Android security remains an unsolved
problem. The vast popularity of Google’s mobile system works
for cybercriminals as honey for flies, so the number of targeting
the attacks is growing. There are also increasing the professionalism of the attackers,
interested not so much vandalism at making money. No
so long ago it was loud even with the detected
by FireEye operation, in which the application pretending to be official store
Google Play stealing login credentials for electronic banking.
Security researcher scientists from Germany and the United States consider
that neither Google nor Android device manufacturers themselves from
it can not do – the system needs a total rethink
its security. For its part, offers not only
thoughts, but also very interesting programming framework that
no help in securing Android.
The problem of adequate protection of Android stems from the large
number of complex programming interfaces (APIs) used
by the system – say researchers from the Technische Universität
Darmstadt and North Carolina State University. It would not be without them
possible rise throughout this vast database software that
Google has a system, but it also leads to the need
use a completely different security model than in the
conventional operating systems (e.g.. Windows). In Android, each
application is assigned to its own security domain, for
possibly a small area, often containing only the necessary
API and created by no files. Thanks are limited
both the consequences of the exploit and the powers that
gets the worm. Not really but you know what permissions
application should be granted, and when to do it, nor how to deal
the threats arising from the interaction between different
API.
Ideas to change the security models of Android has already been
a lot, but they all require changes in the firmware, often in
practice impossible to achieve – hundreds of millions of devices
this system will never be updated to newer versions
the system. For their part, researchers and so offer innovative
Android framework called Security Modules (ASM), which can be
wpinać various security fixes. In his article
Fri
(Which tomorrow is to be presented at the Usenix Security
Symposium in San Diego, CA) explain that the ASM provides a set of
authentication hooks (
You can build your security software. Such a standard
interface for security architectures allow for secure
ongoing to improve the security of Android devices, without
need to modify the operating system itself.
ASM Developers believe that you can not do a single model
security that would meet the needs of all potential
operating system users (individuals, companies or
government). It is therefore necessary to have
programmable interface security in accordance with the expandable
needs, like the Linux Security Modules (GNU / Linux)
or TrustedBSD (BSD). ASM must of course retain the original
existing Android security, providing a more
restrictive security without compromising example. sandboksa and protect
consistency of the kernel, servicing modules running
authentication without any additional overhead on system performance.
In this way it becomes possible to complete separation applications
business and personal equipment with Android, as well as
increase privacy. Available by ASM
Call it permits filter, modify, and anonimizować
data transmitted between applications, thereby protecting the
all on the device data. Then, even if the
an application will request it unnecessary things – eg.
access to the contact list, ASM will be able upon request
filter, so that the manufacturer of the application server will go only
Data apparent.
More about Android Security Modules, can be found at
the project. The researchers also prepared two working examples
applications that implement security known to MockDroida
- Processing allowing Android to control access
Application data and application AppLock,
enables blocking of individual applications.
No comments:
Post a Comment