The present invention relates to a data processing device, in particular an electronic memory component, comprising a plurality of access-secured sub-areas, in particular a plurality of access-secured memory areas, each having at least one assigned parameter (an, an−1, . . . , a1, a0), in particular address.
The present invention further relates to a method of encrypting at least one parameter (an, an−1, . . . , a1, a0), in particular the address, of at least one access-secured sub-area, in particular at least one access-secured memory area, of at least one data processing device, in particular at least one electronic memory component.
In known methods of encrypting confidential data, such as personal data, key data or otherwise sensitive data, a non-volatile memory unit can only be encrypted as a complete physical memory in a generally fixed manner; this means that access can only be denied to memories in their entirety.
This method, known from the prior art, of encrypting entire integrated circuit (IC) areas is considered disadvantageous in view of the high cost together with its technical complexity and lack of flexibility. For this reason, it is desirable to develop alternative methods of encrypting access-secured memory areas or sub-areas.
If, for instance, to control a memory of the size M=2i=2n+1 with i=n+1 address buses, when these address buses are encrypted over the entire address space, modification of one address bus would have the possible effect of modifying a plurality of address buses, indeed even those address buses which ensure that a physically remote memory cell is addressed.
This is not sensible for a number of a memory types, including in particular memories that are organized into areas, such as erasable programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM) or Flash memory. Separation of the address buses into a number of areas and subsequent independent encryption of each of the individual areas is inadequate, however, with regard to security.
Taking as basis the above-described disadvantages and shortcomings and acknowledging the outlined prior art, it is an object of the present invention to develop a data processing device, in particular an electronic memory component, of the above-mentioned type, together with an encryption method related thereto, that the security of the device is increased considerably and the expense associated therewith and the technical complexity are not too great.
The present invention provides a completely novel approach to area-wise encryption of memory contents, i.e., a new method is disclosed for encrypting access-secured memory sectors of non-volatile nature and/or other sub-sectors.
To this end, the present invention allows parts of the (address) parameters of the memory areas to be encrypted in different ways with regard to the object and/or with regard to the customer and/or with regard to the “die”. This means that some sub-areas or sectors of the address do not affect all the addresses, unlike in the prior art.
According to the invention, therefore, encryption of one access-secured sub-area, in particular an access-secured memory area, is performed while taking account of the other respectively available sub-areas, in particular memory areas. This makes it possible to encrypt each sub-area with in each case different parameters.
In an embodiment of the present invention shown in
In this context, it is obvious that although variation of the parameter an, in particular of the address parameter, may influence all the other address buses, variation of the parameter an−1 does not have any influence on the most significant function f1(an).
It is expedient for fi(a) to be any desired one-to-one function, i.e. there are precisely 2i plain/cipher pairs, wherein an unencrypted address an, an−1, . . . , a1, a0 is always transformed into a unique encrypted address a′n, a′n−1, . . . , a′1, a′0. On the other hand, the function fi itself does not have to be bijective, i.e., it does not have to be reversible.
In an advantageous further embodiment of the present invention, not all stages have to be fully performed, i.e., some functions fi may directly reproduce the relevant address bit: a′=a. Alternatively or in addition thereto, the address buses may also be grouped; this may appropriately mean, inter alia, that the inputs to the functions fi and the return values from the functions fi may be several bits wide.
In an advantageous embodiment of the present invention, as shown in
According to a particularly inventive further embodiment, access-secured memory areas may be separately secured, i.e. boundary conditions which require a physical memory are fully utilized by the new method (the wide variety of encryptions is here limited only insignificantly).
As shown in
The present invention finally relates to the use of at least one data processing device, in particular at least one electronic memory component 77, of the above-described type in at least one chip unit, in particular in at least one smart card controller, in at least one reader IC or in at least one crypto chipset, for example, in the field of audio and/or video encryption.
As already discussed above, there are various possible ways of advantageously embodying and developing the teaching of the present invention. Reference is made, in this regard, to the claims subordinate to claims 1 and 6, and the invention will be further described with reference to examples of embodiments shown in the drawings to which, however, the invention is not restricted. In the Figures:
The encryption method according to the present invention for application in an electronic memory component is based on the idea of encrypting unencrypted addresses an, an−1, . . . , a1, a0 of an access-secured memory area only in certain areas, i.e. in dependence on one or more further memory areas, such that encrypted addresses a′n, a′n−1, . . . , a′1, a′0 are formed.
To this end, i=n+1 one-to-one (→2i=2n+1 plain/cipher-pairs) scramble functions f1, f2, . . . , fn, fn+1 are provided, such that, after mapping, the unencrypted addresses of the form an, an−1, . . . , a1, a0 have the following appearance when encrypted by the functions fi (c.f. FIG. 1):
f1(an), f2(f1(an)+an−1), f3(f2(f1(an)+an−1)+an−2), . . . , fn+1(fn(fn−1( . . . )))
This makes it possible to encrypt each sub-area with in each case different parameters.
In this context, it is obvious that although variation of the addresses an, an−1, . . . , a1, a0 may influence all the other address buses, variation of the parameter an−1, does not have any influence on the most significant function f1(an).
As an alternative to that illustrated in
Furthermore, the address buses may also be grouped; this may mean, inter alia, that the inputs to the functions fi and the return values from the functions fi may be several bits wide.
Number | Date | Country | Kind |
---|---|---|---|
102 56 587 | Dec 2002 | DE | national |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/IB03/05271 | 11/19/2003 | WO | 00 | 6/3/2005 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO2004/051482 | 6/17/2004 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
5081675 | Kittirutsunetorn | Jan 1992 | A |
5754659 | Sprunk et al. | May 1998 | A |
7124303 | Candelore et al. | Oct 2006 | B2 |
7151832 | Fetkovich et al. | Dec 2006 | B1 |
20010037450 | Utyansky et al. | Nov 2001 | A1 |
20020048372 | Toh et al. | Apr 2002 | A1 |
20030026423 | Unger et al. | Feb 2003 | A1 |
20040039908 | Rose et al. | Feb 2004 | A1 |
20070118763 | Kamei | May 2007 | A1 |
20080019528 | Kneissler | Jan 2008 | A1 |
20080229115 | Wollnik et al. | Sep 2008 | A1 |
20090052662 | Oxford | Feb 2009 | A1 |
Number | Date | Country |
---|---|---|
0 908 810 | Apr 1999 | EP |
Number | Date | Country | |
---|---|---|---|
20060101284 A1 | May 2006 | US |