Field of the Invention The present invention relates to remote-access apparatuses, especially to methods, and devices for monitoring the usability of an item, wherein the usability is related to the environment in which a remote-access apparatus is or has been present. Background of the Invention

Modern logistics was to a great degree enabled by the development of the product code system and bar codes. Products and pallets could be quickly identified by reading a bar code with an optical reader, and the tracking of items in the logistic chain became feasible. Furthermore, the products could be counted and sold at the point of sale more quickly and more reliably. Beyond consumer product logistics, industrial material management systems, postal and courier services, healthcare and bio-analytics systems and many other fields of human activity have benefitted from the use of bar codes.

Over time, the needs of having various kinds of information available on a product outgrew the capabilities of a simple bar code. To this end, new technologies were developed, such as two-dimensional bar codes. Of such technologies, radio frequency identification (RFID) has rather quickly become the technology of choice for identifying and tracking items. RFID technology has a vast number of applications making use of the ability to read an RFID tag from a distance even without a line-of-sight connection between the reader and the tag. RFID is quickly replacing or at least augmenting the bar code technology in many places.

The basic information that an RFID tag carried has traditionally been electronic product code (EPC) information and/or tag identification information. RFID tags with more capabilities may have the ability to store more information in the tag memory, and even carry out some simpler processing of data. A radio frequency identification tag (RFID tag) typically comprises an RFID chip and an antenna connected to the chip, attached to or protected inside a plastic layer. The RFID chip contains analog and/or digital processing circuitry, e.g. a state machine and/or a processor and memory for storing information, as well as circuitry for receiving radio frequency (RF) energy and modulating the RF wave. The chip is connected to the antenna, whose shape and size depend on the frequency at which the tag operates and the application for which the tag is used. The chip and the antenna are often laid inside a thin plastic container for protection. The tag as a whole is typically flexible to a certain degree, especially the passive tags that do not contain a power supply.

Passive tags use the energy from the radio-frequency electromagnetic field of the read-out signal to power the operations that the tag carries out. Passive tags operate essentially while they are in the reader field, and are essentially inactive at other times. Therefore, the farther the reader device and the weaker the read-out signal, the less energy is available for the tag to use. This in turn means that the tag cannot perform very complex operations that would require a lot of power, since such tags could only be read from a very short distance away. In other words, the read-out distance of a tag is to a large extent determined by the power consumption requirements of the tag. Naturally, the attenuation and power distribution of the electromagnetic signal emitted by the tag is another concern. To tackle this issue, some passive tags have a battery that may be used to power the processing circuitry and thereby allow longer operating range for the tag. Such tags that use an internal power source to energize the response signal may be called semi- passive tags.

Some applications may necessitate more complex processing and/or more demanding properties of the RFID tag. It may become necessary to arrange more internal power such that these operations may be carried out. Some tags may require an internal power source to communicate with a longer range. Such tags may be called active tags. It may also be necessary to provide internal power so that operations can be carried out at times when the power from the external radio frequency electromagnetic field is not available. Usually this is arranged by means of a battery attached to the RFID tag. The battery may power sensors, memory, processors and/or a transmitter of the tag so that sensing, data logging and processing may be carried out at any time. Such more complex tags may be more complicated to manufacture, more expensive and/or more prone to malfunction than passive or semi-passive tags. Moreover, any tags with an internal power supply may have a limited lifetime, since they cannot be operated properly or at all when the battery runs out. Another practical problem is that when an active tag is taken into use, the battery operation needs to be activated. Otherwise, the battery would be in use e.g. already starting from manufacturing of the tag and prior to actual active use.

An example of demanding properties of a tag are measurements. In some applications, wireless measurements regarding the environment may help process or quality control. Active tags are often larger and more expensive than passive tags. Furthermore, in some applications, particularly related to construction engineering, the battery of the active tag may be difficult to change. Therefore, e.g. for process or quality control, passive tags capable of measurements are needed. Often such a measurement is performed to monitor whether a limit has been exceeded or not. One example is in food chain logistics, where temperature should be low enough to ensure preservability of the food items, i.e. the items should remain frozen. In other examples, freezing should not perhaps let to occur at all in order to maintain the properties of the item. Sometimes the user is interested on an integral of a measured value. Both these require a lot of remote-apparatus specific information such as calibration information, information on the limit or limits, or information on the integrated quantity. Thus, it may be difficult to arrange the use of an advanced functionality of a remote-access device e.g. if the use of the advanced functionality requires calculations and/or knowledge of the operation of the whole system and/or chain of operation.

Summary of the Invention Now there has been invented an improved method and technical equipment implementing the method, by which the above problems are alleviated. Various aspects of the invention include a method, an apparatus, a server, a client and a computer readable medium comprising a computer program stored therein, which are characterized by what is stated in the independent claims. Various embodiments of the invention are disclosed in the dependent claims.

In one embodiment, the need for data storage and computing resources is reduced by setting a threshold value for a quantity and comparing the value of quantity with the threshold value for indicating whether an environment condition has been exceeded or not. The threshold value may be assigned for a quantity measurable with a remote-access apparatus. In case the threshold is assigned to the measurable quantity, calibration data for remote- access apparatuses is not necessarily needed. In another embodiment, information indicative of the threshold is stored in the memory of the remote- access apparatus. In one embodiment the threshold value, in terms of the measurable quantity, is stored in the memory of the remote-access apparatus. In one embodiment, a property of a measured distribution is compared to threshold, and in one embodiment, a calibrated remote-access apparatus is used together with a number of uncalibrated remote-access apparatuses. In case comparisons are done for the measurable quantity, the measurements are not easily tampered, since the relation between the measurable quantity and the environment variable in not known.

There may be multiple threshold values, such as both upper and lower thresholds, and one or more of the threshold values may be associated with a time-integral of rate variable that depends on the measurable quantity, whereby an accumulated value of a state variable is compared with the threshold. The threshold values may create a usable range or a number of usable ranges. A threshold value may also be given as a function of a measured value.

The comparison may be carried out by the remote-access apparatus, by the reader, or by a computer in a system where the data from the apparatus is sent for comparison. The remote-access apparatus may indicate just "good/failed", or give more information. Description of the Drawings

In the following, various embodiments of the invention will be described in more detail with reference to the appended drawings, in which

Fig. 1 shows a reader and a transponder,

Fig. 2 shows a reader, a data storage unit, and several transponders attached to items,

Fig. 3 shows an example of the relation between the frequency of a local oscillator in a transponder and the temperature of the transponder, Fig. 4 shows an example of the variation of temperature as function of time, and two threshold values,

Fig. 5 shows the example corresponding to figs 3 and 4 of the variation of frequency as function of time, and two threshold values,

Fig. 6 shows the integral of temperature exceeding the threshold 401 in

Fig. 3,

Fig. 7 shows the integral of frequency going under the threshold 501 in

Fig. 4,

Fig. 8 shows the integral of temperature going under threshold 41 1 in

Fig. 3, Fig. 9a shows three distributions of the measured quantity,

Fig. 9b shows three reference distribution of the measured quantity,

Fig. 9c shows three linear combinations of the reference distributions of

Fig. 9b, Fig. 10 shows a reader and a transponder equipped with a display, and Figs. 1 1 a— 1 1 i

show examples of displayed data on a display of a remote access apparatus.

Detailed Description of the Embodiments

In the following, several embodiments of the invention will be described in the context of radio frequency identification (RFID) tags. It is to be noted, however, that the invention is not limited to RFID tags and systems only. In fact, the different embodiments have applications widely in any environment where advanced functionalities and memory capabilities of small devices with limited power supply are needed.

It has been noticed in the context of this invention that certain advanced applications of RFID tags would require active RFID tags (such tags that have a power supply), and that this would in turn make the tags more complicated and more expensive. Alternatively, straightforward implementation of advanced functionality on passive RFID tags would increase the power consumption of the tags and thereby reduce the effective reading range or operational range of the tag. Various embodiments of the invention are envisioned to be used with RFID tags that are energetically essentially passive, that is, with tags that operate essentially while being in the reader field and being able to draw energy from the field. In this description, RFID inlays and tags are called remote-access apparatuses.

In various embodiments of the invention, some of the processing required by the advanced functionality of a passive RFI D tag may be carried out in the reader device or elsewhere in the system. This makes it possible to reduce the power consumption of the RFID tag and for the tag to have an improved operational range. The invention may help to keep the tag electronics simple and thus usable in weaker reader fields and/or longer reading ranges. On the other hand the invention may also reduce need for powerful readers to compensate for the tag power consumption. The latter benefit may be significant in some environments with many readers in the same area and/or where interfering electromagnetic fields need to be kept to a minimum.

A feature of the present invention is to reduce the computing needs in a system comprising remote-access apparatuses capable for measurements. Figure 1 shows an example of an RFID tag 102 and an RFID reader device 150. The RFID tag comprises an RFID inlay 100 and a substrate 130. Furthermore, the RFID inlay comprises an RFID chip 1 10 and an antenna 140. The chip 1 10 may comprise a radio frequency unit RXTX1 , a control unit CNT1 , and a memory MEM1 . The radio frequency unit RXTX1 may comprise a signal receiver RX1 , and a signal transmitter TX1 . The radio frequency unit RXTX1 may comprise connection terminals T1 , T2, which may be connected to at least one antenna element 140. The antenna elements may from e.g. a dipole antenna or an inductive antenna (coil antenna). The radio frequency unit may send signals Sin to the control unit and receive signals Sout from the control unit. The control unit may receive data DATA from the memory, and write data into the memory. The radio frequency unit RXTX1 may comprise a voltage supply VREG1 , which is arranged to extract operating power from an incoming radio frequency signal. Both the control unit and the memory may receive their operating power from the voltage supply. Therefore, the tag may be essentially passive, i.e. it does not comprise an energy source. Moreover, the RFID chip 1 10, the chip comprises an oscillator 52 to produce a clock frequency †CLK for the radio frequency unit. The clock frequency may depend on the frequency of the local oscillator f.

The transponder may be substantially passive, i.e. energy for operating the radio frequency unit RXTX1 , the control unit CNT1 , the local oscillator 52, and the memory MEM1 may be extracted from a radio frequency field. Energy for operating the radio frequency unit RXTX1 , the control unit CNT1 , the local oscillator 52, and the memory MEM1 may be extracted an interrogation signals ROG sent from a readers.

A passive transponder 100 may comprise a capacitor or a rechargeable battery for storing operating energy extracted from an interrogation signal. Operating energy for operating the radio frequency unit RXTX1 , the control unit CNT1 , the local oscillator 52, and the memory MEM1 may be extracted from one or more interrogation signals sent from one or more readers.

The local oscillator 52 may be used to measure a value of an environment variable, such as temperature, when the frequency of the local oscillator depends on this variable. In other embodiments, the RFID transponder may comprise another oscillator or a temperature sensing unit for this purpose.

A carrier frequency of the response may be modulated at a modulation frequency †LF- The modulation frequency LF may also be called as a "link frequency".

The modulation frequency †LF of the response may, in turn, depend on the clock frequency †CLK of a local oscillator 52. Thus, also the modulation frequency †LF may depend on the temperature of the chip. A change of the modulation frequency †LF may indicate a change in the temperature. Consequently, the modulation frequency †LF may be interpreted to be temperature data. The local oscillator 52 may be e.g. a ring oscillator. A ring oscillator may comprise e.g. a plurality of cascaded logical gates whose operating speed depends on the temperature. The local oscillator 52 may be e.g. a relaxation oscillator. The transponder 100 may be arranged to determine a frequency parameter NC, which depends on the frequency of the local oscillator 52. In particular, the frequency parameter NC may indicate the number of pulses of the local oscillator 52 corresponding to the duration of a frequency-setting parameter. In particular, the frequency parameter NC may be substantially equal to the parameter BLF (backscatter link frequency), as defined in the EPC Gen2 protocol. The frequency-setting parameter may refer to the parameter TRCal, as defined in the EPC Gen2 protocol.

The frequency parameter NC comprises information about the temperature when the frequency of the local oscillator 52 depends on the temperature. The transponder 100 may be arranged to operate such that a response sent by the transponder comprises a binary number corresponding to the value of the frequency parameter NC. The frequency parameter NC may be stored in a memory MEM1 . The transponder 100 may be arranged to operate such that a response sent by the transponder comprises a binary number corresponding to the value of the frequency parameter NC. The transponder 100 may be arranged to operate such that the value of the frequency parameter NC is sent as a binary number only when the interrogation signal contains a request for said value.

The transponder 100 may comprise a temperature sensor, which is different from the local oscillator 52. The temperature sensor may comprise e.g. a P-N junction, a resistive element, whose resistance depends on the temperature, or a thermocouple. The resistive element may be e.g. a NTC or PTC resistor (NTC refers to negative temperature coefficient, and PTC refers to positive temperature coefficient. The resistive element may be a Pt100 sensor.

The temperature sensor may be powered by energy extracted from a radio frequency field. The temperature sensor may be powered by energy extracted from a radio frequency field of one or more interrogation signals. The transponder may be arranged to convert an analog signal provided by the sensor into temperature data. The temperature data may be stored in a memory MEM1 e.g. as binary data. The transponder 100 may be arranged to transmit the temperature data to a reader e.g. when requested by an interrogation signal.

An interrogation signal sent from a reader to a transponder 100 may comprise a frequency-setting parameter TRcal (reference is made to the EPC Gen2 protocol). The transponder 100 may be arranged to set a modulation frequency ("link frequency") fl_F based on the value of the parameter TRcal. The value of the TRcal may be directly proportional to the temporal duration of the data sequence TRcal. The value of the parameter TRcal may be e.g. 50 με.

The transponder 100 may be arranged to set the modulation frequency †LF according to the following equation:

PR L ^{F ~} TRcal

The modulation frequency †LF may also be called as a "backscatter link frequency".

In practice, the transponder may be arranged to calculate the modulation frequency †LF by using integer numbers as follows: f ^{DR '} f ^CLK

^LF ROUND(TRcal ^■ f _CLK ) where DR denotes a division ratio parameter. The value of the division ratio parameter DR may be e.g. 8 or 64/3. † _C LK denotes the frequency of the local oscillator 52. ROUND denotes a rounding or truncating function, i.e. it rounds or truncates an arbitrary number format to an integer number.

When the value of the frequency-setting parameter TRcal is increased, the modulation frequency f _LF may decrease in several (abrupt) jumps J1 , J2, .., as can be derived from the equation. The modulation frequency f _LF may be substantially constant between TRcal values corresponding to two adjacent jumps J1 , J2, provided that the clock frequency † _C LK is constant.

When the value of the frequency-setting parameter TRcal is varied by a small amount in the vicinity of a jump, the clock frequency † _C LK being substantially constant, the modulation frequency f _LF may be abruptly changed from the value f _LF1 to the value f _LF2 . It may be derived from the above equation that

In other words, the clock frequency † _C LK may be calculated from the upper modulation frequency † _L FI and lower modulation frequency † _L FI associated with a single jump.

A first response modulated at the first frequency † _L FI may be provided by sending a first interrogation signal from a reader to the transponder 1 00 such that the first interrogation signal comprises a first frequency-setting parameter TRcaM . A second response from the same transponder 1 00 modulated at the second frequency f _L F2 may be provided by sending a second interrogation signal from a reader to the transponder 1 00 such that the second interrogation signal comprises a second frequency-setting parameter TRcal2.

The time period between sending the first and second interrogation signals may be selected to be so short that the temperature of the local oscillator is not significantly changed during said time period.

Thus, the method for monitoring the temperature may comprise:

- sending a first interrogation signal and a second interrogation signal to the RFI D transponder (100),

- receiving a first response signal from the RFI D transponder (1 00) at a first modulation frequency {† _L FI) and a second response signal from the RFI D transponder (1 00) at a second modulation frequency( f _LF2 ),

wherein the first response signal is a response to the first interrogation signal and the second response signal is a response to the second interrogation signal,

- determining a clock frequency † _C LK from the first modulation frequency( /. _F i) and the second modulation frequency( f _LF2 ).

The determined clock frequency † _C LK comprises temperature-dependent information. For example, a change in the clock frequency c _/ may indicate a change in the temperature. Therefore, the clock frequency †CLK may be interpreted to be temperature data TDATA as such.

The relationship between clock frequency †CLK and absolute temperature values may be established e.g. by calibration measurements.

In addition to height of the frequency jump, a frequency-setting parameter that matches with a jump, may also depend on the environment variable. For example, denoting the frequency-setting parameter that matches with the jump number 1 by TRcaM , TRcaM may depend on the frequency of an oscillator, and thereby also on the value of the environment variable.

As discussed, both the clock frequency and the a frequency-setting parameter that matches with a jump may be measured using a remote- access apparatus. These quantities will thus be called measurable quantities.

The RFID tag may comprise information on whether or not the tag is capable of an advanced functionality. This information is called the advanced functionality code (AFC). In addition to AFC, the tag may comprise further information related the advanced functionality. This information is called advanced functionality data (AFD). The advanced functionality code AFC and the advanced functionality data AFD may for example be such that the AFC informs the reader that the tag is suitable for temperature measurement, and the AFD provides information to be used in the determination of the temperature, such as calibration information. The AFC and/or the AFD may be stored wholly or partially in a password memory of the tag. Naturally, in case the tag contains memory apart from the password memory, AFD or part of AFD can be stored also in that memory. Calibration information means information that can be used to relate a measured value of a measurable quantity to a value of an environment variable, such as temperature. For example, temperature may change the frequency of a ring oscillator, and therefore the oscillation frequency (or oscillation period) of the oscillator may be used to determine the temperature. It is also possible, that the frequency may depend on the strain. Alternatively, if the chip is in a strained state, internal stress is induced in the chip. Moreover, external pressure may induce these internal stress and strain. Therefore, such environment variables may include temperature, pressure, stress and strain.

Furthermore, since the frequency may be a function of temperature, the temperature may be also be a function of the frequency of the local oscillator. Thus, calibration information may comprise e.g. the coefficients of a polynomial that relates the frequency and the temperature. Calibration information may be obtained by making calibration measurements. A feature of the present invention is to reduce the need for calibration information. Various embodiments are disclosed, in which a remote-access apparatus that has not been calibrated can be used to determine an environment dependent usability of an item.

Figure 2 shows an apparatus 700 for monitoring an environment variable, such as temperature. The apparatus comprises one or more RFID transponders 100, a reader 150, and an operation unit 400. In an embodiment, the reader device may comprise the operation unit. The transponders 100a, 100b, 100c send signals, which allow monitoring of the temperatures of the items 300a, 300b, 300c. The item 300a may be e.g. an electronic device, battery, hard disc drive, a package containing an item, a package containing foodstuff, a package containing medicine, or a package containing a chemical substance. The signals may be received by a reader 150. The reader 150 may comprise a body 152 and a user interface 151 . The user interface 151 may comprise e.g. a visual display for visually displaying temperature or usability information determined from signals sent from the transponders 100a, 100b, 100c. The user interface 151 may comprise e.g. a keyboard or a touch-sensitive screen for receiving data and/or commands.

The apparatus 700 may determine the value of an environment variable as follows. With the reader device 150, some property of a remote-access apparatus, an RFID tag 100, may be directly measured. A quantity may be determined from at least one value of the directly measured property such that this relation does not depend on the measurement environment or the individual properties of the remote-access apparatus. Such quantities are called measurable, since they can be measured, i.e. determined from results of direct measurements without calibration. As an example, a directly measured property may be the backscattering frequency of a remote-access apparatus. From at least two of such values, one may determine a measureable quantity, such as the frequency of the local oscillator 52 of the remote-access apparatus. In addition, the backscattering frequency of the remote-access apparatus may change abruptly in response to changing the TRcal value used in the communication between the remote-access apparatus and the reader device. In this case, the backscattering frequency is directly measurable, and the TRcal value, at which this change occurs is measurable, since it may be determined without calibration parameters. In addition, a frequency parameter may be stored in a register of a remote- access apparatus.

Furthermore, the measurable quantity may depend on an environment variable. Therefore, in some cases the value of the environment variable may be determined, once the relation between the measurable quantity and the environment variable is known. This relation can be made known by calibrating the remote-access apparatus or a number of remote-access apparatuses. After calibration, this measurable quantity may be used to determine e.g. the temperature.

The value of an environment variable may change the value of the measurable quantity. However, the dependence between the measurable quantity and the environment variable is not generally known a priori, but needs to be found out by calibration. After calibration, the value of the environment variable may be determined based on the measured quantity. Without loss of generality, the measured quantity will be denoted by and the environment variable by T. Calibration of the remote-access apparatus means that a relation between f and T should be known with reasonable accuracy. Also quite generally, some function may be used to describe this relation, preferably a function h such that T=h(f).

An example of a relation between the frequency of the local oscillator and the temperature 7 ^~ is shown in Fig. 3. The relation is given as a figure of the function that converts a value f of a measurable quantity to a value of an environment variable Γ, the value of the measurable quantity being measurable using said remote-access apparatus. I.e. the frequency of the local oscillator is shown on the horizontal axis and the corresponding temperature on the vertical axis. It is emphasized, that the relation, as described in Fig. 3 serves only as an example, and does not necessarily resemble any actual remote-access apparatus. It is also noted, that the frequency may increase as the temperature decreases.

When such a remote-access apparatus is used to measure the environment, the environment or the value of an environment variable may change as function of time. An example of a temperature as function of time is shown in Fig. 4. The corresponding frequency of local oscillator is shown in Fig. 5 as function of time. The data of Fig. 3 is used to relate the frequency and the temperature shown in Figs. 4 and 5. The variation of the temperature may be due to variable temperature in a fixed location, or due to transportation, where the remote-access apparatus may be exposed to different environ- ments. In many applications, such as logistics, food chain, or construction, the user is often interested, whether the temperature (or the value of another environment variable) is within a predetermined usable range or not, or how often or how long has the remote-access apparatus been in a forbidden environment.

Some examples of such ranges include: (1 ) a threshold for maximum allowed temperature. This type of threshold is typical in food logistics. The range can be e.g. a few degrees of Celsius for fresh food, or around minus 10 °C for frozen items. (2) a threshold for minimum allowed temperature. This type of threshold is typical also in food logistics, but also in many other liquid transportations. For example, products comprising a lot of water should not be frozen, and therefore a lower threshold of around 0 °C might be applicable. Such a threshold applies also to e.g. vegetables and some other fresh food. (3) a combination of an upper threshold and a lower threshold. For example, some fresh food items should be kept in between around 0 °C and 4 °C. The allowed environment would thus be between 0 °C and 4 °C. (4) another combination of an upper threshold and a lower threshold. For example, some items may be stored in a hot environment or a cold environment, but not in between these temperatures. As an example, some prepared food can be stored in a hot environment, where the temperature is above 60 °C, or in a cold environment, where the temperature is below 6 °C, but not in between these temperatures.

A straightforward way to monitor the state of an item would be to first calibrate each remote-access apparatus, store the calibration data, determine and store the range, make measurements concerning the environment variable using the remote-access apparatus and calibration data, and make the comparisons and analysis. This process would yield accurate estimates of the item's state and on the remaining life of the item. It has been found that this type of approach would, however, require many calibration measurements, relatively large amount of data, and some computing capacity in the reader device. Moreover, it has been found that use of the remote-access apparatuses memory and some approximations in the determination of the state of the item decreases both the computing and the memory needs of a monitoring system, and may still result in reasonable accurate life time estimates.

Some products may be considered unsatisfactory, if the environment value has exceeded or gone under a threshold value. This type of threshold will be called a hard threshold. This is shown in terms of temperature in Fig. 4. By using an upper threshold value 401 of 7 ^~ _? =22.5 °C all the values 403 would have exceeded the threshold, and an item could be considered unsatisfactory based on these measurements. In particular, with the hard threshold, the item could be considered unsatisfactory at time 3, which is the time for the first measurement 405, where the temperature exceeds the threshold 401 . Also, by using an lower threshold value 41 1 of 7 ₂ =15 °C all the values 413 would have gone under the threshold, and an item could be considered unsatisfactory based on these measurements. In particular, the item would be considered unsatisfactory at time 1 1 , which is the time for the first measurement 415, where the temperature goes under the threshold 41 1 .

These thresholds are shown in Fig.5 for the frequency. The lower threshold 501 for the frequency (at 50.40 kHz) corresponds the upper threshold 401 for the temperature (at 22.5 °C), and the upper threshold 51 1 for the frequency (at ^=153.88 kHz) correspond the lower threshold 41 1 for the temperature (at 15 °C). This type of measurement, where a single value can be compared to a threshold value to determine whether an item is satisfactory or not are applicable typically e.g. for items that must be kept e.g. in a liquid form, such as unfrozen and unvaporized. Some items may be considered unsatisfactory if a process, such as a degradation process, has proceeded too much. The state of the process can thus be considered indicative of the state of the item. Moreover, the state of the process characterizes how far the process has evolved. Therefore, a threshold value may be applicable for the state of the item such that the item can be considered unsatisfactory, when the value of a state variable has exceeded a threshold. The rate, at which the process evolves may be characterized by a rate. For example, the rate / of many chemical reactions, such a degradation processes, may be described with the Arrhenius equation:

where A and E are constants for the chemical process, T is the temperature and R is gas constant. E is often called the activation energy of the process. From the rate k, one may determine the state of the process by integrating the rate:

This procedure may be used to calculate the state of the degradation process, and in case the state of an item is related to the state of the degradation process, also to calculate the value of a state variable for the item. For example, assume that A and E are known, and 7 ^~ can be measured using the remote access apparatus. Then one can measure the value of a measurable quantity f, and determine the temperature T(f) using calibration information and the value of the measured quantity. Using this temperature, one can calculate the rate / of a process, and by integration, the state K as function of time t. Furthermore, a limit K _t h may exist such that item can be considered usable only when K<K _t h, or in other words, the item may be determined non-satisfactory, when K≥K _t h. Moreover, in case the environment conditions are known, there may exist a factor that correlates the process state to a remaining life. E.g. one may assume certain temperature for the rest of the product life, and using this estimate, convert the remaining life Kth-K to a time value. This may be done e.g. by using the rate k=Aexp[-E/(RT ₀ )] and assuming a known T ₀ for the rest of the life of the item, and possibly assuming the temperature T ₀ to equal the measured temperature T. These procedures, however may be hard to use, may be 5 intensive computationally, and may be intensive in terms of memory requirements for several reasons. First, the determination of the parameters A and E may be difficult. Second, the calculation of the exponential function and its integral can be difficult. And third, to determine the temperature, one needs to know the calibration of the remote access device. One object of the 1 0 invention is to simplify the procedure of estimating a useful life.

As for the exponential form of the rate, the derivative of the exponential function has the same value as the function itself. Therefore, to some accuracy, the exponential function may be considered non-increasing, as

1 5 long as the value of the function is below a threshold. Moreover, the value may considered negligibly small, when it is below a threshold. Still further, one may apply a 1 ^st degree Taylor series at the point, where the rate is considered non-negligible. Thus, above a threshold, the rate may be approximated with a straight line, while below a threshold, the rate may be

20 approximated negligible, i.e. zero. Therefore, an approximation for the reaction rate may be written as

wwhheerree cc iiss aa ccoonnssttaanntt aanndd TT ₁₁ iiss tthhee uuppppeerr tthhrreesshhoolldd vvaalluuee ffoorr tthhee tteemmppeerraattuurree,, bbeellooww wwhhiicchh nnoo ddeeggrraaddaattiioonn iiss aassssuummeedd ttoo ooccccuurr.. TThhiiss ttyyppee ooff tthhrreesshhoolldd,,

2255 wwhheerree aa vvaalluuee iiss tthhee tthhrreesshhoolldd vvaalluuee ffoorr aa rraattee,, wwiillll bbee ccaalllleedd aa ssoofftt tthhrreesshhoolldd..

IInn aaddddiittiioonn ttoo tthhiiss ssoofftt tthhrreesshhoolldd,, aa hhaarrdd tthhrreesshhoolldd ffoorr tthhee ssttaattee mmaayy bbee uusseedd,, ssuucchh tthhaatt tthhee iitteemm iiss ddeetteerrmmiinneedd aass nnoonn--ssaattiissffaaccttoorryy aass ssoooonn aass tthhee ssttaattee vvaarriiaabbllee KK eexxcceeeeddss aa tthhrreesshhoolldd KK _tt hh _,,00 -- FFoorr iilllluussttrraattiioonn ppuurrppoosseess,, aa tthhrreesshhoolldd 77 ^~~ _?? iiss sshhoowwnn iinn FFiigg.. 44.. MMoorreeoovveerr,, uussiinngg tthhiiss aapppprrooxxiimmaattiioonn ffoorr tthhee rreeaaccttiioonn rraattee,,

3300 tthhee ssttaattee ooff tthhee ddeeggrraaddaattiioonn pprroocceessss bbeeccoommeess tthhee iinntteeggrraall ::

An finally, a product may be considered unsatisfactory if K exceeds a given threshold value K _t h _,0 - As seen from figure 3, the relation between the temperature and the frequency of local oscillator is almost linear. Therefore, one may equally accurately write the state of the degradation process as:

where e is a constant and f ₁ a threshold value. Moreover, a product may be considered unsatisfactory if K exceeds another given threshold value K _t h, i. It is also noted that the constant e may be negative, if c is positive and vice versa. Finally, the thresholds K _th ,o and K _t h, i are dependent on the constants c and e, respectively. Without loss of generality, one may use e.g. the reaction rate k=max(0, T-T ₀ ) and the threshold In case of a lower threshold one may use a negative constant c or e, e.g. ο=-λ and use procedures similar to what has been described to the upper threshold. Specifically, as an example, if an upper threshold for temperature is used: k=max(0, T-Ti), the same threshold may be equally used for the frequency: k=max(0,f ₂ -f), where the threshold f ₂ is a lower threshold. These thresholds are depicted in Figs 4 and 5. The lower threshold frequency f ₂ should correspond to the upper threshold temperature T and the threshold K _th for the process should be described in terms of frequency should correspond the threshold KJc for the process described in terms of temperature. Figure 6 shows an example of determining the state of an item using the rate variable. The data for the measurement is given in Fig.4 and the threshold 401 of 22.5 °C is used as the threshold temperature T ₁ to determine the temperature, where reactions start to occur. As long as the temperature stays below T the process does not evolve, and when temperature exceeds the threshold, the process evolves, as characterized by the state variable K. The item may be considered satisfactory, is this state variable stays below a predetermined level. In Figure 6, a threshold value of / _f/7 =15 is shown as an example. The state of the item K exceeds the threshold at time 6, as denoted by the line 605, and the item could be determined unsatisfactory at this time. It is noted that the curve between instances of time, e.g. between times 5 and 6 serve only as a guide to the eye, the values of the state in the figure are updated only at the time of measurements. Computationally it may be possible to update the state more frequently than the measuring frequency. Figure 7 shows the state variable, as calculated using frequency data shown in Fig. 5. In this case, a constant e=- 1 for the approximate reaction rate is used, and therefore, the value of the process state differs from that of Fig. 6. This can be taken into account in determining the threshold for accepting items. For example, in the figure 7, the threshold value K _t h that correspond to the same threshold of figure 6, is only around 7.5, while in the figure 6 it was 15. The difference is mainly due to a relatively small variation of frequency in terms of temperature. With this choice of K _t h, the first measurement exceeding the threshold is at time 6, as in case of Fig. 6, as denoted by the line 705.

Even if not theoretically as well supported, in some cases a lower threshold for the rate can also be used. E.g. if an RFID tag is located on an thermally insulating package, an item is in the insulating package, and the item should not freeze, it might be possible to set a lower threshold to 0 °C (or to other freezing point), and use an integral to approximate the temperature inside the package. Theoretically, the temperature is known to exponentially decrease towards the ambient temperature. In this case one may use a soft lower temperature threshold, and use the steps discussed above. This is illustrated in Fig. 8 where the lower limit temperature is set to 15 °C, and the process state variable is integrated as discussed. It is noted that the state starts to increase at time 1 1 , which is the first instance in Fig. 4, where the temperature goes under the lower threshold 7 ₂ =15 °C. Also other dependence of the rate on the environment variable for the process state variable approach are possible, even if not theoretically always supported. For example, in case the item should be transported in a temperature not in between 6 and 60 °C, one may use a reaction rate variable that is positive only between these temperatures. The reaction rate may be assumed constant or linearly dependent on temperature. Also other forms could be possible, such as k= ax[0,(TrT)(T-T ₂ )], where TV and 7 ₂ are the upper and lower threshold, respectively. This would maximize the rate in between the threshold values, and make the rate zero at both threshold values. Furthermore, if an item should be transported in between the temperatures 0 and 6 °C, one could use a non-zero reaction rate below 0 degrees and above 6 degrees, such as k=max[0 _! C2(T ₂ -T) _! Ci(T-T ₁ )], where 7 ^~ _? and T ₂ are the upper and lower threshold, respectively. The constants Ci and c ₂ introduce different rates for processes exceeding the upper threshold and for processes going below the lower threshold. Without loss of generality, one of the constants Ci and c ₂ may be set to one, and the other, and the threshold for the state variable may be divided accordingly. Finally also nonlinear forms for the rate may be used, such as k=(T-Ti) ^N , when T>T and k=0 otherwise. Here N is a positive exponent, and in the non-linear case different from 1 . Naturally combinations of nonlinear terms are also possible, such 7 ^" - TV)]'}. However, nonlinear terms are computationally more expensive than linear terms.

It is also possible to have two or more state variables: the first state variable increases with a first rate, when the temperature is e.g. below a lower threshold, and the second state variable increases with a second rate, when the temperature is e.g. above an upper threshold. In this case the item's condition could be determined by comparing the first state variable to a first threshold and the second state variable to a second threshold. In case of multiple state variables, at least one state variable is compared with at least one threshold value. The item could be determined non-satisfactory if one of the thresholds is exceeded or all the thresholds are exceeded. Moreover, one may use a hard threshold at the lower threshold temperature and a soft threshold at the upper threshold temperature, or vice versa. By a hard threshold hard it is meant that the item is characterized as non-satisfactory, when a hard threshold is exceeded at least once. A soft threshold is used as a threshold for reaction rate variable, and the item is characterized as non- satisfactory, if the integral of the rate, i.e. the state variable, exceeds a threshold. Without loss of generality, in case of a hard threshold, the measured value may be interpreted as a state variable of the item, and therefore, in both cases the item may be characterized as non-satisfactory, if the state variable exceeds a threshold. o summarize, some forms for the reaction rate are:

Table 1 : Forms for the reaction rate. For more complex usable ranges, these formulas may serve as guidelines, and separate state variables can be used for separate ranges. Moreover, it is noted that the Arrhenius formula k=Aexp{-E/[RT(f)]} may be applicable in some applications, but has the aforementioned drawbacks. A threshold can be applied to each state variable or a single threshold can be applied to a combination of the state variables. The table lists the formulas in terms of the measurable quantity f, such as the frequency of the local oscillator or the TRcal value, where the backscattering frequency changes rapidly. However, as discussed above, the forms given in the table may be used equally well for the environment variable T. In case the measurable quantity is used, calibration data is not needed to calculate the state variable. For this reason, in the table there are given the possibilities of an increasing and a decreasing rate for the case "f>f ₁ or f<f ₂ ", since the degradation rate may increase as function of temperature, but decrease as function of the oscillator frequency. Above, only one remote-access apparatus was considered. However, the concepts may be applied to determine the environment dependent usability of a set of items using energetically essentially passive remote-access apparatuses. Thereby naturally also the environment dependent usability of an item using becomes determined using several energetically essentially passive remote-access apparatuses, including one remote access apparatus. Suppose a set of remote-access apparatuses is being measured. From the measurement, a set of values is measured, each value f being indicative of the value of an environment variable, such as temperature. The one may deduce a statistical measure of the mean, such the average value, the geometric mean, the median value, or another percentile value from the distribution, and use this statistical measure as discussed above. E.g. the hard limits may be associated with the statistical measure, such that the all the measured items are determined unsatisfactory, when the statistical measure, such as the average value, exceeds a limit or falls below a limit. Moreover, soft limits for the statistical measure may be used and an integrator used to update the state variable of the set of items. Still another example is to deduce the temperature of the set of item by comparing the distribution of f to known distributions, and determining the temperature based on the distribution. As an example, a large number of remote-access apparatuses at temperatures 77 may have been measured, resulting in a distribution Then, when determining the usability of a set of items, a distribution φ of the quantity f may be measured. Then by comparing the measured the corresponding temperature may be found. E.g. The temperature T may be considered to equal the value 7V, where the difference between φ and in minimized. Then hard or soft thresholds for this temperature may be used.

For example, due to variations in the remote access apparatuses, a system operator may know that at 6 °C, the average frequency of local oscillators is 158.27 kHz, but the variation is such that 5 % of the remote access apparatuses have a local oscillator frequency less than 1 57.76 kHz. Thus the system operator may set a statistical limit such that all the remote access apparatuses are determined non-satisfactory if the average frequency decreases below 1 58.27 kHz, or if more than 5 % of the measurements are below 1 57.76 kHz, or if both conditions are satisfied. Alternatively one may determine the value, below which 5 % of the measurements are (i.e. the 5 percentile value), and if this value falls below the limit 1 57.76 kHz, all the items may be determined non-satisfactory. This example illustrates the use of a threshold on statistical basis. This example is shown in Fig. 9a. The curve 91 0 shows a reference distribution with the mean 1 58.27 kHz, and deviation such that 5 % of the distribution is below a threshold 91 6 which is at 157.76 kHz. The curve 91 2 illustrates the case, where the whole distribution is shifted towards smaller frequency values, which would indicate an increase in the temperature. The curve 914 illustrates a case, where the deviation is increased. Measurements according to distributions 91 2 or 914 could be considered as an indication of unusable items.

All these examples are based on the assumption that all the items are in the same temperature. In practice this is not necessarily the case. A temperature distribution may result in widening of the distribution of f. Therefore, a hard threshold may be applicable to a statistical measure of the variance, too. The measure of variance may be the biased variance, the unbiased variance, the (biased or unbiased) standard deviation, range, interquartile range, mean of absolute deviation, or the proportion of any of these to a statistical measure of the mean, as discussed above. Moreover, such a threshold for the range may be dependent on the statistical measure of a mean. Thus, a usability of a set of items may be determined to be unsatisfactory, e.g. if the deviation of the measured distribution οί is e.g. too large. As an example, in Fig. 9a, the distribution 914 has a larger deviation than the distribution 91 0, which may be an indication of an underlying temperature distribution. In addition to a percentile value, as discussed above, a threshold could be set also for the deviation.

In addition to a statistical measure of the variance, the skewness of the distribution may also be used to determine the usability. A usable range for the skewness may result in an upper and a lower threshold. A measure of the skewness is related to the third moment of the distribution, and also other measures are known in the art. Further, if the temperature of items is distributed, using the definition of a marginal distribution and the law of total probability, the measured distribution φ may be expressed in terms of the distribution in the known temperatures as:

φ =∑φ \ _τ=Τί Ρ(Τ = ΤΪ) ,

i where P(T=Ti) is the probability of an item to be in the temperature 77. These probabilities may possibly be solved with some numerical or analytical method, and in case a too large portion of items seems to be it the forbidden temperature range, the set of items may be determined unusable. A threshold may thus indicate both a value of a temperature and a percentage value. To illustrate this possibility, Fig. 9b shows an example of three reference distributions corresponding to temperatures Ti=4 °C, Ti=6 °C, and Ti=8 °C. The average frequency corresponding to these temperatures and Fig. 3 are 159.27, 158.27, and 157.26 kHz. These distributions are labeled with 922, 924, and 926, respectively. For purposes of illustration, the deviation is assumed independent of the average frequency, even if in a real case some dependence may exist. Figure 9c shows different linear combinations of these distributions. The distribution 932 shows the case, where equally many remote-access apparatuses are in the temperatures Ti=4 °C, Ti=6 °C, and Ti=8 °C, and therefore P(T=Ti)=1/3 for all these temperatures. The distribution 934 and 936 correspond to cases where 2/3 of the apparatuses is at the temperature Ti=6 °C, and thus P(T=6 °C)=2/3. However, the distribution 934 corresponds to case where one third of the apparatuses is in the temperature Ti=8 °C, and the distribution 934 corresponds to case where one third of the apparatuses is in the temperature Ti=4 °C. Vice versa, by measuring a distribution φ, the probabilities P(T=Ti) may be found in case the reference distributions are known. With statistical techniques, the usability of a set of items can be determined based on a set of measurements. As the usability of a set of items is determined, the usability of a single item belonging to the set becomes also determined. In a statistical method the state variable may be regarded as a state of the whole set of items, e.g. the average frequency of a set of local oscillators may be considered a state of the set of items. In case a percentile value is used, the state variable of the set of items may be the percentile value. E.g. the 5 % value of the distribution. Still further the state variable may be the percentage of the items that fall below a given threshold value. The usability of all the items of the set may then determined by comparing the state variable with a threshold. In a preferred embodiment, the environment dependent usability is determined using directly a value of a measurable quantity or a set of measurable quantities, which are not individually converted to a temperature value using calibration data. Therefore the method of the preferred embodiment may be essentially independent on calibration information of the remote-access apparatus that is used to determine the usability of an item.

It is possible to include one or a few calibrated remote-access apparatuses to the monitoring system. The remote-access apparatuses are used in connection with items, which may be located in an environment, e.g. in a vehicle for transportation, and one or a few remote-access apparatus in the vehicle may have been calibrated. Then, by measuring the value of the measurable quantity of the calibrated remote-access apparatus (or apparatuses) and by using the calibration data, the temperature in the vehicle can be determined. In the following, this temperature is denoted by 77. Further, from reference measurements it may be known that in this temperature, the distribution of the measurable quantity should be φΙτ=η Then, by comparing the measured distribution

the environment dependent usability of the the items can be determined. For example, one may compare the distributions φ &ηό and if they deviate too much, all the items can be considered unsatisfactory. Alternatively, one may compare the individual measured values f of the measurable quantity with a statistical value of and determine the usability of the corresponding item based on the comparison. A very often used statistical value for φ\ _τ=Τί for the comparison is <f>±Nstd(f), where <f> is the average value of f, std(f) is the standard deviation of and Λ/ is an integer, commonly either one or three. The ± means either addition of subtraction, depending whether the limit would be a lower threshold or an upper threshold. These values could be determined using φ\ _τ=Τί and would therefore correspond to the temperature, as measured using the calibrated remote-access device (or devices). Naturally, the measured values of f may also be compared to a threshold value which would correspond to the value of <f>±Nstd(f) as calculated from a distribution of f in a predetermined threshold temperature. In case a calibrated remote-access apparatus is used in a vehicle, the state variable regarding the set of items, or state variables regarding other items in the vehicle may be updated using an integrator and the determined temperature.

In a system for monitoring the state of items by using remote-access apparatuses, the memory of the remote-access apparatuses can be utilized. For example, the threshold value or threshold values can be stored on the memory of the remote-access apparatus. In particular, in case the remote access-apparatus is configured to send an AFC to a reader device, the AFD of a remote-access apparatus may comprise information on a threshold or information on several thresholds. Moreover, AFC of the remote-access apparatus may comprise information on the apparatus' capability for range measurements, e.g. capability of providing information for determining the usability of an item. For the system, it is efficient that the threshold or thresholds are specified in terms of the measurable quantity. I.e. the value of the environment variable is not used to determine the state of the item. As shown in Figs. 4 - 7, the value of the measurable quantity can be equally well used to determine the state of the item. Moreover, Figs. 4 and 5 illustrate how thresholds in terms of the measurable quantity may depend on the thresholds in terms of the environment variable.

In the following, preferred embodiments, where the usability of an item is determined using a remote-access apparatus by using at least one threshold, will be described. In an embodiment the usability of an item is determined by comparing a state variable of the item to a threshold. As for the term "state", it is emphasized, that in case of a hard threshold the state variable of an item is directly related to the value of the measurable quantity, while in case of a soft threshold, the state of the item is related to a time integral of a rate variable. Furthermore, the state may refer to a statistical measure of a set of measured values.

The method for determining the environment dependent usability of an item may comprise determining information indicative of a usable environment for said item, determining the type of threshold, determining at least one threshold value, measuring the value of a measurable quantity, determining the value of a state variable using said measured value, forming usability information by comparing a state variable with said at least one threshold value, and determining the usability of an item by usability information.

In determining information indicative of a usable environment for said item, a system operator determines the usable environment for an item. The system operator here means the person operating or setting up a system for monitoring the states of items. Each type of item may require the determination of the usable environment for the specific item type. For example fresh food needs to be transported and stored in a cool environment. Processed food meant to be served needs to be transported in a hot environment. Some medicines may need to be transported and stored near the room temperature. Due to the large number of possibilities, there clearly needs to be a system operator who sets up a system for monitoring the state of items. The usable environment may be e.g. an allowed temperature range or a forbidden temperature range. In some cases only one threshold is needed (i.e. the usable range extends to either positive or negative infinity). In some cases more than one ranges may be applicable. The range may be determined in terms of an environment variable, but alternatively also in terms of a measurable quantity.

In determining the type of a threshold, the system operator determines, using the said information, whether the process that makes the item non- usable is a slow one or a rapid one and thus determines, whether soft or hard thresholds, or both, are applicable for the problem. Moreover the type in terms of whether a threshold is an upper threshold or a lower threshold should be determined. Moreover, the system operator should determine; is one temperature range enough, or should several ranges be used. Also it must be decided; are the ranges characterized with a single state variable or should there be many. In a preferred embodiment the usable range is relatively simple, and can be characterized with one or two frequency threshold values. In case many state variables are to be used for determining the usability of an item, at least one of them is compared with a threshold value. Moreover, the type of threshold may be statistical or correspond to a single measurement. In case the threshold is a statistical threshold, more than one value associated with the threshold may be needed, such as a percentile value and the corresponding percentage. In determining at least one threshold value, the actual value or values for the threshold or thresholds are determined. The threshold values for the measurable quantity may be determined using the calibration information of the remote access apparatus, if calibration information is available. In case calibration information is not available, the remote access apparatus may be set to an environment, where the environment variable has the threshold value. In this environment, the measurable quantity may be measured, which reveals the threshold value in terms of the measurable quantity. Threshold values for the statistical measures of a distribution may be determined from a database of known values of measurable quantity in known temperatures. At this point it must also be known, whether the measurable quantity increases or decreases with increasing environment variable. As shown in Figs. 3 - 5, the upper threshold of a temperature may become a lower threshold for the frequency. The frequency threshold thus revealed may be stored e.g. to the remote access apparatus' memory. It may also be stored in the RFID reader device, in a detachable memory card used in connection with the reader device, in an external server, or the data may be stored partly in some or all of the previous, including the remote access apparatuses. In a preferred embodiment, the frequency threshold value and the threshold type are stored in the remote access apparatus' memory. In case a soft threshold is used, the frequency threshold may be determined as discussed above. In this case, exceeding the frequency threshold would imply updating a state variable. In addition to the frequency threshold, a threshold for the state variable or state variables should be determined. In case different rates are used for different processes that are characterized with a common state variable, also the relative rate factor Ci (Table 1 , line 7) should be determined. The threshold for the state variable and the relative rates can be determined by experiments, where the evolution on the process is measured together with the evolution of the state variable. Therefore, for a soft threshold, two threshold values may be needed.

The threshold value can be determined for at least the following cases:

(1 ) a threshold value or threshold values for a single item based on a single measured value of a measurable quantity using a remote access apparatus, (2) a threshold value or threshold values for a single item based on a multiple of measured values of a measurable quantity using a set of uncalibrated remote access apparatus, wherein a reference distribution is unknown,

(3) a threshold value or threshold values for a single item based on a multiple of measured values of a measurable quantity using a set of uncalibrated remote access apparatus, wherein reference distributions are known,

(4) a threshold value or threshold values for a single item based on a multiple of measured values of a measurable quantity using a set of uncalibrated remote access apparatus, and at least one value of a measurable quantity using at least one calibrated remote access apparatus,

(5) a threshold value or threshold values for a set of items based on a multiple of measured values of a measurable quantity using a set of uncalibrated remote access apparatus, wherein reference distributions are known,

(6) a threshold value or threshold values for a set of items based on a multiple of measured values of a measurable quantity using a set of uncalibrated remote access apparatus, and at least one value value of a measurable quantity using at least one calibrated remote access apparatus;

In the first case, the usability of an item is determined by comparing a measured value with a threshold value. The threshold value may be e.g. a frequency of a local oscillator corresponding to a temperature. The local oscillator is a local oscillator of the remote-access apparatus in connection with the item. The threshold may be set e.g. using calibration data of the device, using calibration data of other similar devices, or by setting the item to a known temperature and determining the value of the measurable quantity. Further, the threshold may be set on statistical basis. E.g. if a distribution \ _τ=Τί of the measurable quantity in a temperature Ti is known, a threshold may be set for example to is the average of the reference distribution, Λ/ is an integer, and is the deviation of the reference distribution. This is illustrated in Fig. 9b. For example, the item has a usable temperature range of 4 - 8 °C, a lower threshold for the frequency may be found from the known distribution corresponding to 8 °C, and the lower threshold may equal to e.g. Similarly, the upper threshold for the frequency may be taken as In Fig. 9b the distribution 926 corresponds to 8°C, and the lower threshold 919 for the frequency is shown. Similarly, the distribution 922 corresponds to 4°C, and the upper threshold 918 for the frequency is shown. It is noted, that the reference distributions may have been measured using such remote-access apparatuses that are not present in the measurements when determining the usability of an item.

In the second case, a threshold may be set based on the measured distribution. The usability of an item is determined by comparing a measured value with a threshold value. The threshold value may be e.g. a frequency of a local oscillator corresponding to a temperature. The local oscillator is a local oscillator of the remote-access apparatus in connection with the item. The threshold may be set for example to <f>±Nstd(f), where <f> is the average of measured values, Λ/ is an integer, and std(f) is the deviation of the measured values. This can be illustrated e.g. with Fig. 9a. Suppose the distribution 910 is measured, all items that have a frequency value below the line 916 may be considered unsatisfactory. Even if not shown, it is evident that an upper threshold may be used. As the actual threshold may depend on an average value and a deviation value, the threshold value may be e.g. the constant N. It is also possible to set a threshold to <†>-†TH, and use f-m as the threshold.

In the third case, a threshold may be set based on known distributions. The usability of an item is determined by comparing the usability of an item is determined by comparing a measured value with a threshold value. The threshold value may be e.g. a frequency of a local oscillator corresponding to a temperature. The local oscillator is a local oscillator of the remote-access apparatus in connection with the item. By measuring the value, a their distribution ^ becomes measured. By comparing the measured 0 to the set of known distributions, the corresponding temperature and the corresponding reference distribution may be found. The threshold may be set for example to is the average of the reference distribution, N is an integer, and is the deviation of the reference distribution. This is illustrated in Fig. 9b. The distribution 928 is an example of a measured distribution. By comparing this to the known distributions 922, 924, and 926, the measured distribution can be determined to be almost the distribution 924. Therefore, the temperature can be considered to be the temperature corresponding to the distribution 924 (which was Ti=6 °C, as discussed earlier). Then, based on the distribution 924, one may set a threshold value, such as the one denoted by 929.

In the fourth case, at least one calibrated remote-access apparatus is used to measure the temperature. Therefore, one may select the correct reference distribution \ _τ=Τί since the temperature Ti is measured. In contrast to case (3), no comparison between distributions is needed. The threshold value may be set as as discussed in case (3). It is evident that instead of a calibrated remote-access apparatus, other temperature sensor can be used. Further, by using a calibrated remote-access apparatus, or another sensor, it may be feasible to make measurements and record data to obtain the reference distributions for cases (3), (5), and (6).

In the fifth case, a set of items may be determined unusable based on the measured distribution. The distribution may be compared to reference distributions, and if too much deviation or skewness is noticed, all the items are determined unusable or "possibly unusable". In the latter case, the possibly unusable items may be sent for a more detailed examination. Measured distribution may be compared to reference distribution to find a temperature value, as in case (3). In addition to deviation or skewness, percentiles can also be used, as discussed above. For the threshold values of the percentiles, the case (1 ) is referred to. Finally, it may be possible to calculate probabilities for an item to be in an unallowed temperature, and set a threshold for the probability.

In the sixth case, a set of items may be determined unusable based on the measured distribution. Furthermore, the temperature may be determined using one calibrated remote-access apparatus within the set of items. The measured distribution may be compared a reference distribution, and the reference distribution can be selected based on the measured temperature, as in case (4). If too much deviation or skewness is noticed, all the items are determined unusable or "possibly unusable". In the latter case, the possibly unusable items may be sent for a more detailed examination.

Also other methods for determining a threshold value or threshold values may be used.

Some of the above methods need information on the distributions as function of the environment variable, e.g. temperature. This information may be obtained by measuring the behavior of remote-access apparatuses in different environments. Preferably these measurements are done when the temperature value is accurately known. These measurements may be done in a normal life-cycle of remote-access apparatuses. Thus, by using apparatuses it may be possible to initially use a threshold scheme corresponding to case (2), but as more information becomes gathered, the case (3) may become more applicable. Also, as discussed in case (4) it may be possible to use a calibrated remote-access apparatus to record this information.

In measuring the value of a measurable quantity, the value of the measurable quantity is measured with a reader device. In practice, the reader device may measure several directly measurable quantities, and based on these measurements the reader device may measure a value of the measurable quantity, as discussed above. The reader device may also receive the threshold type and threshold value from the remote access apparatus. In a preferred embodiment, the measurable quantity is measured periodically, e.g. once in a specified time interval. In addition, the reader may receive the time for the previous measurement e.g. from the remote access apparatus, and the value of the time interval e.g. from a database. By comparing the time of the current measurement, the time of the previous measurement, and the specific time interval, the reader device may also determine, whether all the needed measurements are done, or not. If the reader device notices that not all measurement have been made, it may give some penalty to the measurements. E.g. in case a hard limit is used, and the reader device notices absence of some measurements, the device may change the frequency value such that is in the unacceptable range. Comparison of this value to a threshold will be done in a subsequent step, and the comparison will result in classifying the item as non-satisfactory. In case of a soft threshold, a penalty scheme may be applied e.g. by modifying a time integrator or by modifying the measured value of the measured quantity. As the integrator will be discussed later in the context of "updating a state variable", implementation of a penalty scheme in the integrator will be discussed in that context.

In case a soft threshold is used, the penalty scheme me be applied by changing the value of the measured quantity. The reader device may change e.g. the frequency value in such a way that some penalty will be given to state variable. E.g. if a soft lower threshold of 158.27 kHz (corresponding to an upper threshold of 6 °C in Fig. 3) is used and a frequency of 158.77 kHz (corresponding to 5 °C in Fig. 3) is measured the item is in the usable range. However, if the reader devices notices the absence of e.g. two measurements, the reader device might change the measured frequency value to give some penalty to the state of an item for an operator not performing the measurements. E.g. the measuring device might assume, that the remote-access apparatus has been in 10 °C (corresponding to 156.25 kHz) during the absent two measurements, and therefore use lower value for frequency. The corrected value could be assigned e.g. such that during the three steps, one of which is measured, and two of which are absent, the degradation process should have advanced 2x(158.27-156.25)+0=4.04. The same advance in the state variable would be achieved with a frequency of 158.27-4.04/3=156.92 kHz. (It is noted that this would correspond to temperature 8.69 °C, even though the temperature is not used in the procedure). Therefore, the reader device may assign this frequency value to for the frequency of the remote access apparatus' oscillator. When this frequency value is used in updating the state variable, proper penalty is given to the state. The value of the calculated frequency may depend also on the integration algorithm used to update the state variable, and a person skilled in the art may take this into account.

The absent measurements may pose a problem for the system to determine the operator responsible for a faulty item. For example, suppose that a first operator delivers an item to a second operator. The first operator does not comply with a measurement schedule, but the second one does. Thus, when the second operator makes his first measurement, he recognizes that a measurement schedule has not been followed, and may therefore apply a penalty for the item. This may result the item to be determined as non- usable. However, the operator responsible for the fault is not the second operator, but more likely the first operator, who delivered the item to the second operator. However, the second operator is the one making the measurement. Thus the operator making the measurement is not necessarily responsible for a faulty item. Naturally it is also possible that a hard threshold is exceeded in the first measurement by the second operator. Then, if the first operator has followed a measurement schedule, the second operator is responsible for the fault. Therefore, information on whether the item is determined as non-satisfactory on the basis of a penalty scheme or on the basis of a measurement is also useful. The usability information of an item may comprise such information. Moreover, the second operator may also perform a first measurement, but omit some measurements. In such a case, the determination for non-usefulness of the item may be done on the basis of a penalty scheme, but still the second operator is the one responsible for the fault. Therefore, information on whether the measurer or a previous operator is responsible for the fault is also useful. In order to determine the operator responsible for a fault, information indicative of the operator making the measurement may be stored together with the information on the updated state variable. Also the time of the measurement may be stored. Depending on the application, a flag indicating the application of a penalty scheme may be used to indicate that a penalty scheme has been applied at some point during the item's life or to indicate that a penalty was applied at the point the item was determined non-satisfactory. In addition, two flags might be used, the first the flag indicating if the penalty was applies at some point during the item's life, and the second flag indicating that the penalty was applied at the point the item was determined non-satisfactory. However, in case the first measurement by the second operator indicates an forbidden condition, e.g. too high a temperature, and the first operator has not performed measurements, the operator responsible for the fault be hard to solve with these technical apparatuses. In determining the value of a state variable using said measured value, the system analyses the type of the threshold. If the threshold is a hard threshold, the value of the state variable is the value of the measurable quantity. If the threshold is a soft threshold, a state variable is updated using the measured value of the measurable quantity, and information on the threshold.

In forming usability information by comparing a state variable with said at least one threshold value, the reader device or another device may compare the state variable such as the measured frequency or the integrated process state to a threshold value. In case a soft threshold is used, the state variable may be updated before the comparison. As discussed above, a calculated frequency value can be used as the state variable, or the calculated frequency value can be used to update the state variable. The calculated frequency may give penalty for the item's state in case measurements have not been done according to a measurement schedule. The comparison may be done by the reader device. Based on this comparison the reader device may generate usability information. The usability information is preferably written to the remote-access apparatus. The usability information can be stored also elsewhere, such as the reader device, a detachable memory card or an information server. Usability information may be one bit indicating that the item is either usable or not. Usability information may also comprise another bit of information indicating whether the item was determined non- satisfactory on the basis of a penalty scheme or on the basis of a measurement. Usability information may also comprise a third bit of information indicating whether a penalty scheme was applied to the item's state during the item's life or not. Moreover, usability information may comprise a fourth bit of information indicating whether the operator responsible for a fault is the one making the measurement or a previous operator. In determining the usability of an item by using usability information, a device will determine the usability of an item based on usability information. The device may be an RFID reader device or another device, such as a computer. The usability information may be formed in the device or elsewhere. As an example, the state variable may be sent from a reader device the device to another device, and the another device may generate usability information by comparing the state variable with a threshold value. Here the state variable may mean the value of the measurable quantity or the state variable obtained by integration of a rate. In addition, an RFID reader may read the usability information from a remote access apparatus and send this information to the another device. The another device may then determine the usability based on this information. Usability information may also be stored on a reader device. Moreover, the threshold values or usability information be stored in one or some of the remote-access apparatus, another remote-access apparatus, the reader device, the another device, a memory card, or an external database.

The method may further comprise updating a state variable and storing a time. These steps are preferred in case of soft thresholds, but the latter may be used with hard thresholds, too. In updating a state variable, a state variable is updated using the value of a measurable quantity, the previous value of said state variable, and preferably also two values of time. This step is used with soft thresholds. Denoting the state variable at time t, with K the state variable is updated with a time integrator, such as the forward Euler integrator as Kj ₊₁ =Kj+(tj+i-tj)k. Here t,-+i is the time at which the measurement is made and t, is the time at which the previous measurement was made, k is the rate, which may depend on temperature or frequency, the threshold value(s) and some other constant, as discussed above and in Table 1 . A person skilled in the art may use also other time integrators. In a preferred embodiment, measurements are made in specified time intervals, and in that case the difference Ι,+ι-Ι, is constant. Thus, if one could rely on such measurement frequency, this term could be assumed constant, and the values of time need not to be known. However, as discussed above, the time may be needed to monitor the measurement frequency, and possible to use some penalty scheme, if all the required measurements are not done. The value of the state variable K, as well as the latest time of measurement, t may be stored on the remote-access apparatus' memory. Moreover, if a simple time integrator is used, such as the one described above, only one pervious value of the state variable needs to be stored. Once the old value is read and the new value obtained by updating, the new value may be stored to the remote access apparatus, and the new value may replace the old value. In case some measurements have not been according to a measurement schedule, a penalty scheme may be applied at the point of updating a state variable. Suppose, that normally the forward Euler integrator is used, but as in the above example, two measurements are missing. Then, instead of updating the state variable using the Euler formula: K _i+1 =Ki+(ti ₊ i-ti)k, the state variable may be updated e.g. as Here the subscript i-2 is used for the state variable, since two latest measurements are missing. Moreover, instead of the real difference between measurement times, f, ₊ rf/-2, the scheduled time difference At=t _i+1 -ti is used instead. And finally, for the missing measurements the penalty rate k _P is used. In the above example, the penalty rate could be taken to be KHz, which are the frequencies corresponding to a lower threshold of 1 58.27 kHz and the frequency 156.25 kHz corresponding to an assumed penalizing environment.

In storing a time, the time of measuring a value is stored. The value is stored preferably such that it is available for the reader device, when making subsequent measurement. The time may be used to check if all required measurements have been made, and it may be used to update the state variable. Preferably the time is store in the remote-access apparatus.

The method may also comprise storing information indicative of an operator, information indicative of a reader device, and information indicative of the time of fault.

The method may also comprise determining at least one of an estimate of the remaining life, and possibly based on this remaining life, determining a best before date for an item or a price of an item. Such information may also be displayed on a display in corresponding system.

A remote-access apparatus in a preferred embodiment is energetically essentially passive, is capable of sending information, such as information indicative of the value of a measurable quantity and information indicative of a threshold. The apparatus also comprises memory, which may comprise information indicative of a threshold. This information may comprise a threshold value and information on the threshold type. Furthermore, the remote access apparatus may comprise means for receiving data and storing the data, and the memory may comprise information indicative of the usability of the apparatus. An energetically essentially passive remote access apparatus comprises a voltage supply VREG1 (Fig. 1 ), which is arranged to extract operating power from an incoming radio frequency signal for the control unit and the memory.

An remote access apparatus is capable of sending information indicative of the value of a measurable quantity, if it is capable of sending at least one signal from which the value of the measurable quantity can be determined without calibration. Generally a remote access apparatus may send signals on different backscattering frequencies, and from these frequencies the frequency of the local oscillator 52 of the remote-access apparatus can be determined without calibration. In addition, from these frequencies, the TRcal value at which the backscattering frequency abruptly changes can be determined without calibration.

Typically a remote access apparatus comprises memory and means for receiving and storing data to said memory, as described in Fig. 1 (RXTX1 , CNT1 , MEM1 ). Such data may be sent by the reader device, and received by the remote access apparatus. Moreover, the apparatus may store the data to its memory. In the preferred embodiment the memory comprises a threshold value and information on the threshold type. In the preferred embodiment, the threshold value is a threshold for the measurable quantity, e.g. the frequency of the local oscillator or the TRcal value at which the backscattering frequency abruptly changes. Information on threshold type comprises information, whether the threshold is soft or hard, and whether the threshold is an upper threshold or a lower threshold.

In case of a soft threshold, the determination of usability needs two threshold values: a first threshold value for the measurable quantity and a second threshold value for the state variable. In an embodiment, the memory of the remote-access apparatus comprises information on a state variable. The memory may further comprise a threshold value for the state variable. Moreover, the memory may comprise usability information on the usability of the item. The usability information may comprise the information on whether the item is usable or not, but may also comprise information on whether a penalty was applied some time during the item's life, information on whether a penalty was applied at the point the item was determined non-satisfactory, and information indicative of the operator responsible for the fault. Still further, the memory may comprise another threshold value and information on the type of the another threshold. Even further, the memory may comprise information indicative of the dependence of a rate variable on a measurable quantity, such as a formula of the Table 1 . Moreover, the memory may comprise the value of another state variable, and a threshold value for the said another state variable. In addition, the memory may comprise the time of the last measurement and possibly also information indicative of the operator that performed the last measurement. And finally, the memory may comprise information relating the state variable to a time variable, such as a value of a factor to convert the state variable to a time variable. All these data may be stored as AFD in a remote access-apparatuses memory, and the password memory area of a remote-access apparatus may also be used as a data storage. In an embodiment, the remote access apparatus has also means for showing information indicative of the state of the item. Referring to Fig. 10, This is may be a display unit 1 001 . The display unit may be used to show the state of the item. For example, if a hard threshold is used to determine the usability of the item, the display may show e.g. either GOOD, or FAILED indicating the usability of the item. This is depicted in Figs. 1 1 a and 1 1 b. In case soft limits are used, the system may compare the state variable to the threshold of the state variable, and indicate the state with a user friendly form. E.g. if K _th is the threshold value for the state variable K, the proportional value of the state variable K/K _t h may be used to indicate the condition of the item. E.g. if K/Kt _h < 25 %, the display might show "GOOD", if 25 % < K/K _th < 1 00 %, the display might show "OK", and if K/K _t h≥ λ 00 %, the display might show "FAILED" (Figs. 1 1 a - 1 1 c). Furthermore, the ratio can be showed on the display as such (Fig. 1 1 d), or the remaining life, in terms of the state variable (e.g. λ -Κ/Kth) can be shown (Fig. 1 1 e). Furthermore, the remote access apparatus may comprise a factor to convert the state variable to a time variable. Therefore, the system can estimate a remaining life of the item, provided that storage conditions are known. Usually these conditions are not known, but the system may assume that the conditions are stationary, i.e. the present storage conditions remain for all useful life. The system may e.g. calculate the rate by assuming the present environment conditions, and using this rate, convert λ -Κ/Kt to a time value. Thus, an estimate for the remaining life of the item may be calculated. The remaining life can be shown as a "best for" time e.g. in days, as shown in Fig. 1 1 f. Moreover, since the reader device may know the present date, the best before date may be calculated. The calculated best before date of the item may be shown on the display, as depicted in Fig. 1 1 g. Such a best before date would be a real-time estimate for the best before date and would change dynamically, if the storage conditions change. This would allow for extended best before dates, provided that the storage conditions are reasonably good. Moreover, a dynamical best before date would allow dynamical pricing of items. Items that are near the end of their lives may be discounted (Fig. 1 1 h), while items that have a long remaining life estimate might be regularly priced (Fig. 1 1 i). This would allow for efficient pricing and circulation of foodstuff in a store. The display may be energetically passive, meaning that the display needs energy only to change the displayed information. An example of such a display is the zenithal bistable liquid crystal display (ZBD). Other zero-power displays include the bistable nematic (BiNem®) paper-like liquid crystal display technology and displays using polymer stabilized cholesteric liquid crystals (ChLCD).

In other embodiments, such a display may be located in some other device, such as the reader device. A price label printer may be integrated to a reader device to print price labels for discounted items.

In a preferred embodiment, the reader device comprises means for measuring the value of a measurable quantity using a remote-access apparatus, means for receiving information from the remote access apparatus, and means for comparing the value of a state variable with a threshold value. Furthermore, the reader device of the preferred embodiment is configured to measure the value of the measurable quantity and to form usability information by comparing the value of a state variable with said threshold value. The reader device has means to measure the value of the measurable quantity, if the reader device has means to change the TRcal value used in communication between the reader device and the tag, and means to detect the backscattering frequency. In case the TRcal value is directly readable from a remote-access apparatuses register, the reader device may be capable of reading it.

By definition, a reader device has means to receive information from the remote access apparatus; e.g. the module RX1 of Figs. 1 and 9 is a signal receiver for this purpose. Means for comparing the value of a state variable with a threshold value may be a computer code for the purpose.

The reader device may be configured to form usability information by comparing the value of a state variable with said threshold value. The state variable may be the value of the measurable quantity. Moreover, the reader device may also comprise means to update the value of a state variable. This may be a time integrator software that updates the value of the state variable as discussed above. Then the reader device may be configured to form usability information by comparing the updated value of a state variable with said threshold value.

Moreover, the reader device may comprise means for sending information to the remote access apparatus. Many reader devices have these means, e.g. the RFID tag of Figs. 1 and 10 has the signal transmitter TX1 for this purpose. If the item is determined non-satisfactory, the reader device may send the corresponding usability information to the remote-access apparatus. Also other information such as information indicative of the time, place, operator, or reader id of the instant when a threshold is exceeded, may be stored. This information may be stored e.g. in the remote-access apparatus, in another remote-access apparatus, in the RFID reader device, in a detachable memory card used in connection with the reader device, in an external server, or the data may be stored partly in some or all of the previous. In case the information is stored on a device different from the remote-access apparatus, the id of the remote-access apparatus will also be stored. Such information helps to indentify the operator who is responsible for the fault. In case of absent measurements, the reader device should be capable of determining whether the fault occurs due to penalty reasons or due to wrong environment. For these reasons, the reader device may send information indicative of the operator, e.g. the reader device id, to the remote access apparatus. The reader device may also comprise a clock for obtaining time information needed for updating the state variable. Moreover, the time may be sent to the tag.

It is also possible, that the reader device is not equipped with computing capacity. In these cases another device, such as a computer, may be used in connection with the reader device. The computer may, for example receive the value of the measurable quantity, and other information, such as the threshold values and the old value of the state variable, and the computer may then update the state variable and perform the comparisons. The computer may use the reader device to sent the data to the remote access apparatus.

The process, as discussed above, is hard to tamper, particularly when the value of the measurable quantity is used to characterize the state of the item. Because the calibration is not known to any party, it may be hard to tamper the measurements in a reliable way. In contrast, if the temperature was used to characterize the state of the item, an operator could more easily tamper the results for better life time estimate.

The present application has been filed together with the applications titled "Data storage on a remote-access apparatus", "A method for measuring environment using a calibration database", "Temperature monitoring system", "Temperature managed chain", all on the same date, and by the same applicant. For the purposes of storing and using application data in a password, the application "Data storage on a remote-access apparatus" is referred to. For the purposes of forming and using calibration data related to temperature determination, the application "A method for measuring environment using a calibration database" is referred to. For the purposes of determining temperature information in various systems, the application "Temperature monitoring system" is referred to. For the purposes of monitoring temperature of in a temperature-managed chain, the application "Temperature managed chain" is referred to. The various embodiments of the invention can be implemented with the help of computer program code that resides in a memory and causes the relevant apparatuses to carry out the invention. For example, a terminal device may comprise circuitry and electronics for handling, receiving and transmitting data, computer program code in a memory, and a processor that, when running the computer program code, causes the terminal device to carry out the features of an embodiment. Yet further, a network device may comprise circuitry and electronics for handling, receiving and transmitting data, computer program code in a memory, and a processor that, when running the computer program code, causes the network device to carry out the features of an embodiment.

It is obvious that the present invention is not limited solely to the above- presented embodiments, but it can be modified within the scope of the appended claims.