK) of a received communication signal, the device comprising: an estimator (41) of the correlation of the received signal providing a demodulation signal, a first filter (42) intended to receive said demodulation signal and to output a first phase signal (sl), a second filter (44) causing a different signal integration time through the filter than said first filter and for receiving said demodulation signal and outputting a second phase signal (s2), said second phase signal being out of phase with said first phase signal, a phase comparator (47) for comparing the phases of said first and second phase signals (si, s2) to deduce the frequency shift coded components (FSK) of said received communication signal and providing a demodulated signal at output, and an extraction module (48) of the FSK message which receives the demodulated signal and which supplies the demodulated FSK message.   2. Device according to claim 1, in which the device further comprises an adaptive filter (22) intended to eliminate components of the received communications signal and to output a filtered signal (28), in which the estimator of the correlation (41) receives the filtered signal (28) as a demodulation signal.   3. Device according to any one of the preceding claims, in which said filters (42, 44) are low-pass filters.   4. Device according to claim 3, wherein said filters (42, 44) are high-pass filters.   5. Device according to any one of the preceding claims, in which the adaptive filter (22) comprises a Wiener filter.   6. Device according to any one of the preceding claims, in which the adaptive filter (22) receives the communications signal as the first input, and a delayed communications signal as the second reference input.   7. Device according to any one of the preceding claims, in which the adaptive filter (22) comprises at least one finite impulse response filter.   8. Device according to any one of the preceding claims, produced in the form of a digital circuit for processing samples of the received communications signal.   9. Device according to any one of the preceding claims, produced in the form of an analog circuit.