EP2002554A1

EP2002554A1 - Method and system for information transmission

Info

Publication number: EP2002554A1
Application number: EP07710825A
Authority: EP
Inventors: Andreas Stucki; Andreas Martin HÄBERLI
Original assignee: Kaba AG
Current assignee: Dormakaba Schweiz AG
Priority date: 2006-04-03
Filing date: 2007-03-30
Publication date: 2008-12-17
Also published as: CN101449479A; BRPI0710043A2; TWI443980B; RU2446561C2; NZ571613A; MY148051A; CA2648128A1; AU2007234292B2; JP2009532958A; US20090161734A1; CN101449479B; WO2007112609A1; IL194444A; ZA200809386B; IL194444A0; JP2013211864A; MX2008012725A; AU2007234292A1; RU2008143184A; KR20090012232A

Abstract

The invention relates to the transmission of information between a data transmitting device and a read and/or write module, in particular, for application in access control. According to the invention, data for transmission is given by the data transmitting device as a digital signal which is converted into an ultra broadband signal by means of the frequency spread method and capacitively and/or resistively transmitted to the write and/or read module via the body of the user.

Description

METHOD AND SYSTEM FOR INFORMATION TRANSMISSION

The invention relates to a method for transmitting information from, for example, a portable device to a writing and / or reading module, a system for carrying out this method, a data transmitting device and a writing and / or reading module.

For the transmission of information signals, in particular digital signals, between transmitters and receivers, a plurality of channels is available. One such channel is the capacitive (more precisely: capacitive / resistive) coupling between a portable device and a read / write module. Particularly interesting in the application is such a coupling, if it takes place over the human body as a transmission medium. Corresponding systems are disclosed, for example, in US Pat. No. 4,591,854, US Pat. No. 5,914,701 and US Pat. No. 5,796,827. A user is carrying a portable device. As soon as the user touches or is in the immediate vicinity of a touchpad coupled to a read / write module, information flows. For example, a unique access code may be transmitted from the portable device to the read and write module.

For practical applications, especially the access control in the broadest sense, the following requirements arise, which in combination still no existing System. and which has hitherto prevented a commercial breakthrough of this type of information transfer:

A. Signal-to-noise ratio. A favorable signal-to-noise ratio is only possible with a large amplitude of the transmission signal. In - - high-impedance - human body existing, impressed by electrical equipment

Potential fluctuations are of the order of magnitude of a few 100 mV in the

Range up to 1 MHz. For a transmission system is a much larger

Signal amplitude (i.e., high voltage on the body) but not tolerated by the user. The method should therefore also work in case of unfavorable signal-to-noise ratio.

B. Inexpensive Components of the Portable Device: While, for example, for RFID information transmission, the simplest passive components suffice, a portable device for capacitive transmission must have an active transmitter with voltage source, and the problem arises of synchronization with the receiver. Accurate clock (quartz oscillator or similar) are expensive; with less accurate clocks increases the synchronization effort.

C. Speed: The entire information transmission process, including synchronization, can last no more than a few seconds, better still less than a second, depending on the application, at most 300 ms or at most 200 ms.

It has already been proposed in US Pat. No. 5,914,701 to use the "direct-sequence spread spectrum" modulation method for transmitting information, thereby reducing the noise sensitivity (which is probably meant in particular for interference sensitivity) and making it possible for several Transmitters are active at the same time, each transmitter has its own modulation code (spreading code). In fact, the "spread-spectrum" method, which has long been known per se, is known to reduce the susceptibility of signals to signals and to signal-specificly code signals, but there are also disadvantages: the computational effort of the read and / or write module and the synchronization effort The above-mentioned document US 5,914,701 contains no indication as to how the synchronization can be accomplished without violating requirements B and C. Moreover, depending on the application, it may also be disadvantageous if several portable devices are simultaneously possible For example, for the secure access control application, it should rather be ensured that the data received from the read and / or read module originates only from the user who is in the immediate vicinity of a control surface of the module and for example this touches.

On the basis of this prior art, the object is to provide a method for the transmission of information by which disadvantages according to the prior art are overcome, and which satisfies the requirements A to C at least partially. Preferably, the method should have the advantages of "intrabody" capacitive information transmission across the human body and be able to ensure that data received from a read and / or write module is from the portable device that the user in close proximity to the module is referring to wearing.

This object is achieved in that data to be transmitted by a (often carried by the user, portable) device represented as a digital signal and this converted by means of the frequency spreading in an ultra-wideband signal and Capacitive and / or resistive -via the user's body or directly - is transmitted to a writing and / or reading module.

Under the capacitive and / or resistive signal transmission via the human body, the signal transmission between a transmitter (transmitter) and a receiver (receiver) is understood, via a transmitter interface, a signal from the transmitter into the body coupled and from the body into a receiver Interface can be coupled. The coupling through the body is primarily resistive. Depending on the situation, the coupling between the transmitter and the receiver interface is primarily capacitive or primarily resistive or a combination of both. A primary resistive coupling between interface and body takes place when the interface has an electrode directly touched by the body; otherwise i.A. capacitive shares. This type of signal transmission via capacitive and / or resistive coupling is also called "intrabody" signal transmission In the literature (especially in US 5,914,701), intrabody signal transmission is modeled primarily by capacitive coupling.

Ultra-broadband is defined as the use of a frequency range of at least 20% of the center frequency or at least 500 MHz for the transmission of information. For the inventive method transmission frequencies of over 100 MHz are disadvantageous or not feasible, so that in the following "ultra-broadband" with "at least 20% of an average transmission frequency", i. if appropriate "at least 20% of a carrier frequency" is equated.

Frequency-spread ultra-wideband signals are used in the prior art where interference with other transmission channels is to be prevented (eg in personal area networks). Such signals are (eg. UMTS) also used to communicate with a large number of users at the same time collision-free. The invention makes use of the new realization that the transmission of a frequency-spread ultra-wideband signal can also be advantageous for point-to-point transmissions without interfering, other information channels - such point-to-point transmissions are involved the capacitive and / or resistive transmission by the human body.

It has been shown that the capacitive and / or resistive transmission of an ultra-wideband signal is advantageous, in particular with regard to the signal-to-noise ratio at low voltage amplitudes. In particular, the procedure according to the invention makes it possible to operate with voltage amplitudes of a few mV in the body - corresponding to, for example, up to 3 V or less on electrodes - which is below the potential fluctuations introduced by electrical devices in the human body anyway. The procedure according to the invention allows the signal to "disappear" as a pseudo-noise signal in the noise or interference (for example by a factor of 10) and therefore does not cause a measurable influence on the current flows in the human body.

In the embodiment of the method according to the invention, an amplitude at the coupling electrodes of 5 V, particularly preferably 3 V, is also preferably not exceeded.

The code frequency ("chipping frequency") is then, for example, at least one fifth of the signal center frequency, preferably at least half of that, due to the definition of "ultra-broadband" Signal Center Frequency .Im special In the preferred case, the chipping frequency is equal to the modulation frequency and thus equal to the center frequency.

The data word may be modulated prior to spreading by a digital data modulation technique, such as Phase Shift Keying (PSK), in particular the binary one

Phase shift keying method (BPSK) or also a quadrature or other phase shift keying method. Preferably, such a data modulation method is combined with a coding which makes the signal insensitive with respect to the absolute phase, eg by looking only at phase differences (differential coding) In the example PSK this results in a combination DPSK (differential phase shift Keying; differential phase shift keying), for example a DBPSK modulation This method, in combination with the procedure according to the invention, has the advantage that the absolute phase need not be known, but in differential phase shift keying only the relative phase between one symbol and the next one Symbol of meaning Alternatively to the differential coding also another eg error-correcting code with similar characteristics can be used, bpsw a rotation-invariant code.

Preferably, the data word contains one or more bits, the receiver a consistency test (checksum test in the broad sense), for example. A cyclic

Allow redundancy test. This is particularly preferred in the context of differential coding or rotationally invariant codes: By applying the

Consistency tests on the data obtained can be determined if at the

Decoding due to systematic phase rotations between two adjacent code symbols instead of the data word an artifact was obtained. Such systemic phase rotations may be due to a non-fixed

Relationship between the frequency of the transmitter-modulated Carrier oscillation on the one hand and the sampling frequency (possibly with phase correction of the code used in the receiver). If the consistency test (eg CRC test) results in a lack of consistency, the resulting symbol sequence is discarded and a new evaluation with a systematic phase rotation - for example by π / 2 - from symbol to symbol is made.

As an alternative to the differential phase shift keying, another modulation can also be used, for example a (non-differential) phase shift keying (PSK modulation), another modulation, or no modulation at all. In such a case, the absolute phase of the received signal may need to be known. For the determination of an absolute phase and frequency of the received signal, for example, a phase-locked loop (PLL) can be used, as it is known per se from the prior art. Although this involves a relatively large effort and a certain settling time, depending on the application, this embodiment is also practicable or even preferred.

According to the invention, the de-spreading (the receiver-side reverse operation for frequency spreading) is particularly preferably combined with the demodulation. According to the state of the art, modules for spreading and de-spreading on the one hand and modulation and demodulation on the other hand are independent of one another. This is illustrated in FIG. 11, where a method is shown, as used, for example, for contactless data transmission. Data to be transmitted ("data") are first modulated, for example by the PSK method, whereby this modulation can also be understood as a coding, followed by spreading and modulating onto a carrier frequency The procedure has the advantage that standard components can be used, ie, for example, a per se known spreader / de-spreader can also be used for a newly developed system However, the preferred embodiment of the invention has the advantage that no separate synchronization processes are necessary for the de-spreading and the demodulation, which is advantageous for the operation in the "burst" mode. Rather, the signal acquisition performed for the de-spreading is also used for the The de-spreading provides directly code symbols, which may still need to be decoded / demodulated (reversing operation of the modulation, for example the DPSK modulation), but for which synchronization is no longer necessary A chipping sequence (a code cycle) is particularly preferably the length of a code symbol.

Preferably, the signal is modulated on the frequency spread and before transmitting to a carrier signal. The carrier frequency may, as already mentioned, be equal to the chipping frequency. The modulating on a carrier signal has the advantage that a large part of the signal power is obtained in a frequency range which is sufficiently far away from the fault-prone very low frequencies (50 Hz, etc.). The exemplary embodiments described below all include modulating onto a carrier, even if this is not a necessary prerequisite of the invention, and the frequency-spread signal (it corresponds to a base band signal in terms of telecommunications) is also transmitted directly - possibly after suitable filtering with a low-pass filter can be.

With the method according to the invention it is possible to transmit in a "burst-mode", ie immediately and without time-consuming synchronization steps The acquisition (synchronization) and the tracking need a higher signal-to-noise ratio than the actual reception Therefore, according to a preferred embodiment the codes added up or averaged ("combining"). In order not to be dependent on a pilot sequence (eg preamble), the Averaged codes using a non-data aided (NDA) combiner. That requires that the code symbols are estimated. be, for example, with a DPSK demodulator.

Preferably, at least two, preferably immediately consecutive, signal sequences of the (possibly estimated) length of a data bit are correlated with the stored code and the results are added in a coded-symbol-value-adjusted manner, the added value can be used to calculate the acquisition and the, tracking 'signals with a particularly good signal-to-noise ratio' tracking 'is the tracking of the receiver with respect to the transmitter frequency, for example, by means of the "early-late-gating" method. The sign correction is preferably carried out in PSK modulation by means of a multiplication with a DPSK demodulation value which estimates the relative "sign" or the relatively complex argument of two data signals (corresponding to code symbols).

It can also be provided that the data word to be transmitted is continuously sent repeatedly so that a continuous bit stream is transmitted. The receiver can be configured so that the recording of the data word can start at any start time, ie as soon as the receiver has detected and acquired an incoming signal. The combination of these measures ensures that the transfer and recording of data can begin at the earliest possible date. There is then practically no time delay between the time when the receiver recognizes that a signal is received and the beginning of the data recording.

As an alternative or in addition, it can of course also be provided that the device transmitting the data contains means by which the signal is only caused is sent out at intervals. These may include manual activation of the device, activation by a motion detector, a suitable wake-up circuit, or any other means.

According to a specific embodiment, a plurality of simultaneously applicable correlators are available for the correlation method. This can take account of the fact that the frequency relationship between the transmitter-side "chipping" frequency and the receiver-side sampling frequency may not be exactly known - namely, if the transmitter's clock is not very accurate There is a significant correlation between the different chipping frequencies and only at the appropriate chipping frequency and possibly at adjacent frequencies.The different correlators can match a fake sample signal of a chipping code with different sampling frequencies (or - this is equivalent and comes to the same - correspond to the fake strobe signals of chipping codes with different chipping frequencies at a fixed sampling frequency.) This means that the length and phase of the various correlators are matched to the corresponding frequency offset t Typically, the correlators are in a receiver-A quantized btastintervall, Vz prefers the chipping length (corresponding to the double chipping frequency).

According to a first variant, this multiple of correlators - the correlator angle - covers the full width of the chipping frequency uncertainty. Then the acquisition can be completely parallel. As a further alternative, the correlator bank may cover only a portion of the possible frequency inaccuracy, and the sampling frequency, or equivalently, the correlator bank as a whole, may be varied stepwise across the full width of possible frequencies until a correlator yields (partially) significant data signals (code symbols) parallel acquisition). A correlator bank may be fixed and have a fixed frequency relationship to the fixed sampling frequency. Alternatively, a fine tuning may be provided according to which the sampling frequency is slightly adjusted based on correlation values (eg obtained from the tracking).

According to an alternative embodiment, which is particularly suitable for systems with relatively accurate transmitter-side clocks, only a single correlator is used. This can be used in the case of sufficiently high accuracy together with a fixed sampling frequency. Alternatively, the sampling frequency may be incrementally changed over a certain range until significant data signals are found.

For the acquisition, two criteria are available, of which at least one, preferably both are used: an amplitude or amount criterion and a time criterion. The amplitude or magnitude criterion is based on a comparison of a - supposed - magnitude maximum (peaks) with a noise level. If the average noise level has been exceeded by a certain threshold, for example between 2 and 5dB, a code symbol is suspected. The time criterion will satisfy two consecutive peaks if their time interval equals at least about one bit length. A bit length is at the same time a correlator (code) length or predetermined fraction thereof.

If multiple correlators are present, the signal is searched for peaks not only as a function of time but also as a function of the correlator or its number. There is a third criterion for the acquisition: different peaks should be assigned to the same correlator or maximally adjacent correlators since the transmission frequency exceeds the burst length must be approximately constant, ie it can not be that the chipping frequency between different bits changes greatly.

The method is particularly preferably carried out in such a way that it is precluded from the outset that more than one portable device (as transmitter) is involved in a data exchange. This can be ensured, for example, by the fact that the transmission frequency (or the center frequency) is less than 10 MHz, advantageously not greater than 2 MHz, particularly preferably not greater than 1 MHz. In addition, the transmission power should be so small that the capacitive-resistive coupling only works at very small distances. The fulfillment of these conditions is particularly interesting for the application "secure access control", because then the condition is met that the signal radiation (so to speak the information transmission past the human body as a transmission medium) is not measurable, then it is ensured that the receiver In fact, information received from the portable device actually originates from the user who touches the user interface or who is in the immediate vicinity of him / her, but at higher frequencies than the specified one, a signal emitting electrode would also act as an antenna.

The invention also relates to a system for the transmission of data which has at least one data transmitting (eg portable) device and at least one writing and / or reading module. Likewise provided by the invention are a corresponding data transmitting (for example portable) device and a writing and / or reading module. The device includes two electrodes, between which a time-dependent voltage can be applied so that the smallest current flows in the human body of a user when one of the two electrodes is located in the immediate vicinity of the body and the other a little further away. The write and / or read module has a detector, which detects an electrical voltage or electric currents between a first and a first second electrode detected. The first electrode is generally arranged in such a way that, in the operating state, it is located in the immediate vicinity of the user's human body. It may, for example, be designed as a control surface, a pusher surface, door knob, etc. For example, a conductive plate can serve as the second electrode. The currents in the body of the user cause in the operating state, a capacitive and / or resistive coupling between the electrodes of the data transmitting device and those of the read and / or write module.

The device is now controlled so that the time-dependent electrical voltage in the operating state emits an ultra-wideband frequency-spread signal. The read and write module has a data acquisition and decoding unit which decodes an incoming frequency spread ultra wideband signal.

The system according to the invention, the device according to the invention and the writing and / or reading module according to the invention can be designed such that they can carry out the method according to any of the preferred embodiments described above and below.

Hereinafter, preferred embodiments of the invention will be described in detail with reference to drawings. Showing:

1 is a diagram of an embodiment of the inventive method,

2 shows the process steps taking place in the portable device according to an embodiment of the method according to the invention, 3 shows the processing of the received signal in the writing and / or reading module up to scanning according to an embodiment of the method according to the invention, FIG.

4 shows the processing of the scanning signal in the writing and / or reading module according to an embodiment of the method according to the invention,

5 shows the processing of the scanning signal in the writing and / or reading module according to an alternative embodiment of the method according to the invention,

6 shows an embodiment of the signal detection for the method according to the invention,

FIG. 7 shows an embodiment of tracking,

8 shows an embodiment of the decoding of the code symbols,

9 shows a schematic representation of the "combining signal" as a function of time,

10 very schematically a system according to the invention with a portable device according to the invention and a writing and / or reading module according to the invention, Fig. 11 is a schematic representation of a system for information transmission by means of radio waves according to the prior art.

The system according to FIG. 1 has a data-transmitting, portable device 1 and a writing and / or reading module 2. These are able to communicate capacitively and / or resistively via the body of a user 3 or via direct capacitive / resistive coupling between transmitter and receiver. The latter is, for example, the case when a user holds a tag for reading directly to a receiver electrode connected to the receiver.

According to the invention, data from the portable device is transmitted capacitively and / or resistively via the body of a user to a writing and / or reading module using the spread spectrum method as an ultra-wideband signal. The data 11 may, for example, be present digitally in an EEPROM memory and be subjected to the frequency spreading method 12 for the data transmission, whereupon they are modulated onto a carrier signal 13. The frequency spreading method may, for example, be in the form of a direct sequence frequency spreading method and thus include modulation with a periodically repeated "chipping" sequence The "chipping sequence" is of the type of a pseudo-random "bit Sequence and is also called "code", "spreading code" or "chipping code". The time duration T _c = l / fc of a single chip is less than the symbol length (bit period) T _B = 1 / f _β (/ ^ bit frequency).

Other embodiments of the frequency spreading method are conceivable, for example, the "frequency hopping" or the "pulse position modulation" method. Embodiments of the method according to the invention are discussed below, which are based on the preferred direct sequence spread spectrum (Direct Sequence Spread Spectrum).

Short codes, ie codes of a length of, for example, at most 10 T _B , particularly preferably codes of length Tβ, are particularly preferred for direct-sequential frequency spreading. In most embodiments, no measures are provided for CDMA (Code Division Multiple Access).

In the read and / or write module, the data is again multiplied by a signal 21 of the carrier frequency (demodulation), whereupon the received signal is synchronized by a correlator 22 with a code signal generated in the writing and / or reading module. This is followed by decoding 23 (i.e., the generation of a bit sequence from the received signal), whereupon the decoded data is processed in a data processing unit 24.

The processing of the data can consist, for example, of verifying an identification code: in the case of consistency, in the application example of access control, the release of an object takes place, for example by means of a control signal to a mechatronic unit. Alternatively, the processing of the data may involve one or more further steps and / or other steps than simply verifying a code. In addition, a further data exchange via the capacitive and / or resistive information transmission and / or other channels can be initiated. Overall, the capacitive / resistive information transfer is typically unidirectional; any other channels may also be unidirectional or allow for information transfer in the other direction or bidirectional information transfer. For example, the transmission of a message in the way described here can serve to build a communication channel (capacitive-resistive and / or by other means). Examples of information transmission methods are also described in the international patent application PCT / CH 2006/000518, to which express reference is made here.

With reference to FIGS. 2 to 8, implementations of the method according to the invention in a read and / or write module will be described below with reference to exemplary embodiments. The following problems are addressed:

1. A direct-sequence-frequency-spread signal arrives at the receiver as a series of "chips." If the ratio between chipping frequency and bit frequency is N, then N chips form a code symbol, all N intercepts

Chips a new code icon. In order to be able to read the data at all, the receiver must determine where each new code symbol begins in an incoming chip sequence, in order by chip-wise multiplication (or, depending on the representation / resolution of the data by an XOR operation) the chipping code also stored at the receiver to obtain a bit sequence.

The process of detecting the position of code symbols of a bit length in a chip sequence is hereafter called "acquisition", a chip-wise multiplication (or XOR) operation performed for the acquisition, and addition of the code with chip subsequences "Correlation". The coded-symbol-averaged averaging of the results of the correlation with different chip subsequences is referred to as "combining".

2. In the case of imperfect synchronization between the carrier frequency generator on the transmitter side and the sampling frequency generator on the receiver side, systematic, non-signal-related, time-dependent deviations arise as the scanning signal rotates as numerical values in the complex plane can be represented. In any operation in which successive symbols are compared, this fact must be taken into account. All multiplications are considered complex in the following description. "Phase" of a value in this context denotes the argument of the complex value.

3. Specifically, these systematic variations affect DPSK demodulation, where artifacts can arise when the systematic deviation from symbol to symbol is π / 2 or more.

FIG. 2 shows the implementation of an embodiment of the method according to the invention in the portable device.

The data to be sent ("data in") are retrieved, for example, from a data memory of the portable device, for example consisting of a short bit sequence of, for example, between 40 and 500 data bits, particularly preferably 150 or fewer data bits. Sequence may be appended with an introductory or final "sync word" 31 - a bit sequence known to the receiver in advance.

As a possible alternative to the addition of a "sync word", the bit sequence may be cyclically encoded The cyclic encoding has the advantage that the method can proceed faster during decoding (as further clarified by the following description of FIG becomes).

In a further step 32, the data for the error detection and / or - correction are coded. This can be done, for example, by adding at least one CRC bit as CRC word happen, which allows a cyclic redundancy check (CRC). Instead of CRC tests, other methods can be used which can check and / or correct the consistency of the received data. Possible are systematic or non-systematic codes.

The entire sequence of bits - including the optional components "sync word" and coding bit (s) - is hereafter called "data word".

Preferably, the data word is sent continuously repetitively, at least during one transmission period, i. An uninterrupted bitstream is created. The recording of the data word by the receiver can begin at any start time, which will be explained below.

In a next step 33, a modulation with the method of the differential phase shift keying (DPSK) method is carried out in the method described here. Then, the resulting digital signal is multiplied by a - generally pseudo-random - chipping code 35 and then modulated onto the carrier signal 36. Finally, on the output side (optional), a low-pass filter 37 is arranged, through which frequencies of more than twice the carrier frequency (in the event that fc∞fc _on er) or of more than fc + fc _a never cut off _r . In the event that the chipping frequency is smaller than the carrier frequency, a bandpass filter is advantageously used instead of the low-pass filter 37. The emitted signal is designated Tx in the figure.

For the generation of the chipping code 35 and the generation of the carrier signal 36, a (single) clock generator 38 is available. Adding the syncword, calculating a CRC code, DPSK modulation, and possibly even (possibly) upsampling, spreading, and then modulating (then digitally) onto the carrier or just one or more These steps can be precalculated and do not have to be done 'online' during data transfer.

The receiver has, for example, a wake-up circuit (not shown in the figures), which determines when the input electrode has an increased signal level, which is the case when a user is in the immediate vicinity of the electrode The user acts as an antenna for electromagnetic radiation, primarily in the frequency range between 50 Hz and 100 kHz and thereby causes an increase in the "noise" level (actually, it is an interference level, since it is at the captured signals is not noise in the strict sense). Only after the wake-up circuit has detected such an increased signal level, the actual receiver electronics is put into operation.

A wake-up circuit of the receiver may also be based on other principles. For example. may alternatively be provided that the circuit is woken up by the signal, rather than by the noise / interference level. As yet another alternative, the wake-up circuit may include two switching elements. One responds to the noise / interference level while a second selectively searches for receipt of a signal, with one or the other wake-up circuit depending on the situation the detection of an increased noise / interference level by the first wake-up circuit of the receiver is activated only when the activated selective wake-up circuit detects the presence of a signal. FIG. 3 shows the input-side processing of the signal Rx received capacitively and / or resistively by the body 3 and received at the electrode. An input-side low-pass filter 41 has the same cut-off frequency as the low-pass filter 37 of the portable device and cuts off noise components which are above the cutoff threshold. After demodulation by renewed multiplication with a carrier signal 42, a second low-pass filter 43 may be arranged whose cut-off frequency corresponds to the chipping frequency fc. The obtained signal is sampled (step 44), the sampling frequency being preferably 2 * f _c . Also on the side of the read and / or read module, a clock 46 is present.

In FIGS. 4 and 6 and as a variant to FIG. 4, FIG. 5, it is shown how the method step of the acquisition can be implemented using a non-data-assisted 'combination'.

The sampling signal Sa is correlated according to FIG. 4 with the predetermined chipping code signal, which is generated by a chipping code generator 51. In the ideal case of noise-free signals with known carrier-phase relationship, the sample-wise multiplication and addition of the results yields the transmitted code bit (or the transmitted code-bit sequence for codes of more than one chip) if the sampled signal and the generated one In the figure, cascaded shift registers 52.1, 52.2 of the length of one chipping code period (ie in the preferred case corresponding to one chipping code period) are shown in phantom and approximately 0 (due to the pseudo-random nature of the chipping code) Code bit period) for the correlation of the sample signal with the chipping code. To do this, the samples are multiplied value by value with values of a register 53 containing the code. This gives two multiplies per chip when the sampling frequency equals twice the chipping frequency. If the signal-to-noise ratio is good enough, it is sufficient to use one shift register for the sample signal and one register with the code to calculate the sample rate Absolute value of the added products to effect the acquisition and the tracking: Whenever the sum obtained in the Absblutwert exceeds a certain threshold, the contents of the shift register corresponds to exactly one chipping sequence, and the (complex) sum is a code - Icon. This is illustrated in Figure 9 again for the ideal case of no-noise signals and known carrier-to-carrier phase relationship. In the figure, the sum _{Σcorr of} the results of the value-by-value multiplication, corresponding to the output of the adder 54.1, is plotted as a function of time. From the signal in FIG. 9, the sequence of code symbols 1, -1, -1 results (before decoding, ie the sequence does not necessarily represent a data bit sequence). Since the distance between the code symbols is known and corresponds to a bit period T _b , it is after the first recognition of a code symbol in principle no longer necessary to make the correlation with the sample-frequency clocking, but it is sufficient perform the calculations with a bit frequency clocking (in FIG. 9, only values of the symbols at the peaks, not the data between the peaks) are calculated, which will be described later in more detail.

In the case of problematic signal-to-noise ratios, according to a particularly preferred embodiment of the invention, non-data-assisted "combining" is proposed: Figure 4 illustrates a double "combining"; but the concept can equally well be extended to triple, quadruple, N-fold combining. For this purpose, a plurality of shift registers 52.1, 52.2 are connected in series and multiplied by the code value-by-value and the values obtained are added. For the result an estimate of the relative (compared to the respective next result) "sign" is made in order to be able to add the approximate absolute values of the results The estimation can be performed by a DPSK demodulation type operation 55 (Sign (Re (^) ^ L ₁ ))), where "sign" the signum function, "Re" the real part, "*" the complex Conjugation and r>"r ^ -i denote two consecutive values). The result of this comparison gives the relative "sign" of the two values, by multiplying one of the values with this result both have the same sign, and in a subsequent addition (adder 56) the signal components representing the code symbols are added constructively, regardless of their value while the noise is averaging, resulting in an improvement in the signal-to-noise ratio of a maximum of 3 dB The sum obtained and its absolute value (absolute value 57) no longer contain the actual data (due to the elimination of the relative "sign"). However, due to the improved signal-to-noise ratio, they are more usable for acquisition purposes. As an alternative to the DPSK demodulation-type operation, other possibilities for estimating the relative sign are also conceivable, for example decision tree-based. For example, the complex plane may be divided into sectors, the estimate being based on a comparison of the sectors in which there are consecutive values.

In the case of triple or multiple combining, the DPSK demodulation-type operation bspw takes place in each case between the output of a specific adder (for example the first adder) and in each case one output of all other adders. Alternatively, the combining can also be done cascaded in different ways, etc.

Depending on the application, it is desirable that the portable device consists of inexpensive components as possible. Then, the clock of the portable device may be chosen to be integrated with an integrated semiconductor device (chip) and may therefore be relatively inaccurate. It can deviate up to 2% from the clock of the write and / or read module. According to the prior art, such a situation can be met by systematically varying the sampling frequency in a "tuning" process until a high correlation value is obtained between the sampled input signal and the stored code. Also for the inventive method such a tuning process comes into question. However, for the important "access control" application, it is important that the entire acquisition process does not take more than a fraction of a second, and if the timer of the portable device is too inaccurate, it can not tune during that time.

With reference to Figure 5, therefore, a particular variant of Combinings and (in combination with Figure 6) of the acquisition is explained, which is particularly suitable for systems in which either the transmitter or the receiver - or both - has an inaccurate clock (no quartz) and where a rapid acquisition, such as real-time acquisition, is important. While this variant is explained here using the example of "ultra-broadband capacitive and / or resistive data transmission via the human body with the aid of the frequency spreading method and operation in burst mode", it is also suitable for any other systems in which comparable requirements play a role play.

According to Figure 5, a code bank is created by the stored code 51 is scanned with different sampling frequencies. Specifically, in the example shown here, in the sampling step 62, sampling of the code 51 is simulated with a sampling frequency of 2 * (# chips) + n (per bit length), where n varies between a (negative) minimum value minO and a maximum value maxO. Minimum and maximum values depend on the clock accuracy. In an example with 511 chips / bit and an inaccuracy of + 2%, minθ - 21, maxO-2 \, so that the correlator length is varied between 1001 and 1043. As mentioned above results due to the lack of exact match the clock of transmitter and receiver and a systematic phase rotation. By correcting the phase by a value 2π / n per carrier period (π / n per sample) this becomes compensated (phase rotation 63). The result is maxO-minO codes of length each 2 * (# chips) + n.

Of course, it is not necessary for the particular variant of the correlation described here that the series of codes (the code bank) is determined in each case on the basis of a scan. The code bank can also already be stored in the electronics and, for example, be created directly in the register 73. Also hardware-based solutions are of course possible.

For each code of the code bank, a correlation is performed with a method analogous to FIG. Again, each correlator means for a simple "Combining" or a multiple Combining (in the figure 2-fold Combining) have.

On the output side, maxO-minO signals result as a function of time, whereby as in FIG. 4, I / Q represents the complex, i. a phase information containing - signal values and Abs absolute values, possibly with the above-described improved signal-to-noise ratio by combining.

FIG. 6 shows the actual acquisition step. The correlated absolute value signal is examined for maxima as a function of time t and, if several correlators and thus several signals are present, in function of the correlator index c (ie the frequency difference) (81). Corresponding maximum values are compared with a threshold value 82 (step 83); the threshold value 82 is a maximum noise, which can also be determined from the absolute value signal itself, for example, from the noise of the other correlators or an average over all correlators. Values that this Threshold exceed are stored in a data storage element (eg, a FIFO memory with a few memory locations), wherein also the time t and - if necessary - the correlator index c is stored. In a further step 84, the time and correlator difference between stored values is compared. If stored maxima are meaningful and represent code symbols, they are present at regular intervals of a code length + a few chips and have a matching or at most a deviating correlator index. Reference numeral 85 denotes the peak search.

Once a series of such meaningful maxima has been found, the acquisition process is terminated and the time t, as well as the regular time interval T and, if applicable, the correlator index C of the "correct" correlator are forwarded to a next stage, which is referred to here as "tracking" and which is shown schematically in FIG. By "tracking" is meant a process in which the receiver timing is tracked to the transmitter, s.u. For this, the "early-late gating" can be used. The regular time interval T corresponds to the bit length (and, if 1 code per bit has been selected, to a code length). It may not be taken from the acquisition stage, but the known bit or code length (Tb) may be used.

The tracking stage has one or more tracking receivers which each process the signal of a correlator. Typically, a tracking receiver is available for the signal of the correlator C selected in the acquisition method as well as for the two adjacent correlators CI and C + \. Methods are conceivable where the CI, C and C + 1 receivers help each other to track the correlation peak position. This means that if a tracking receiver loses the signal, it receives information about the signal position from another tracking receiver. The example shown in FIG. 7 uses "gating" for tracking.

In each gating receiver, the signal absolute value of the extrapolated respectively next peak position is first compared with its neighboring values. For this purpose, incoming absolute values of the corresponding corrector are first processed. A decimator 91 selects, each time from the consolidated time t of a peak (representing a code symbol), a sequence of three values corresponding to t + T (the presumed time of the next peak), t + TA and t + T + 1 (the immediately adjacent ones) signal values). In a subsequent decision maker, four cases are distinguished:

1. The largest value of the three data points is the middle one (corresponding to t + T). Then the peak position was extrapolated correctly. The position of the next code symbol is t: = t + T.

2. The three values represent a monotone decreasing sequence. Then the next peak is earlier than extrapolated. The position of the next code symbol is t: ~ t + T-1.

3. The three values represent a monotonically increasing sequence. Then the next peak is extrapolated later. The position of the next code symbol is set as t: = t + T + \.

4. The smallest value of the three data points is the middle one (corresponding to t + T). Then it can be assumed that the peak position was completely missed. The

Peak position t is discarded for the respective gating receiver. Unless all

Gating receivers a position once or several times reject (or if only a gating receiver is present, the gating receiver discards a position), the process is completely aborted and the acq eatsungsstufe set in motion again, possibly after further correction mechanisms.

A second decimator 93 selects from the complex signal values the one which, in its temporal position t, corresponds to the absolute value selected by the decider. It can be provided that, in addition to the selected signal, the two temporally adjacent signal values are also selected and the three summed up in a subsequent adder 94. The summation of three temporally adjacent signal values makes sense if the sampling frequency is twice the chipping In embodiments with higher sampling frequencies, summation over more than three temporally successive values may also be expedient In these embodiments, the "gating" function may also include more elaborate algorithms, always following the principle that within a time window the peak of the peak is searched and, if it does not correspond to the predicted time, the prediction for the peak position is adjusted accordingly.

On the output side, a code symbol or, as a function of time, a sequence of code symbols results per gating receiver.

The decoding of the found code symbols will now be described with reference to FIG. In the case of several gating receivers, the signal-to-noise ratio is improved in a first step, if appropriate, by adding the signals representing the code symbol determined with different gating receivers. For this purpose, it is first decided for each gating receiver (decision stage 101), whether an evaluation of the code symbols makes sense. When The decision criterion is, for example, the information on how often in the above distinction the situation has occurred that the middle data point represents the smallest of the three values (case 4). If this has occurred at least once, at least twice, at least three times, ..., depending on the selected criterion, the corresponding sequence of code symbols is considered to be insignificant and, for example, the determined value is replaced by zero. Otherwise one takes msgLen + l code symbols for further processing (msgLen: number of data bits including sync word and coding (CRC) bit (s)). The additional code symbol is used for DPSK encoding. The three code symbols are combined by an adder 103, wherein there is still a comparison 104.1, 104.2 of the phases by a value 2% l {T _carr lδ * T _sa ), where δ is the deviation of the respective correlator associated with the gating receiver Index of the selected, central 'correlator C represents, T _carι = - \ lf _catτ the reciprocal of the carrier frequency and T _sa -Vf _S a the reciprocal of the sampling frequency. In the example shown, δ = Λ for the top strand (adjustment 104.1) 3 = 0 for the middle strand and δ = l for the lowest strand (alignment 104.2).

The resulting sequence of summed code symbols may be executed "off-line" - that is, unlike the stages described above, not necessarily in the data transmission real-time - by a DPSK demodulator 106 from the summed code symbols - These represent phase difference data signals, because of the input-side DPSK modulation 33 - Generates a bit stream For DPSK demodulation can here also the formula Sign (Re (T _5. • rζ_ _x )) or another, for example. based approach.

In connection with the DPSK demodulation may still be the problem that due to the uncertainty described above regarding the Phase between two consecutive code symbols, a rotation of the phase by more than π / 2 takes place. Then the formula Sign (Re (T ₄ .T ^ ₁ )) does not yield the transmitted data bit sequence but as artifact a non-meaningful symbol sequence. For this reason, the symbol sequence of length msgLen is first taken (107) and examined for consistency by means of error detection and correction, for example error test and / or correction 111. If a cyclically decodable code is used, the error test consists of a single calculation on the present symbol sequence, even if the order is not known. If this is not the case, the code symbol sequence must be rotated maximally msgLen times (112) as shown in order to get to the correct starting point. As soon as the error test displays the consistency of the symbol sequence, this is the searched bit sequence. The method with error test 111 and possibly multiple rotations 112 can also be regarded as a cyclic decoding.

Additional measures can be provided with which the plausibility of the received symbol sequence is examined, for example by reserving at least one bit for a consistency test in addition to the one described above in the data word. The test can be designed as a CRC test or equivalent test. Such a test can detect if, for example, one of the decoded code symbols is wrong. This can also reduce the likelihood that a code symbol will be corrected in the wrong direction due to false error correction.

The symbol sequence is passed on for further processing (search of the sync word, described below). Otherwise, the phase between two code symbols is rotated by ΔΦ _m = mπ / 2 for m = 1, 2, 3 and the DPSK demodulation 106 and the subsequent consistency test are carried out again until a consistent symbol sequence - the searched bit sequence - has been found , In the Rotation ΔΦ _m is a relative rotation, ie the phase of each code symbol is rotated by a value ΔΦ _m further than the previous one, ie, for example, for m = 1, the phases of consecutive symbols are around 0, π / 2, π, 3π / 2, etc. rotated.

This phase rotation ΔΦ _m is actually a fine correction, which is carried out in addition to the (coarse, ie sample-wise) phase rotation 63 according to FIG. 5 and takes into account the fact that even the most suitable correlator found has a small deviation exists between the transmitter and receiver side clocking, which manifests itself in a phase rotation between the code symbols. The necessity of a phase rotation ΔΦ _m can be avoided if the correlators of the correlator bank are chosen closer to one another than described above, so that a fine correction is no longer necessary.

From the data bit string, it is not known a priori which bit of the data will start, i. the desired information can not be taken a priori.

Because the data is constantly being transmitted repeatedly and the start time of the receiver-side signal acquisition is not known in advance, the actual beginning of the data bit sequence must be determined in a first step. For this purpose, a syncword is preferably used for the search of the starting point, as has already been described at the outset. The sync word is searched for by the msgLenic rotation 121 of the bit sequence 122. As soon as it has been found, the sync word and possibly the coding bits are removed (strip

123) and the remaining bits are forwarded as data (Data out) for further processing. The method of searching for the sync word and, if necessary, multiple rotations 121 may also be considered as an anticyclic (ie, based on the identification of a well-defined initial point of the message) decode become. The procedure described here works if the pattern of the sync word does not occur in the data bits or only with very little probability.

Depending on the embodiment, the error test can only display consistency if the sequence of the code symbols has a unique, fixed starting point (completely anticyclical error test). In such a case, searching for the sync word (and thus the sync word itself) is not necessary. Then, the starting point is recognized directly by the error test. In this case, the starting point of the data word results from the code results (comma-free codes).

On the other hand, even if there is a sufficiently long sync word, the CRC test can be omitted. The consistency test, which is necessary to eliminate the above-mentioned phase-related uncertainty in DPSK demodulation, is then limited to a search of the syncword bits, which in turn includes the maximum msgLen-amount rotation of the bit string. If the sync word is found, consistency is assumed and at the same time the starting point is found. However, this embodiment involves a reduced security, since a bad bit is not recognized, and error corrections are not possible. The embodiment is therefore particularly suitable for systems that are robust to data transmission errors.

As already mentioned, the data word can also be cyclically coded. Then the search for the sync word is omitted.

An inventive system for the transmission of information is shown in FIG. Between a portable device 1 and a write and / or Read module 2 can be capacitively and / or resistively exchanged over the body of a user 3 data. For this purpose, the portable device 1 has two electrodes 201, 202, between which a time-dependent electrical voltage can be applied by a signal generator 203. In the figure, also a data memory 204 is illustrated, which contains data to be transmitted. The portable device is worn, for example, by the user on the body, for example in the pocket. In this case, the first electrode 201 is closer to the body than the second electrode 2O2. Under certain circumstances, the first electrode can also touch the body directly. The writing and / or reading module 2 has a detector 213 (in the simplest case essentially consisting of an amplifier, but the detector can also be more complicated and have, for example, discriminating means, etc.) which has a voltage between a first receiver electrode 211 and a second receiver electrode 212 can detect. The first receiver electrode 211 may be formed as an operating or contact surface. It may be part of the read and / or write module, or the module may also comprise only contact means (eg, a wire contact) to a non-module metallic body, such as a door handle or door knob, which serves as the first electrode acts. Also, the second electrode is not necessarily a component of the module itself. If the user 3 is in the immediate vicinity of the first receiver electrode - for example by a finger touching or almost touching it - a time-dependent voltage can be induced between this and the second electrode, which depends on the voltage between the electrodes of the portable device. This principle is known per se from the introductory mentioned writings and will not be explained further here. Voltage signals detected by the detector may be processed by a data acquisition and decoding unit 215 in the manner described above.

The embodiments described above are mere examples of how the invention can be implemented. Other implementations are possible without having to forego the essential advantages of the invention. First Example is. mentioned that the acquisition - with the aid of a correlator bank or even with a simple correlator - is also possible without multiple "combining", and then no estimation of the relative sign is possible, for example by means of DPSK demodulation-like operation for the determination of the relative "sign". necessary.

Other methods of digital data modulation than BPSK can also be used, for example QPSK or others. Moreover, the relationship between code length and bit length proposed here is by no means a necessity; other, preferably defined relationships between a bit (symbol) length and a code are possible.

While acquisition by non-data-aided combining has been described above, the invention is also feasible with data-aided acquisition methods.

Finally, it should be mentioned that the terms "portable device" and "read and / or read module" have been chosen for a better understanding of the invention and the

Arrangement of the corresponding elements not binding. In particular, it can also be provided that a device to be worn by the user has a

Has receiver for the capacitive-resistive coupling and as writing and / or

Read module is used. Alternatively and / or additionally, the device provided with the transmitter may also be at least temporarily bound to a location, i. it does not have to be portable at all times.

Claims

PATENT CLAIMS

A method of transferring data between a device and a read and / or write module, wherein the data is capacitively and / or resistively transferred to a read and write module by the device using the frequency spreading method as an ultra-wideband signal ,

2. The method according to claim 1, characterized in that a direct-sequence frequency spreading (direct sequence spread spectrum) method is used.

3. The method according to claim 2, characterized in that the length of a code code is selected as the length of a frequency spreading code.

4. Method according to one of the preceding claims, characterized in that a maximum amplitude of the ultra-wideband signal is 5 V or less.

5. The method according to any one of the preceding claims, characterized in that the data to be transmitted are first modulated with a method of digital data modulation and then frequency-spread.

6. The method according to claim 5, characterized in that the binary Phasenumtastung is used as a digital data modulation method.

7. The method according to claim 5 or 6, characterized in that the data modulation method is combined with a coding, which makes the signal insensitive to the absolute phase.

8. The method according to claim 7, characterized in that the coding in combination with the data modulation method corresponds to a differential Phasenumtastung.

Method according to one of the preceding claims, characterized in that the signal received by the read and / or write module is sampled, whereby a sequence of samples is generated and that subsequences of the sequence of samples are correlated with a series of values representing the represent stored frequency spread code.

Method according to claim 9, characterized in that a plurality of consecutive subsequences of the sequence of samples are correlated with the sequence of values at the same time, and that results of the

Correlation calculations for. the data acquisition can be combined.

11. The method according to claim 10, characterized in that the combination of the results is not supported by data.

12. The method according to claim 11, characterized in that a value is estimated for the combining, which is characteristic of the

Data content is in each case two results of the correlation calculations, and that one of the two results is corrected with this value and added to the other two results.

13. The method according to any one of claims 9 to 12, characterized in that the subsequence or the subsequences of the sequence of samples is / are correlated in parallel with different sequences of values.

14. The method according to claim 13, characterized in that the different sequences of values correspond to the frequency spread code sampled at different sampling frequencies.

Method according to claim 13 or 14, characterized in that, in the absence of correlation between the subsequence and each of the different sequences of values, the subsequence is correlated with either a further group of sequences of values corresponding to the with matched sampling frequencies, or the sequence of samples is recreated at a new sampling frequency and correlated again with the various sequences of values

16. The method according to any one of claims 5 to 8 and one of claims 9 to 15, characterized in that the correlation of the received signal with the stored frequency spreading code is used in a reversal process of the digital data modulation method for synchronization.

17. The method according to claim 16, characterized in that results of the correlation of the received signal with the stored frequency spreading code directly code symbols are removed.

18. The method according to any one of claims 9 to 17, characterized in that for the timing acquisition at least two criteria on the results of

Correlation calculations are applied, of which a first criterion is a comparison of the amount of the results with a noise level and a second criterion a comparison of the time interval of a

Amount maximum to a last amount maximum with a bit length includes.

19. The method according to any one of claims 9 to 18, characterized in that a sampling frequency is adjusted depending on the result of the correlation calculation, whereby a tuning and / or a fine tuning is achieved.

20. The method according to any one of the preceding claims, characterized in that a center frequency of the ultra-wideband signal is not greater than 2 MHz.

21. The method according to any one of the preceding claims, characterized in that the transmitter continuously emits the data repetitively, such that immediately after the end of the transmission of a data word, the transmission of the data word starts again.

22. The method according to any one of the preceding claims, characterized in that the data comprise at least one data bit for a data consistency test and / or for error correction.

23. The method of claim 2, wherein the data is transmitted differentially phase-shifted, and wherein the signal received from the write and / or read module after a frequency spread inversion process as a

Symbol sequence is shown, characterized in that means of the

Data consistency a symbol sequence subsequence of the length of a data word is checked and in the absence of consistency, a phase rotation between each two adjacent symbols is performed.

24. The method according to any one of the preceding claims, characterized in that the data have a previously known sequence for the synchronization.

25. The method according to any one of claims 1 to 23, characterized in that the data are transmitted cyclically and / or anticyclically coded.

26. The method according to any one of the preceding claims, characterized in that the length of the frequency spreading code corresponds to an integer multiple of a data bit length and that the

Frequency spreading takes place synchronously with a data bit sequence representing the data.

27. The method according to any one of the preceding claims, characterized in that the ultra-wideband signal via the body of a User or via direct capacitive and / or resistive coupling between transmitter and receiver is transmitted to the receiver.

28. System for the transmission of data, comprising at least one device and at least one writing and / or reading module, the device having two electrodes (201, 202) and a signal generator (203), by which between the two electrodes (201, 202) is a time-dependent electrical voltage can be applied, and wherein the writing and / or reading module, a detector (213), with which an electrical voltage or an electric current between a first and a second receiver electrode (211, 212) is detectable, and a A data acquisition and decoding unit (215) for detecting data from a signal detected between the first and second receiver electrodes (211, 212), characterized in that the signal generator (203) is programmed and / or driven to provide an ultra Broadband signal is generated by which data using the frequency spreading method are shown and that the

Data acquisition and decoding unit (215) comprises means for recovering the data from the ultra-wideband signal.

29. Device for the transmission of data to a read and / or read module, comprising two electrodes (201, 202) and a signal generator (203), by means of which a time-dependent electrical voltage can be applied between the two electrodes (201, 202), characterized in that the signal generator (203) is programmed and / or driven so that an ultra-wideband signal can be generated by it, are represented by which data using the frequency spreading method.

30. A writing and / or reading module for receiving data which can be transmitted capacitively and / or resistively by a device, comprising a detector (213), with which an electrical voltage or an electric current between a first and a second receiver electrode (211 , 212), and a data acquisition and decoding unit (215) for detecting data from a signal detected between the first and second receiver electrodes (211, 212), characterized in that the data acquisition and decoding unit (215) comprises means for recovering comprising data from an ultra-wideband signal, by which data is represented using the frequency spreading method.

31. Writing and / or reading module according to claim 30, characterized by a

Wake-up circuit through which the data acquisition and

Decoding unit (215) is activated as soon as a noise and / or interference level has reached a certain value and / or as soon as the

Arrival of a signal is detected.