Rate adaptive pacemaker

The invention relates to a rate-adaptive pacemaker according to the preamble of claim 1.

Pacemakers are known which set the adaptive heart rate as a function of the load on the pacemaker wearer.

DE-C-34 19 439 describes a pacemaker which measures the blood temperature of the venous blood in the heart with a temperature sensor and adjusts the adaptive heart rate as a function of the measured value. This principle is based on the knowledge that the blood temperature of the human being increases during exercise. The blood temperature is assigned to the physiologically sensible adaptive heart rate by means of a characteristic curve which assigns a value of the adaptive heart rate to each value of the blood temperature.

A disadvantage of this known pacemaker is that the relationship between blood temperature and physiologically sensible heart rate is generally different for each person, so that the pacemaker must be individually calibrated for each pacemaker wearer.

In addition, a change in the blood temperature independent of the load - for example, due to aging of the temperature sensor or a shift in the temperature sensor in the body of the pacemaker wearer - also leads to a change in the adaptive heart rate, which is not physiologically sensible.

A large number of arrangements for impedance measurement in the region of the thorax or in the heart for obtaining an impedance signal for ratanadaptive pacemakers are known, so that the technique of intracardiac impedance measurement as such is familiar to the person skilled in the art. Most of these arrangements, however, aim to make a statement about the respiratory or minute volume as an expression of the physical strain on the patient and as an actual rate control parameter.

The so-called ResQ process is also known (<u>R</u>regional <u>E</u>ffective <u>S</u>lope <u>Q</u>uality) (SCHALDACH, Max: Electrotherapy of the Heart, 1st edition, Springer-Verlag, p.114ff.), in which the temporal course of the intracardiac impedance is used to determine the physiologically sensible adaptive heart rate.

This method is based on the knowledge that the intracardiac impedance in a certain time window after a QRS complex - the so-called "region of interest" (ROI) - has a particularly significant dependence on the load on the organism.

The slope of the impedance curve in the ROI is therefore determined and the difference between the slope of a rest or reference curve and the slope of the currently measured impedance curve (load curve) is calculated. The adaptive heart rate is set depending on this difference. The assignment of the calculated gradient difference to the heart rate to be set also takes place here by means of a characteristic curve. Since this relationship is usually different for different people, the pacemaker must be calibrated individually for each wearer, and the calibration must be repeated if the patient's health, exercise capacity or life circumstances have changed significantly, and the position of the ROI must also be checked.

It is therefore, in particular, the object of the invention to create a generic pacemaker which can do without a patient-specific calibration process and which adapts itself automatically to changed boundary conditions.

This object is achieved by the features specified in claim 1.

The invention includes the idea that - due to the connections via the autonomic nervous system (ANS) reflecting the overall stress situation (physical and psychological stress) of a patient excellently - the time course of the intracardiac (especially right ventricular) impedance in a meaningful, no patient-specific setting to use the required size for rate adaptation.

The pacemaker according to the invention evaluates the intracardiac impedance, in particular the unipolar measured right ventricular impedance, in a wide range, which includes the ROI ranges usually set for individual patients. In it, it determines a relation between the rest or reference and load curve, specifically using arithmetic processing of the time integral of the impedance over the range mentioned.

For this purpose, the output of the corresponding integrator stage is in particular connected to an integral value memory, in which a reference integral value determined in at least one previous cardiac cycle is stored, and the rate determining device comprises an arithmetic unit connected to the output of the integrator stage and the integral value memory for calculating a secondary impedance variable from the respective primary impedance variable and the reference integral value in accordance with a predetermined arithmetic relationship. In an advantageously simple embodiment, the arithmetic unit is a subtraction stage for forming the difference value between the primary impedance variable and the reference integral value - however, other arithmetic processing or, if necessary, a multi-stage threshold value discrimination can also take place.

The definition of the time range or integration limits does not require any patient-specific programming after the implantation, but these can be stored in a fixed value memory (at least indirectly) connected to a control input of the integrator stage, in particular during the manufacture of the pacemaker. The limits of the predetermined section are determined as a result of examinations of the range of the temporal course of the impedance relevant to the rate adaptation in a patient population.

The rest or reference curve mentioned is preferably “carried along”, that is to say averaged from impedance values ​​obtained over a predetermined time period of a few (for example three) minutes, wherein either a moving averaging or an averaging in each case for successive separate periods can be carried out. This enables rapid adaptation to changing boundary conditions - such as stimulation parameters, medication or lifestyle habits - and prevents the patient from being endangered by persistence of the pacemaker at unphysiologically high rates.

In the case of a transition from spontaneous to stimulated cardiac activity (when there is an increase in stress) or vice versa, the current impedance curve is expediently defined as a new reference curve. To avoid sudden rate changes, however, the rate should not be changed at this time or not significantly, which requires the introduction of a rate offset amount, which is then gradually reduced again over a predetermined time or number of cardiac cycles. The implementation of this function - in addition to a corresponding design of the pacemaker or sequence control - serve a control connection to the output stage and an offset memory.

An expanded functionality is provided by equipping the pacemaker with a sensor - at least indirectly connected to a control input of the integrator stage and / or a control input of the rate determination device - for an activity variable whose output signal sets at least one of the limits of the integration range and / or a characteristic curve of the rate determination device . A particularly simple and at the same time expedient sensor is a digital motion sensor which, in particular, causes a switchover between different programmed time range or integration limits and / or a corresponding switchover of the processing characteristics (characteristic curve) of the rate determination device.

In a further special embodiment, the rate determination device comprises a differential element having a data input for the impedance variable, with which very slow changes in impedance for the pacemaker control are rendered ineffective. As a result, with a drastically reduced calculation effort and power consumption, an effect similar to that of carrying the impedance-rest curve is achieved.

In an expedient development, the differential element has a time constant control input connected to the sensor for the activity variable, by means of which the differential time is set to a value which is substantially smaller in the idle state than when the patient is active. If a digital motion sensor - already mentioned above - is used, the setting is made in particular as a switchover between preset time constants.

The characteristic curve determining the dependence of the stimulation rate on the impedance variable as an essential operating parameter of the rate determination device is preferably not static, but is optimized continuously or at certain time intervals. The goal of the optimization is first of all to adapt the variation range of the impedance variable to the permissible variation range of the heart rate.

The range of variation is not known at the start of operation, but is determined by continuously measuring the impedance during operation and optimized after each measurement. At the start of operation, an estimate for the lower and the upper limit of the variation range is specified as the starting value.

A distinction must now be made between two cases in the optimization: On the one hand, it can happen that a measured value of the impedance exceeds the previously determined range of variation upwards or downwards. In this case, the range of variation is expanded accordingly and thereby updated. The growth time constant of this adaptation process is preferably of the order of a few seconds in order to achieve a rapid adaptation and thus to prevent an excessive heart rate. On the other hand, the case may arise that the impedance no longer fully utilizes the previously determined range of variation over a longer period of time. In one embodiment of the invention, the range of variation is therefore slowly reduced again in this case. The time constant of this adaptation process is preferably in the order of several weeks.

The characteristic curve is adjusted during the optimization to the current value of the range of the activity variable (impedance). The characteristic curve thus assigns the base rate to the lower limit of the range of variation of the activity variable and the maximum stimulation rate corresponding to the upper limit of the activity variable. By changing these limit values ​​during optimization, the characteristic curve also changes.

In an advantageous variant, the calculated adaptive heart rate is therefore statistically evaluated. From the statistical distribution within the permissible range of variation from the base rate to the maximum stimulation rate, information can be obtained for the optimization of the characteristic curve. A physiologically meaningful statistical distribution of the activity variable in the form of a frequency distribution function can be determined for each patient depending on his load profile and other factors by setting the course of the characteristic curve between the support points so that the frequency distribution function of the adaptive heart rate of the physiologically meaningful frequency distribution function is as possible comes close.

The characteristic curve can be implemented, for example, as a polynomial, the characteristic curve being completely determined by the coefficients of the polynomial:<maths id="math0001" num=""><math display="block"><mrow><msub><mrow><mtext mathvariant="italic">HR (A) = K</mtext></mrow><mrow><mtext mathvariant="italic">0</mtext></mrow></msub><msub><mrow><mtext mathvariant="italic"> + K</mtext></mrow><mrow><mtext mathvariant="italic">1</mtext></mrow></msub><msub><mrow><mtext mathvariant="italic">A + K</mtext></mrow><mrow><mtext mathvariant="italic">2</mtext></mrow></msub><msup><mrow><mtext mathvariant="italic">· A</mtext></mrow><mrow><mtext mathvariant="italic">2</mtext></mrow></msup><msub><mrow><mtext mathvariant="italic"> + K</mtext></mrow><mrow><mtext mathvariant="italic">3</mtext></mrow></msub><msup><mrow><mtext mathvariant="italic">· A</mtext></mrow><mrow><mtext mathvariant="italic">3</mtext></mrow></msup><mtext mathvariant="italic"> +</mtext><mtext> ...</mtext></mrow></math><img file="EP0793976A2_D0001.tif" /></maths>

The coefficients <i>K</i>_<i>i</i> are then determined in such a way that, on the one hand, the limit values ​​of the range of variation of the activity variable are assigned the corresponding limit values ​​of the permissible range of variation of the heart rate, and on the other hand the frequency distribution function of the adaptive heart rate comes as close as possible to the physiologically meaningful frequency distribution function.

The following must therefore apply:<maths id="math0002" num=""><math display="block"><mrow><mtable><mtr><mtd><mrow><mtable><mtr><mtd><mrow><msub><mrow><mtext mathvariant="italic">HR (A</mtext></mrow><mrow><mtext mathvariant="italic">MIN</mtext></mrow></msub><mtext mathvariant="italic">) = TR</mtext></mrow></mtd></mtr><mtr><mtd><mrow><msub><mrow><mtext mathvariant="italic">HR (A</mtext></mrow><mrow><mtext mathvariant="italic">MAX</mtext></mrow></msub><mtext mathvariant="italic">) = MSR</mtext></mrow></mtd></mtr><mtr><mtd><mrow><msub><mrow><mtext mathvariant="italic">V (HR) = V</mtext></mrow><mrow><mtext mathvariant="italic">REF</mtext></mrow></msub></mrow></mtd></mtr></mtable></mrow></mtd></mtr></mtable></mrow></math><img file="EP0793976A2_D0002.tif" /></maths> With<dl id="dl0001"><dt><i>A</i>_<i>MIN</i><i>, A</i>_<i>MAX</i></dt><dd>Limits of the range of variation of the activity quantity</dd><dt><i>TR</i></dt><dd><u>T</u>arget <u>R</u>ate (base rate)</dd><dt><i>MSR</i></dt><dd><u>m</u>aximal <u>S</u>timulations<u>r</u>ate</dd><dt><i>V</i></dt><dd>Frequency distribution function of the heart rate</dd><dt><i>V</i>_<i>REF</i></dt><dd>physiologically useful frequency distribution function</dd></dl>

To calculate the frequency distribution function of the adaptive heart rate, for example, the entire permissible variation range is divided into equidistant frequency intervals, and the time in which the heart rate was within the frequency interval during an observation period is determined for each frequency interval. In order to suppress the effect of cyclical fluctuations in the adaptive heart rate in the daily or weekly cycle, the observation period is preferably of the order of several weeks.

The combination of activity sensor and differential element results in the following mode of operation: If the motion sensor detects a movement of the pacemaker wearer, the differential time <i>T</i>_<i>D</i> and the differential transmission factor <i>K</i>_<i>D</i> of the differential link is also greatly increased. By increasing the differential time<i>T</i>_<i>D</i> it is achieved that long-term stresses are supported by an increased heart rate. By simultaneously increasing the differential constant<i>K</i>_<i>D</i> the continuity of the output signal of the differential element is ensured. Otherwise, the output signal of the differential element would change if the differential time<i>T</i>_<i>D</i> jump, which makes no physiological sense. After the end of the physical stress, the differential time<i>T</i>_<i>D</i> and the differential transmission factor <i>K</i>_<i>D</i> reset to the values ​​before the stress phase.

Advantageous developments of the invention are further characterized in the subclaims or are shown in more detail below together with the description of a preferred embodiment of the invention with reference to the figures.

Show it:<dl id="dl0002"><dt>Figure 1</dt><dd>a pacemaker as an exemplary embodiment of the invention as a block diagram,</dd><dt>Figure 2</dt><dd>a characteristic curve and an optimized characteristic curve for assigning the activity variable to the adaptive heart rate in the pacemaker shown in FIG. 1,</dd><dt>Figure 3</dt><dd>the frequency distribution functions of the adaptive heart rate which result from the characteristic curves shown in FIG</dd><dt>Figure 4</dt><dd>an embodiment of an impedance measuring and processing device, which can be used in a pacemaker similar to that shown in Fig. 1.</dd></dl>

FIG. 1 shows, as an exemplary embodiment of the invention, a rate-adaptive pacemaker 1 as a functional block diagram, only the components that are important for the explanation of the invention being shown.

The impedance measuring device and processing device 3 measures the right ventricular intracardial impedance via a unipolar measuring electrode 2 in the right ventricle of the heart H. <i>Z.</i>. The measurement is carried out in a clocked manner by applying a measuring voltage to the measuring electrode 2 and measuring the current between the measuring electrode 2 and the housing 1a of the pacemaker acting as a counterelectrode at eight equidistant times within a fixed preprogrammed time range. The impedance<i>Z.</i> results as the quotient of the measuring voltage and current. The clock frequency of the impedance measurement here is a few tens to approximately 100 Hz. According to the generally known sampling theorem, this ensures that the discrete-time impedance signal represents the actual course of the impedance<i>Z.</i> reproduces sufficiently well.

The impedance <i>Z.</i> shows relatively large changes during a cardiac cycle. So is the impedance<i>Z.</i> usually minimal at the beginning of a cardiac cycle immediately after a QRS complex and then increases again until the next QRS complex; their course over time, as will be explained in more detail below with reference to FIG. 4, results in the rate setting parameter after integration and arithmetic processing.

The measuring device and processing device 3 is followed by a differential element 4 with a differential time <i>T</i>_<i>D</i>. This differential element 4 has the task of making slow changes in the intracardiac impedance<i>Z.</i>, whose time constant over the differential time <i>T</i>_<i>D</i> to filter out.

In addition to the impedance measuring device, a motion sensor 5 is provided, which delivers a binary signal (motion yes / no) depending on the motion state of the pacemaker wearer. The differential time is dependent on this movement signal<i>T</i>_<i>D</i> of the differential member 4 set. When the pacemaker wearer is at rest, there is a differential time<i>T</i>_<i>D1</i> = 10 min set. That means changes in impedance<i>Z.</i>whose time constant is greater than 10 min do not lead to a relevant change in the adaptive heart rate. If the movement sensor detects a movement of the pacemaker wearer which is associated with a physical load, the differential time is opened<i>T</i>_<i>D2</i> = 5 h set.

The output signal of the differential element 4 is fed to a rate determination device 6, which has the adaptive heart rate at its output <i>MR</i> issues.

The rate determination device 6 has a characteristic curve element 10, which corresponds to each value of the impedance variable, which has been reworked in the differential element <i>A</i> a value of the adaptive heart rate by means of a characteristic curve <i>MR</i> assigns. The characteristic curve is continuously set and optimized during operation of the pacemaker by means of an adjusting device 8, by processing the impedance variable to the permissible variation in the heart rate<i>MR</i> is adjusted by a base rate <i>TR</i> as a lower limit and a maximum stimulation rate <i>MSR</i> is specified as the upper limit.

The range of variation of the activity size <i>A</i> is determined by a (not shown) discriminator unit, which is continuously the maximum <i>A</i>_<i>MAX</i> and the minimum <i>A</i>_<i>MIN</i> the impedance size during the past four weeks.

The characteristic <i>K</i> is implemented in the characteristic element 10 as a polygon with a total of twelve equidistant support points. A support point is due to the lower limit<i>A</i>_<i>MIN</i> the range of variation of the processed impedance variable and the base rate <i>TR</i> and a second support point through the upper limit <i>A</i>_<i>MAX</i> the range of variation of the impedance size and the maximum stimulation rate <i>MSR</i> given, the relationships mentioned above apply. The location of the remaining (ten in the example) support points can be derived from a further optimization goal, which consists in the frequency distribution function<i>V</i> the adaptive heart rate <i>MR</i> as good as possible on a reference curve <i>V</i>_<i>REF</i> adapt.

The adaptive heart rate is therefore determined by an evaluation unit 9 <i>MR</i> statistically evaluated. The frequency distribution function is used continuously for this<i>V</i> the adaptive heart rate <i>MR</i> certainly. This happens in that for each of the twelve support points of the characteristic<i>K</i> lying eleven frequency intervals <i>ΔHR</i>_<i>i</i> in each case the percentage of time within an observation period for which the adaptive heart rate is determined <i>MR</i> within this frequency interval <i>ΔHR</i>_<i>i</i> lay. The determined frequency distribution function<i>V</i> with a reference curve <i>V</i>_<i>REF</i> compared which is a physiologically meaningful frequency distribution function of the adaptive heart rate <i>MR</i> represents. This is done in the support points<i>V</i>_<i>i</i> the difference <i>ΔV</i>_<i>i</i> between the measured frequency distribution function <i>V</i> and the reference curve <i>V</i>_<i>REF</i> formed and fed to the actuator 8.

Agree the measured frequency distribution function <i>V</i> and the reference curve <i>V</i>_<i>REF</i> match, the characteristic curve is optimally adapted. Otherwise the characteristic curve<i>K</i> optimized by the actuator 8. The frequency distribution function lies in the frequency interval between the first and the second support point of the characteristic<i>V</i> the heart rate over the reference curve <i>V</i>_<i>REF</i>, this means that heart rates within this frequency interval occur too frequently. The slope^<i>dHR</i>/_<i>there</i> the characteristic curve must therefore be raised in this frequency interval. For this purpose, the second support point is shifted against the first support point in the direction of falling activity values. On the other hand, the frequency distribution function lies<i>V</i> below the reference curve <i>V</i>_<i>REF</i>, this means that heart rates occur too seldom within this frequency interval. The slope^<i>dHR</i>/_<i>there</i> the characteristic curve must therefore be reduced in this frequency interval. For this purpose, the second support point is shifted relative to the first support point in the direction of increasing activity values.

The new location of the other support points is determined in the same way. The frequency distribution function lies in a frequency interval between two support points<i>V</i> above the reference curve <i>V</i>_<i>REF</i>, the support point belonging to the higher frequency is shifted in the direction of decreasing activity or impedance variable values.

A driver or output stage 13 is connected downstream of the control device to stimulate the heart. The pacemaker 1 works according to the demand principle, ie it only stimulates the heart if there is no contraction of the heart by natural stimulation within a certain waiting time after a previous contraction of the heart. The driver stage 13 therefore has a comparator unit 11, which takes the electrocardiogram (ECG) on the heart 1 via the measuring electrode 2, determines the natural heart rate therefrom and compares this with the adaptive heart rate. If no natural contraction of the heart 1 is detected within the waiting time after a previous contraction, the comparator unit 11 controls a pulse generator 12 which emits an electrical stimulation pulse on the stimulation electrode 2 (which also serves as the measuring electrode).

To control the general pacemaker functions and to carry out the impedance measurement and processing, a time control (controller) 14 is provided, which outputs control signals (symbolically represented by a single arrow) to the functional components.

Figure 2 shows the characteristic as an example <i>K</i> of the characteristic curve element 10 from FIG. 1 and an optimized characteristic curve <i>K</i>_<i>OPT</i>. The characteristic<i>K</i> is formed by a polygon with a total of twelve support points and assigns each value to the range of the activity size from <i>A</i>_<i>MIN</i> to <i>A</i>_<i>MAX</i> a value of the adaptive heart rate <i>MR</i> to.

The limits of the range of variation of the activity quantity <i>A</i> are not constant here. Rather, the pacemaker continuously "learns" during operation which range of variation results from the loads occurring in the daily life of the pacemaker wearer. The range of variation of the activity size<i>A</i> is constantly being redefined. This also changes the position of the characteristic curve<i>K</i>.

The lower limit <i>A</i>_<i>MIN</i> the range of variation of the activity size <i>A</i> and the base rate <i>TR</i> form a support point and the upper limit <i>A</i>_<i>MAX</i> the range of variation of the activity size <i>A</i> and the maximum stimulation rate <i>MSR</i> another support point of the characteristic <i>K</i>. The characteristic curve runs between these two support points<i>K</i> After the pacemaker has been started up for the first time, it is initially linear. However, the course is changed as part of an optimization process so that the frequency distribution function of the adaptive heart rate<i>MR</i> comes as close as possible to a physiologically meaningful reference curve. The optimized characteristic<i>K</i>_<i>OPT</i> shows in the lower range the range of the heart rate <i>MR</i> a smaller slope ^d<i>MR</i>/_d<i>A</i> on than the linear characteristic <i>K</i>. This means that the optimized characteristic curve<i>K</i>_<i>OPT</i> lower heart rates <i>MR</i> more often. The slope is corresponding^d<i>MR</i>/_d<i>A</i> in the upper range of the heart rate variation range <i>MR</i> with the optimized characteristic <i>K</i>_<i>OPT</i> larger than with the linear characteristic <i>K</i>. High adaptive heart rates<i>MR</i> close to the maximum stimulation rate <i>MSR</i> therefore occur with the optimized characteristic <i>K</i>_<i>OPT</i> less often.

FIG. 3 shows frequency distribution functions of the adaptive heart rate <i>MR</i>, which result from the characteristic curves shown in FIG <i>K</i> and <i>K</i>_<i>OPT</i> can result. The frequency distribution function<i>V</i> it arises from the characteristic curve <i>K</i> resulting distribution of adaptive heart rate <i>MR</i> across the entire range of the heart rate <i>MR</i> from the base rate <i>TR</i> up to the maximum stimulation rate <i>MSR</i> The maximum of the frequency distribution function <i>V</i> is in the upper frequency range close to the maximum stimulation rate <i>MSR</i>, ie the heart beats relatively often in the upper frequency range, which makes no physiological sense. The physiologically meaningful frequency distribution function of the adaptive heart rate is through the reference curve<i>V</i>_<i>REF</i> given. The maximum in the lower frequency range is close to the base rate<i>TR</i>. The characteristic<i>K</i> of the characteristic curve element is set according to the principle outlined in FIG. 2 in such a way that the frequency distribution function <i>V</i> the shape of the reference curve <i>V</i>_<i>REF</i> assumes.

An expedient embodiment of the essential components for impedance measurement and processing of a pacemaker 1 'which is slightly modified compared to FIG. 1 is shown in FIG. 4 in the form of a functional block diagram. Based on the reference number 3 selected in FIG. 2, the actual impedance measuring device is designated 3.1 and the impedance processing device 3.2.

The impedance measuring device 3.1 connected on the input side to the intracardiac measuring, sensing and stimulating electrode 2 comprises, in a construction known per se, a scanning pulse generator 300, a filter stage 301 for the separation of interfering (eg from breathing activity) signal components, a current meter 302 and an impedance calculation stage 303 Impedance measurements are clocked by the sequence controller 14 (cf. FIG. 1) in synchronism with stimulated or spontaneous cardiac events.

The measurement signal Z 'passes from the output of the impedance calculation stage 303 to the input of an integrator stage 304, in which a fixedly programmed number (for example eight) of impedance measurements from a cardiac cycle is subjected to an integration. The number is permanently stored in an integration limit memory 305 and determines the count of a counter 306 which counts the measurement control pulses from the sequence control and which stops the integrator when the programmed number is reached. The counter 306 simultaneously triggers the transfer of the integration result A ″ to a subtraction stage 307 on the one hand and via a delay element 308 to a FIFO memory 309 on the other hand.

A predetermined number of impedance integral values ​​from the past (for example the last three operating minutes of the pacemaker in each case) are constantly stored in the memory 309, each of which is taken over by the output signal of the counter 306 into an averaging stage 310 and subjected to a current averaging.

The output signal of the averaging stage 310 is (in addition to the current impedance integral value as the primary impedance variable A '') as the reference value A ''_ref the subtraction stage 307, which forms the difference between the current integral value and the current time average of the integral values ​​as a reference value and outputs it as a secondary impedance variable A ', which here represents the rate control parameter.

It should be noted that the signal connection of the sequence controller 14 to the measuring electrode 2 shown in FIG. 1, with which the cardiac actions or intracardiac EKGs are also recorded, distinguishes between spontaneous and evoked cardiac actions and thus deleting the FIFO 309 when the event type changes enables a flip-flop 311 connected to the sequence control 14 on the input side and to an erase input of the FIFO on the output side. Its output signal is also fed to a modified rate determination device 6 ', where a rate offset is added to the calculated stimulation rate with each change of event type, the amount of which is selected as a function of the previous and the current rate value so that the rate jump does not exceed a predetermined amount, and that of the subsequent cardiac events are gradually reduced to zero. The specific circuitry means for realizing this additional function are available to the person skilled in the art from known arrangements for rate smoothing or adjustment.

The further processing of the secondary impedance variable A 'corresponds - apart from the elimination of the differential element replaced by the components for forming the moving average - to the explanation given for FIG. 1.

The embodiment of the invention is not limited to the preferred exemplary embodiments specified above. Rather, a number of variants are conceivable which make use of the solution shown even in the case of fundamentally different types.

In particular, stage 3.2 can have a large number of alternative designs in which, for example, the formation of a moving average value as a reference value is replaced by time averaging with fixed starting points at predetermined time intervals or instead of on the basis of the output signals of integrator stage 304 on the basis of the output signals of impedance calculation stage 303, ie from ( Impedance, time) value pairs, can be performed. Instead of a fixed predetermined number of impedance values ​​to be integrated, a fixed temporal integration range can also be programmed. It is furthermore advantageously possible to provide means for setting the respectively valid integration range as a function of the signal from the motion sensor (or another activity sensor).