EP4205005A1

EP4205005A1 - Method for implementing and using cryptographic material in at least one system component of an information technology system

Info

Publication number: EP4205005A1
Application number: EP22761462.5A
Authority: EP
Inventors: Viktor Friesen; Viktor Pavlovic
Original assignee: Mercedes Benz Group AG
Current assignee: Mercedes Benz Group AG
Priority date: 2021-08-31
Filing date: 2022-08-02
Publication date: 2023-07-05
Also published as: EP4297332A3; EP4297334A2; US20240356746A1; EP4297332A2; KR20240038749A; JP2024536707A; CN117882072A; EP4297335A3; EP4297334A3; EP4297333A3; EP4297335A2; WO2023030811A1; DE102021004427B4; DE102021004427A1; EP4297333A2

Abstract

The invention relates to a method for implementing and using cryptographic material (KM) in at least one system component (SK) of an information technology system (IT-S) for performing at least one operation, wherein a state of the system component (SK) as described by at least one variable (VAR) is checked at least at a first time, the cryptographic material (KM) is supplemented with additional data, wherein the additional data describe possible states of system components (SK), and the cryptographic material (KM) is used by the system component (SK) when the additional data of the cryptographic material (KM) comprise at least the state that the system component (SK) has at the first time. The method according to the invention is characterized in that the additional data are formed by at least one condition (BED, BED_KM*, BED_KM ^Prov, BED_KM ^Type(skid), BED_KM ^{Type(skid), Prov}), at least one role (ROLE_KM*) and/or at least one target component identity (ZKIDENT_KM).

Description

Mercedes-Benz Group AG

Method for implementing and using cryptographic material in at least one system component of an information technology system

The invention relates to a method for implementing and using cryptographic material in at least one system component of an information technology system according to the type defined in more detail in the preamble of claim 1.

The procedure for implementing and using cryptographic material essentially deals with the fact that in different life cycle phases of system components of an information technology system that are equipped with cryptographic material, cryptographic material corresponding to the current security level of the respective system component or a sub-module of the system component must be used. This relates in particular, but not exclusively, to the area of so-called CarlT security for vehicles.

Modern vehicles, but also other systems, are characterized by increasing networking. The vehicles are not only connected to generally accessible systems such as the Internet, but also to dedicated systems operated by the vehicle manufacturer or OEM, for example manufacturer-specific applications and a manufacturer-specific server, which is often also referred to as a backend server. These are developed, marketed and operated by the manufacturer exclusively for their own vehicle fleet. All this together is also known as the vehicle ecosystem.

In practice, the diverse communication relationships between the individual system components within such a vehicle ecosystem result in a large number of new interfaces and applications, all of which are must be protected by suitable cryptographic methods, such as mechanisms, protocols, etc. These cryptographic methods, which are known in principle from the prior art, require a wide range of cryptographic material such as asymmetric key pairs, symmetric keys, certificates, random numbers, hash values, etc. All of these cryptographic materials are coordinated with one another and are often exchanged before they are put into operation of crypto material, also during ongoing operation, in the vehicle ecosystem in all participating system components and must be used by them. The provision of the cryptographic material, the so-called provisioning, takes place, for example, as part of the manufacturing process, but it can also take place during the operation of the system components. The system components include in particular control units installed in the vehicles, but also other components such as applications installed in the vehicle, OEM applications available on a smartphone, vehicle-external devices and the like.

In order to reliably prevent protected communication from being accessed, it is absolutely necessary to separate cryptographic material used during the development phase and cryptographic material used during the productive phase, i.e. the real use of the fully developed product comes, is differentiated. The cryptographic material for the development phase is also referred to below as test crypto material. In the development phase, as long as the crypto material is secret, only such test crypto material may be used. The productive crypto material, and here in particular secret productive crypto material, must be protected from unauthorized access during the entire life cycle of the system component. Ideally, this protection should be guaranteed by adhering to a special secure process and using special protection mechanisms in the respective development departments and production facilities. In the development phase, however, this is usually only insufficiently guaranteed, since various components of the information technology system are usually not or not yet implemented or implemented in this development phase. For example, the secure generation of the required cryptographic material may not yet have been fully implemented or tested. The secure transmission of the cryptographic material from the generation server, the so-called crypto material server, to the system manufacturer, for example a supplier, may not yet be guaranteed. The secure introduction of the crypto material into the system component has not yet been implemented or has not yet been finally implemented. The secure storage of the cryptographic material in this system component may not yet have been implemented because, for example, the target system has not yet installed a hardware security module (Hardware Security Module; HSM), although this is to be installed at a later date. The secure use of the cryptographic material in the system may not yet have been implemented, for example because the development interfaces required for testing and debugging are available, which have read and write access to the entire system and thus in particular also access to the installed cryptographic material allowed during the development phase. However, these are necessarily available during development and are usually openly accessible to all people and/or companies involved in the development process.

This means that it cannot be guaranteed that the crypto material used during the development phase will not be manipulated or read. However, the structure and form of this test crypto material is identical to the productive crypto material used later. It is therefore also suitable for use later in operation, i.e. in the productive phase of the respective system component. It is therefore particularly important that the test crypto material, which is relatively poorly secured and therefore highly likely to be corrupted, is not misused in the production system. It is even more serious when productive crypto material ends up on a system component in the development phase. This material, which has a very high protection requirement, can then also be manipulated or read out in whole or in part. It is no longer secret and cannot be used. If the error goes unnoticed or is deliberately covered up, the system component that later switches to the productive phase loses its protection. In practice, the safety regulations and guidelines are implemented by the people working with the system components in the phases. Therefore, errors, oversights and deliberate manipulations cannot be ruled out. Therefore, there is always a risk that test crypto material will remain in a system component that has already switched to the productive phase or that the productive crypto material will be used on a system component that is still in the development phase. Any crypto material used intentionally or accidentally in the development phase is very likely to be corrupted due to the many interfaces etc. that are still open. These dangers ultimately lead to an increased security risk. This represents a serious disadvantage of the existing procedure.

A method for the secure use of cryptographic material is already known from DE 10 2020 003 072 B3. Several system components of a networked system are equipped with cryptographic material. The cryptographic material has a mark identifying the cryptographic material as development or production material. The individual system components have a binary status flag which indicates whether the corresponding system component is in the development phase or in the productive phase in its life cycle phase. The marking of the cryptographic material is then compared with the binary status flag of the respective system component. If the marking and the flag match, ie if the cryptographic material is marked as development material and the system component is in the development phase, then the corresponding cryptographic material is used by the system component. If, on the other hand, the marking and the binary status flag do not match, safety measures are initiated. The verification of the correspondence between the marking of the cryptographic material and the binary state flag takes place prior to the respective use of the cryptographic material or once for all cryptographic material in the system component. The match can be checked, for example, when the system component is started up or when the cryptographic material is made available. As a security measure, a warning message can be issued, the delivery of the cryptographic material can be canceled or prevented and/or after Cryptographic material that has already been introduced into a system component can be deleted as required.

The disadvantage here, however, is that to distinguish whether cryptographic material may be used on the system component, a distinction is only made between two life cycle phases of the corresponding system component and two different properties (test or productive) for the cryptographic material. In real use, a differentiation into just two groups will not suffice, since the conditions for the secure use of cryptographic material are too different for different cryptographic materials and system components. For example, if a secret is created and stored directly in a hardware security module and never leaves it, the cryptographic material can be used securely during the development phase if necessary. However, the prerequisite here is that the hardware security module in the development phase does not have or use any special diagnostic interfaces which would enable the secret to be read out. A secret included in software, on the other hand, is part of a program code and can thus be read out or reconstructed comparatively easily via various interfaces. A secure use of such a secret in a later productive phase of the system component is then not possible.

The present invention is based on the object of specifying an improved method for the implementation and use of cryptographic material in at least one system component of an information technology system, which allows a particularly flexible, secure implementation and use of cryptographic materials that differ in a variety of ways in various system components. and thereby ensures that, in the case of a system component which itself or in which at least one sub-module is operated in an insecure mode, cryptographic material suitable for insecure operation is used by the system component or at least the sub-module and in the case of a system component which itself or in which at least one sub-module is operated in a secure mode, cryptographic material suitable for secure operation is used by the system component or at least the sub-module. According to the invention, this object is achieved by a method for implementing and using cryptographic material in at least one system component of an information technology system having the features of claim 1. Advantageous configurations and developments result from the dependent claims.

In a method for implementing and using cryptographic material in at least one system component of an information technology system of the type mentioned at the outset, the cryptographic material has additional data which, according to the invention, are formed by at least one condition, at least one role and/or at least one target component identity.

The method according to the invention allows a particularly flexible, secure implementation and use of cryptographic material in the most varied of system components and/or their sub-modules. Various system components such as a control device, a cloud server, an application running on a mobile device, a vehicle or an external device require the use of a wide variety of cryptographic materials such as symmetric or asymmetric keys, certificates, random numbers, hash values or the like at different times. Thus, a specific target system component, which is to use cryptographic material, can also require the use of different cryptographic materials in different states. The state of the respective system component is described using at least one variable. The corresponding system component can assume a large number of different states, i.e. it can also have more than two states, such as being in a development phase or a productive phase of a life cycle. Individual sub-modules of a system component can also be in different life cycle phases or states. For example, a network interface can be in a first, insecure state, and a memory element of the system component can be in a second, secure state. A corresponding information technology system can be networked or isolated. Cryptographic material is typically generated by a dedicated system, for example a crypto material server, and transmitted to a target system, i.e. a target system component, as part of a special data packet. According to the invention, the cryptographic material is supplemented with additional data. A corresponding condition defines the circumstances under which the use of the cryptographic material by the system component is permitted or not permitted. The corresponding conditions can be of the most diverse nature. For example, the condition can require that the respective system component is in a certain phase of a product life cycle, or that the system component has a hardware security module and that this is in a secure state so that the hardware security module can securely record cryptographic material and can put down. The cryptographic material can also include several conditions. This allows the implementation and use of cryptographic material in various system components to be controlled in an even larger number of different application scenarios.

Cryptographic material intended for a system component fulfills a specific role there. If several cryptographic materials of the same type are used by the same system component, for example several public 4096-bit RSA keys, the system component itself cannot distinguish which role the new cryptographic material should play, i.e. how and where it should be in the system component to install and how to use it. This information pertaining to the roll of cryptographic material may be attached to the cryptographic material in the form of the roll. The role differs from the condition in that the role does not place any conditions on the current status of the system component.

For a particularly efficient distinction in which system component cryptographic material may be used, the cryptographic material can be supplemented with a target component identity. The target component identity includes a clear identification of the system components that are allowed to use the corresponding cryptographic material. If the corresponding cryptographic material is applied to a system component that is not in the Target component identity is included, the use of the cryptographic material on this system component is prevented. This makes it particularly easy and quick to provide which system components may use which cryptographic material and which may not.

The target component identity can be directed to an entire class of systems, system components and/or their sub-modules, for example to the class “head unit”, “engine control unit”, “flash memory” or the like, as well as to exactly one special system component , such as the hardware security module with serial number GHNS-1934952.

With the help of at least one condition, at least one role and/or at least one target component identity, it is thus possible to control very efficiently and flexibly which cryptographic material may be used in which application scenarios, i.e. in which states of which system component and/or its submodules, and which not.

The cryptographic material is typically designed as a crypto material (data) packet and exchanged between systems or system components or introduced there. The cryptographic material package includes at least the actual cryptographic material, for example an asymmetric pair of keys, and the respective condition, role and/or target component identity in the form of additional data attached to the cryptographic material. For this purpose, the attached data can be concatenated with the cryptographic material, for example. The crypto material package can also include further data. For the sake of simplicity, only the cryptographic material is discussed in this text.

An advantageous development of the method according to the invention provides that the cryptographic material comprises at least one role specific to the target component. If one and the same cryptographic material is sent to different system components, it can also fulfill different roles on the different system components. In this way, target component-specific roles can be introduced into the cryptographic material, which tell the cryptographic material on which system component it should be used and how. This allows control even more extensively and flexibly how cryptographic material is implemented on various system components and used there.

According to a further advantageous embodiment of the method according to the invention, all conditions are defined by a system component-external creator and evaluated by at least one evaluator running in an environment of the system component, with the creator and the evaluator jointly specifying which variables are used by the creator when defining conditions should. For this purpose, the creator and the evaluator can use a fixed set of possible variables and select suitable variables depending on the condition. The creator selects suitable variables for each condition.

Likewise, the number and type of conditions is determined by the creator. The name and permissible value range used for the respective variable is then determined jointly by the creator and the evaluator, with the evaluator determining which names and value ranges the creator may use. However, the creator decides for himself which names and value ranges he ultimately uses.

The creator is, for example, the already mentioned crypto material server. The evaluator can be trained by hardware and software or by a human being. The evaluator can be included in the system component or it can also be implemented externally to the system component, but it can be provided in the same environment as the system component. This allows the evaluator to capture the variables underlying the system component and describing the corresponding state of the system component. An evaluator external to the system component enables the condition(s) imposed on the system component by means of the cryptographic material to be checked for a system component that has not yet been booted. Thus, for example, verified cryptographic material corresponding to the system component can be stored in a memory without the corresponding system component having to be started up beforehand.

In particular, the evaluator is able to check at any point in time whether a possible condition for the respective system component is met or not. If the creator and the evaluator agree on a common set of relevant variables, it can be ensured that the variables required to evaluate a certain condition can actually be recorded by the evaluator. This increases the reliability in the application of the method according to the invention.

A further advantageous embodiment of the method also provides that at least one variable has one of the following value ranges:

- BOOLEAN;

- INTEGER; or

- STRING.

All variables can have the same value range, i.e. be of the BOOLEAN, INTEGER or STRING type, or individual variables can also have different value ranges. By using the different value ranges or types, a wide variety of variables can be used to define and subsequently check conditions.

Correspondingly, a variable of the BOOLEAN type assumes the value TRUE or FALSE. With the help of the INTEGER and STRING value ranges, states can be described in even more detail or in a more complex way. For example, a specific state can be described using numbers and/or character strings.

A range of values can range from 1 to 10, for example. The variables used to define a state can be in the form of a vector, for example. For example, the vector may include a first variable of the BOOLEAN type and a second variable of the INTEGER type. This variability vector is time dependent. At a certain point in time, for example, the first variable can have the value TRUE and the second variable have the value 24. In this way, a condition for each system component can be evaluated by the evaluator whose environment is in a state described by the current values of the variables at a specific point in time t, i.e. it can be checked whether the variables at point in time t meet the defined conditions fulfill or not. The corresponding variable vector can also include at least one empty element. This is the case, for example, when the evaluator cannot capture a particular variable. Instead of an empty element, a placeholder can also be used, such as the number "0", "9999" or a character string such as "NAN".

Likewise, variables can be of a structured type, for example variables can be a:

- ARRAY or a

- RECORD, where structured types can be formed from both unstructured and structured types.

According to a further advantageous embodiment of the method, at least one condition is defined in a target-component-specific and/or operation-specific manner. Thus, the cryptographic material may be used and/or used by various system components to perform various operations such as installing software, locking or decrypting certain data, creating or verifying a signature, or the like. In order for the creator to be able to mark a condition as operation-specific, i.e. only valid for a specific operation, a number of operations known within the system component must be communicated to the creator, analogous to the number of usable state variables, so that they can be used when defining conditions can. This set of operations can be general, i.e. valid for all target components, or target-component-specific.

In order for a specific condition to be considered fulfilled on different system components and/or when performing different operations, it can be the case that for at least two different system components and/or operations a specific variable has a different value, i.e. TRUE once and FALSE once, or once 0 and once 256.

A further advantageous embodiment of the method according to the invention also provides that for a target-component-specific and/or operation-specific Condition on a different target system component and / or in a different operation at least one different variable is used. Not only can a specific variable assume different values in order to fulfill a condition on different system components and/or to carry out different operations, but different variables can also be required for this purpose. In particular, the quantity and/or type as well as a size of a permissible value range of the variables relevant to a condition can differ between different system components and/or operations. For example, in order to fulfill a condition on a first system component, the variables 1 and 2 must assume the values 0 and TRUE and on a second system component the variables 1, 2 and 3 must assume the values 0, TRUE and FALSE. On a third system component, however, to fulfill the same condition, the variables 1, 2, 3, 4, 5 and 6 must have the values TRUE, [0..100], [-100..100], 0, "engaged" and FALSE assume. In general, at least two BOOLEAN expressions can be OR linked, which will be discussed below.

According to a further advantageous embodiment of the method, an evaluator checks a condition by means of an evaluation function, wherein at least two evaluators who differ from one another use an evaluation function that differs from one another, in particular an operation-specific evaluation function. It is generally possible for all evaluators to use the same evaluation function. However, the provision of different evaluation functions increases flexibility in the implementation of the method according to the invention for controlling the conditions under which cryptographic material is used by system components.

In general, a generic evaluation function can also be used, whereby all cryptographic material, or the conditions covered by the cryptographic material, can be evaluated by providing only one function. However, the evaluation function can also be designed specifically for different cryptographic materials, for example depending on the role of the cryptographic material. In particular, a distinction is again made for different evaluation functions for different operations. A further advantageous embodiment of the method according to the invention also provides that at least one condition is defined in a machine-processable definition language, with an executable language that can be interpreted by an interpreter or one of the following formal logics being used:

- propositional logic;

- propositional logic with relations; or

- Propositional logic with relations and functions.

The definition of the condition in a machine-processable definition language makes it possible to use existing languages for carrying out the method according to the invention. Using a definition language that can be evaluated, conditions can be formulated about the valid variables that describe a state of the system component, for example as BOOLEAN-valued expressions that are defined and processable by means of the definition language. If the cryptographic material has several conditions, a set of variables relevant to the corresponding condition can also contain different variables. All variables of a corresponding set of variables can be of the same type, i.e. BOOLEAN, INTEGER or STRING, or at least two variables of a set of variables relevant to a respective condition can differ in their type. Also, the same or different machine-processable definition languages can be mixedly used for different conditions comprised by the cryptographic material.

In the well-known propositional logic, all variables relevant to a specific condition have the value range BOOLEAN. The individual variables can therefore only have the values TRUE or FALSE. Logical formulas are formed with the help of well-known propositional logical connectives, i.e. "A", "V", or the like, as well as brackets "(" ")". Examples of propositional formulas are:

- Variabiei;

- Variable2 A TRUE;

- (Variablei A (Variable2 v Variables)). In propositional logic with relations, variables relevant to a specific condition can have different value ranges. There is a finite set of relations, with each relation being assigned a fixed arity, and each relation and the position being assigned a fixed range of values. A relational term is then obtained by applying a relation to a number of constants and/or variables with the appropriate value ranges that corresponds to the arity. When evaluating a relational term for a fixed value assignment of the variables occurring in this term, the result is always a Boolean value, i.e. TRUE or FALSE. An example of a two-digit relation that requires INTEGER as the value range in both places, for example, is the less than or equal relation "<", relational terms based on this relation would be, for example:

variable < variable2; 7 < variables.

Logical formulas are formed with the help of BOOLEAN constants, BOOLEAN variables, relational terms and well-known propositional logical connectives A, V, etc. as well as brackets "(" and ")" to determine the evaluation order. An example of a propositional formula with relations that uses the two-digit relations "<" and "=" is:

(Variablei A ((Variable2 < Variables) v (Variables = Variable4))).

In the case of propositional logic with relations and functions, a finite set of functions is also used, with each function being assigned a fixed number of digits, and each function and each digit being assigned a fixed range of values. In addition, the value range of the result is defined for each function. A function is then applied to a number of constants, variables and/or functions with the appropriate value ranges that corresponds to the arity and returns a value from the specified result value range. An example of a two-digit function that requires INTEGER as the range of values in both digits and returns an INTEGER value as the result is the addition "+". Logical formulas are formed with relations as in propositional logic, with the difference that relational terms are not only formed by applying relations to constants and/or variables, but also to functions applied to constants/variables with the corresponding result Value range can be formed. An example of a propositional formula with relations and functions is:

- (Variablei A ((Variable2 < (Variables + Variable?)) v (((Variables - Variables) + Variable2) = Variable4))).

In particular, the use of an executable language that can be interpreted by an interpreter offers advantages. Such languages are very “powerful” due to their executability. Thus, before executing a corresponding program code, no previous translation/compilation, or transformation into a formula tree and insertion of variable values, is necessary. An interpretable executable language can be, for example, a scripting language such as Python.

According to a further advantageous embodiment of the method according to the invention, at least two conditions are formulated in a definition language that differs from one another, in particular two target-component-specific conditions are formulated in a definition language that differs from one another. If a first system component uses a first language and a second system component uses a second language that differs from the first language, the provision of two conditions formulated with different definition languages ensures that both system components can also evaluate the condition relevant to them. The variables relevant to the conditions are formulated accordingly in the respective definition language. It is also possible for two conditions valid for the same system component to be formulated in a definition language that differs from one another. So it can be advantageous to formulate a certain condition with a certain definition language. This is the case, for example, if certain variables can only be recorded and/or processed with a certain definition language or if this is specified due to a wide variety of boundary conditions.

A further advantageous embodiment of the method according to the invention also provides that the cryptographic material comprises at least two conditions and all of the conditions must be met in order for the cryptographic material to be used by the system component. Referring back to the use of formal logic, this means that the conditions used are AND-linked. So must all Conditions must be met for the cryptographic material to be used by the system component. For example, the fulfillment of a single condition for the use of the cryptographic material by the system component is necessary, but not necessarily sufficient, since at least one further condition must be met.

According to a further advantageous embodiment of the method, the evaluator supplies, if cryptographic material is to be used by a system component whose target component-specific condition does not include the class of the corresponding system component, if an operation is to be carried out using the cryptographic material which is not covered by an operation-specific condition or when an evaluator cannot capture a variable required to check a condition, the response for that condition defaults to:

- TRUE;

- FALSE; or

- a response attached to the cryptographic material as a default response.

This ensures an even more reliable implementation of the method according to the invention. In general, a system component, or a class of the system component, from which certain cryptographic materials are to be used, cannot be included in a target-component-specific condition. So that the cryptographic material can still be used by the corresponding system component, the value TRUE is output by default after the target component-specific condition has been checked. If, in such a case, the use of the cryptographic material is to be prevented instead, the value FALSE is output by default. To increase the flexibility of the method according to the invention, the standard response in the form of TRUE or FALSE can also be appended to the cryptographic material. This can reduce the risk that a system component whose default response is set to TRUE will have the cryptographic material applied when it really shouldn't. In this case, the default response in the form of FALSE is appended to the cryptographic material. The formulations made above apply correspondingly to an operation-specific condition, as well as in the event that an evaluator cannot record a variable required to check a specific condition.

A further advantageous embodiment of the method according to the invention also provides that all additional data comprised by the cryptographic material is cryptographically secured against being compromised, in particular using at least one digital signature and/or a symmetrical integrity protection mechanism. In particular, a keyed hash message authentication code (HMAC) can be used as a symmetrical integrity protection mechanism. Securing the cryptographic material against compromise increases the protection even further when the method according to the invention is carried out. In this way, the additional data attached to the cryptographic material package in addition to the cryptographic material is protected against unauthorized manipulation. For example, the cryptographic material together with the associated additional data can be digitally signed by the creator using his asymmetric private key. In particular, secure procedures and secure systems are used for signing. The public key or a certificate containing the public key is then distributed to all evaluators concerned. The corresponding public key or certificate is introduced in a tamper-proof manner in the corresponding systems. For example, the public key or the certificate is stored in a read-only memory (ROM) or in a memory that can only be written once (WORM). Correspondingly, the crypto-material package includes such a digital signature. Accordingly, the cryptographic material is only used by the system component if a check of the signature attached to the cryptographic material using the introduced public key or the certificate supplies a positive result.

According to a further advantageous embodiment of the method according to the invention, at least two different actors are authorized to digitally sign the cryptographic material, with all actors authorized to sign receiving their own leaf certificate belonging to a common certificate hierarchy with the associated private key. In particular, it makes sense to give several actors permission to digitally sign albeit multiple actors create cryptographic material. A system performing the signature then signs the corresponding cryptographic material with the private key belonging to its leaf certificate. Not only the generated signature is then added to the cryptographic material, but also the entire certificate chain. The certificate chain attached to the cryptographic material is then used by a corresponding evaluator or a system comprising the evaluator in order to check the correctness of the signature created by the signing system.

If the cryptographic material is already a digital certificate, it is signed from the start. If additional data, i.e. at least one condition, role and/or target component identity, is then included in the certificate, this is also signed by the creator of the certificate. In this way, the additional data, or just parts of it, can be included in the corresponding certificate as one or more certificate extensions and thus co-signed. If all additional data is included in the certificate, the dedicated creation of the digital signature for the cryptographic material can be dispensed with.

To ensure the best possible protection, the method according to the invention should be used as early as possible when developing a corresponding system component. Since the condition check by the evaluator depends on variables, these variables must be protected against manipulation as best as possible, since corrupted variables can simulate the existence of a non-existent state. Accordingly, constants used to evaluate a condition are preferably stored in a write-once memory such as a WORM memory. The values of the variables on which the conditions depend are preferably systematically adapted to the current system state during the entire life cycle of the information technology system or of the individual system components.

A vehicle ecosystem is preferably used as the information technology system. In general, the method according to the invention can be used for different system components in different types of networked or non-networked information technology systems. Use is particularly efficient when information technology system is a vehicle ecosystem, since the corresponding requirements in terms of complex development with many partners involved often make compliance with safety guidelines very difficult in practice. The risk of a potential security gap can be drastically minimized by the implementation of the method according to the invention and thus a self-check by the respective system component.

Further advantageous refinements of the method according to the invention for implementing and using cryptographic material in at least one system component of an information technology system also result from the exemplary embodiments, which are described in more detail below with reference to the figures.

show:

1 shows a schematic representation of a first use of cryptographic material on a system component of an information technology system;

FIG. 2 shows a schematic representation of a use of alternative cryptographic material on the system component shown in FIG. 1;

3 shows a schematic representation of a use of alternative cryptographic material on a further system component; and

4 shows a flowchart of an exemplary processing of cryptographic material on a system component when executing an operation.

FIG. 1 shows a first exemplary embodiment for using cryptographic material KM from at least one system component SK of an information technology system IT-S. A state of the system component SK is described using at least one variable VAR. In the example in FIG. 1, the set of all variables VAR is referred to as VAR and the individual status variable used here as “productive”. The status variable productive indicates whether the system component SK is in a productive phase of its life cycle, i.e. it is a productive system or not. The cryptographic material KM is a self-signed test root certificate TestRootCert belonging to a test root key pair (TestRootPub, TestRootPriv), which is introduced into the target system as a trust anchor to be able to set up test connections with communication partners by checking the trustworthiness of the certificates signed with TestRootPriv received from partners using TestRootCert.

The TestRootCert certificate was created under insecure conditions, for example the private key TestRootPriv is not securely stored. Thus, TestRootCert must not be used in production systems. Accordingly, a condition BED comprised by the cryptographic material KM is defined as productive=FALSE. The BED condition states that the TestRootCert certificate may only be used by the target system, i.e. the system component SK, if it is not in the productive phase.

FIG. 2 shows a similar exemplary embodiment, in which the cryptographic material KM now includes a securely generated certificate ProdRootCert that can be used securely in productive systems. Accordingly, the ProdRootCert certificate may always be used by the target system, i.e. the system component SK, for example also in the development and productive phases. ProdRootCert may also be used in the development phase because it does not contain any secret parts. Accordingly, the BED condition is not productively dependent on the status variable, but is always considered to be fulfilled (TRUE). Since the condition BED does not depend on any variable VAR, one could also choose an empty set of state variables, i.e. BED _TestRootcert ({}):=(TRUE) instead of BED _TestRootcert ({productive}):=(TRUE).

FIG. 3 shows a further exemplary embodiment of the implementation and use of cryptographic material KM in a system component SK. The system component SK has two status variables here, namely the two elements HSMProvisioningSicher: BOOLEAN and HSMVerschlaecher: BOOLEAN of the set of variables VAR. The set of variables VAR here corresponds to a vector with two elements. The state variable HSMProvisioningSicher indicates whether a hardware security module HSM is already in a state in which it can securely receive and store cryptographic material KM. The state variable HSMVerschleichSecure indicates whether the hardware security module HSM is already in a state in which it can securely use secret keys stored in it for encryption. The cryptographic material KM is shown in FIG 3 around a 256-bit long, secret symmetric key AESKey, which is to be used for AES encryption in production systems.

Two operations are to be protected by conditions BED. On the one hand, this involves the provision of cryptographic material KM, ie the secure introduction and secure storage of a key in the system component SK. Encryption is also involved, ie the secure use of a key for encryption in the system component SK. The secure use of the key in the target system is ensured by two conditions BED BEDAESKe _y ^provisionin9 and BEDAESKe _y ^encryption9 . The condition BED BEDAESKe _y ^Provisionin9 states that the key AESKey may only be brought into the target system if HSMProvisioningSicher = TRUE. The condition BED BEDAESKe _y ^Encryption9 states that the key AESKey may only be used for encryption in the target system if HSMVerschliffSicher = TRUE.

FIG. 4 shows a flow chart of a method according to the invention. It is assumed that a cryptographic material packet KM-P has the following form: KM-P=(KM, ZKIDENTKM, BEDKM*, ROLEKM*, Sign™, ZK), where ZKI DENTKM is the entirety of the target component identities permitted for the cryptographic material KM ZKI DENTKM, BEDKM* denotes the entirety of the cryptographic material KM, possibly target component-specific and/or operation-specific conditions BED, and ROLEKM* denotes the entirety of possibly target component-specific roles ROLEKM*. At least one of the variables mentioned ZKI DENTKM, BEDKM*, ROLEKM* can also be missing in the crypto material package KM-P. Sign is the signature created by the creator of the cryptographic material KM (KM, ZKIDENT™, BED™*, ROLE™*) and ZK is the certificate chain, with the help of which the system component SK can check the correctness of the Sign™ signature. For unique addressing of a system component SK, it is also assumed that the system component SK has the identity skid and the target system component type Type(skid). For example, the identity skid and the target system component type Type(skid) on a system component SK can be included in the variable VAR. A system component type can also be understood as a class of system components SK, for example all system components SK of the class/type “head unit”, “engine control unit” or the like. It is also assumed that the evaluator wants to check whether the cryptographic material KM contained in the crypto material package KM-P may be installed in the system component SK for an operation Prov in the system component SK. The operation can generally be any operation, such as performing a decryption process. In the example in FIG. 4, the operation is the actual provisioning of the cryptographic material KM.

First, the Signature Sign™ is checked. If the signature check is unsuccessful, an exception handling follows according to a first link LINK1. If, on the other hand, the signature check is successful, it is checked whether the cryptographic material KM, here in the form of the cryptographic material package KM-P, includes at least one target component identity ZKIDENT™. If this is the case, then it is checked whether the identity skid of the system component SK is included in the target component identity ZKIDENT™, and if not, according to the first link LINK1, exception handling is initiated. However, if this is not the case, i.e. the crypto material package KM-P does not include a target component identity ZKIDENT™, this step is skipped.

It is then checked whether and, if so, which conditions BED * are contained in the cryptographic material KM. If the cryptographic material package KM-P does not include a BED™* condition, then a second link LINK2, which could mean installation of the cryptographic material KM, is followed.

It is then examined whether at least one condition BED™* is target component-specific.

If this is not the case, a check is made as to whether at least one BED™* condition is operation-specific.

If this is not the case, then the corresponding condition BED™ is evaluated and the process is continued according to the first or second link LINK1, LINK2.

If, on the other hand, it is determined that the at least one condition BED™* is operation-specific, then a check is then carried out as to whether a Provision of the cryptographic material KM is provided, so a provisioning. If this is the case, the operation-specific condition BEDKM ^Prov is checked. The procedure then proceeds analogously to the links LINK1, LINK2.

On the other hand, if at least one type-specific condition was recognized in the crypto material package KM-P, a check is made to determine whether at least one of the type-specific conditions is also valid for an entire class of system components SK, ie whether Type(skid) is contained in BEDKM*. The set of conditions valid for the respective classes referenced by Type(skid) is referred to below as BEDKM ^Type(skid) *. If this is the case, provisioning may take place (see first link LINK1). If this is not the case, it must be checked which standard procedure should be used. So whether provisioning may take place or not. For this purpose, for example, a standard response can be included in the crypto material packet KM-P.

It is then checked whether the type-specific condition BEDKM ^Type(skid) * is also operation-specific. If this is not the case, the condition BEDKM ^Type(skid) is evaluated and the links LINK1 , LINK2 are followed. If, on the other hand, the conditions BEDKM ^Type(skid) * are operation-specific, it is checked whether there is at least one operation-specific condition in BEDKM ^Type(skid) *, here as a representative example for the operation Prov ("Provisioning"), i.e. a condition BEDKM ^Type(skid) ' ^Prov , (this step is not shown in Figure 4). If this is not the case, exception handling is initiated according to the first link LINK1. If this is the case, however, if BEDKM ^Type(skid) * contains a condition BEDKM ^Type(skid) ' ^Prov , then this is evaluated and the procedure is continued in the same way as the links LINK1 , LINK2.

After evaluating the individual conditions BEDKM, BEDKM ^Prov , BEDKM ^Type(skid) or BEDKM ^{Type(skid), Prov,} it is checked whether the cryptographic material KM, here in the form of the crypto material package KM-P, includes at least one role ROLEKM*. If this is not the case, the cryptographic material KM is made available for the system component SK without a role. If, on the other hand, this is the case, it is then checked whether ROLEKM* includes a role for the target component type Type(skid). If this is not the case, exception handling occurs again. If this is the case, however, the cryptographic material KM is provided taking into account the role ROLEKM*.

Claims

Method for implementing and using cryptographic material (KM) in at least one system component (SK) of an information technology system (IT-S) for performing at least one operation, wherein at least at a first point in time a state described by at least one variable (VAR) of the system component (SK) is checked, the cryptographic material (KM) is supplemented with additional data, with the additional data describing possible states of system components (SK), and the cryptographic material (KM) is used by the system component (SK) if the additional data of the cryptographic material (KM) include at least the state that the system component (SK) has at the first point in time, characterized in that the additional data of at least one condition (BED, BEDKM*, BEDKM ^Prov , BEDKM ^Type(skid) , BEDKM ^{Type(skid ) Prov} ), at least one role (ROLEKM*) and/or at least one target component identity (ZKIDENTKM). be loaded. Method according to Claim 1, characterized in that the cryptographic material (KM) comprises at least one target component-specific role (ROLEKM*). The method according to claim 1 or 2, characterized in that all conditions (BED, BEDKM *, BED _K M ^Prov , BED _K M ^{Type (skid)} , BED _K M ^{Type (skid)} ' ^Prov ) from a system component (SK) external Creator defined and evaluated by at least one in an environment of the system component (SK) running evaluator, the creator and the evaluator together determine which variables (VAR) may be used by the creator in the definition. Method according to one of Claims 1 to 3, characterized in that at least one variable (VAR) has one of the following value ranges:

- BOOLEAN;

- INTEGER; or

- STRING. Method according to one of Claims 1 to 4, characterized in that at least one condition (BEDKM ^Prov , BEDKM ^Type(skid) , BEDKM ^Type(skid) ' ^Prov ) is defined in a target-component-specific and/or operation-specific manner. The method according to claim 5, characterized in that for a target component-specific and / or operation-specific condition (BED _K M ^Prov , BED _K M ^{Type (skid)} , BED _K M ^{Type (skid)} ' ^Prov ) on a different target component and / or at a different operation at least one different variable (VAR) is used. Method according to one of Claims 3 to 6, characterized in that an evaluator checks a condition (BED, BEDKM*, BEDKM ^Prov , BEDKM ^{Type (skid)} , BEDKM ^{Type (skid)} ' ^Prov ) by means of an evaluation function, with at least two mutually deviating Evaluators use an evaluation function that differs from one another, in particular an operation-specific evaluation function. Method according to one of Claims 1 to 7, characterized in that at least one condition (BED, BEDKM*, BEDKM ^Prov , BEDKM ^Type(skid) , BEDKM ^Type(skid) ' ^Prov ) is defined in a machine-processable definition language, with a executable language that can be interpreted by an interpreter or one of the following formal logics is used:

- propositional logic;

- propositional logic with relations; or

- Propositional logic with relations and functions. Method according to Claim 8, characterized in that at least two conditions (BED, BEDKM*, BEDKM ^Prov , BEDKM ^Type(skid) , BEDKM ^Type(skid) ' ^Prov ) are formulated in a definition language that differs from one another, in particular two target component-specific conditions

(BED _K M ^Type(skid) , BED _K M ^Type(skid) ' ^Prov ). Method according to one of Claims 8 or 9, characterized in that the cryptographic material (KM) comprises at least two conditions (BED, BEDKM*, BEDKM ^Prov , BEDKM ^Type(skid) , BEDKM ^Type(skid) ' ^Prov ) and all the conditions ( BED, BEDKM*, BED _K M ^Prov , BED _K M ^Type(skid) , BED _K M ^Type(skid) ' ^Prov ) must be fulfilled in order for the cryptographic material (KM) to be used by the system component (SK). Method according to one of Claims 5 to 10, characterized in that if cryptographic material (KM) is to be used on a system component (SK), its target component-specific condition (BEDKM ^{Type (skid)} , BEDKM ^{Type (skid)} ' ^Prov ) the class of the corresponding System component (SK) not included if an operation is to be carried out using the cryptographic material (KM) which is not covered by an operation-specific condition (BEDKM ^Prov , BEDKM ^Type(skid) ' ^Prov ) or if an evaluator is to check a state cannot capture required variable (VAR), the evaluator for the given condition (BED, BEDKM*, BEDKM ^Prov , BEDKM ^Type(skid) , BEDKM ^Type(skid) ' ^Prov ) returns the following response by default:

- TRUE;

- FALSE; or 27

- a response attached to the cryptographic material (KM) as a standard response. Method according to one of claims 1 to 11, characterized in that all of the cryptographic material (KM) included additional data are cryptographically protected against compromise, in particular using at least one digital signature and / or a symmetrical integrity protection mechanism. Method according to Claim 12, characterized in that at least two different actors are authorized to digitally sign the cryptographic material (KM), with all actors authorized to sign receiving the individual private keys assigned to them together with the associated individual leaf certificates and the associated certificate chain . Method according to one of Claims 1 to 13, characterized in that a vehicle ecosystem is used as the information technology system (IT-S).