Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
157 tokens/sec
GPT-4o
43 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

AdSplit: Separating smartphone advertising from applications (1202.4030v1)

Published 17 Feb 2012 in cs.OS

Abstract: A wide variety of smartphone applications today rely on third-party advertising services, which provide libraries that are linked into the hosting application. This situation is undesirable for both the application author and the advertiser. Advertising libraries require additional permissions, resulting in additional permission requests to users. Likewise, a malicious application could simulate the behavior of the advertising library, forging the user's interaction and effectively stealing money from the advertiser. This paper describes AdSplit, where we extended Android to allow an application and its advertising to run as separate processes, under separate user-ids, eliminating the need for applications to request permissions on behalf of their advertising libraries. We also leverage mechanisms from Quire to allow the remote server to validate the authenticity of client-side behavior. In this paper, we quantify the degree of permission bloat caused by advertising, with a study of thousands of downloaded apps. AdSplit automatically recompiles apps to extract their ad services, and we measure minimal runtime overhead. We also observe that most ad libraries just embed an HTML widget within and describe how AdSplit can be designed with this in mind to avoid any need for ads to have native code.

Citations (262)

Summary

  • The paper introduces AdSplit, which decouples ad code from the host app to mitigate permission bloat and security risks.
  • The authors implemented automatic recompilation to isolate ad components, achieving efficient processing with 183 events per second on older hardware.
  • AdSplit’s architecture improves security and privacy, paving the way for modular, ad-integrated mobile platforms in future research.

AdSplit: A Novel Approach to Addressing Permissions and Security in Smartphone Advertising

The paper "AdSplit: Separating smartphone advertising from applications" addresses a crucial issue in smartphone application design: the integration of advertising libraries into app code and the resultant permission bloat and security vulnerabilities. The authors propose AdSplit, a modified architecture on Android that separates advertisements from the host application. By running ads as separate processes under distinct user identities, AdSplit mitigates several security concerns while reducing unnecessary permission requests associated with ads.

Key Insights and Contributions

The primary motivation for AdSplit stems from the security and privacy challenges linked to embedded advertising libraries in mobile applications. Typically, these libraries request additional permissions, contributing to "permission bloat" and potentially scaring users away due to intimidating permission prompts. Moreover, malicious applications could impersonate advertising behaviors, thereby defrauding advertisers by logging user interactions deceptively. AdSplit endeavors to resolve these challenges by separating ads from their host applications both logically and in terms of system processes.

The authors conducted an empirical paper of approximately 10,000 Android applications to highlight the degree of permission bloat attributed to advertising. It was discerned that a significant proportion of permission requests were primarily for the benefit of advertising libraries. AdSplit was shown to be effective in reducing this bloat by managing advertisements in a separate manner.

A crucial feature of AdSplit is its ability to automatically recompile applications, extracting advertising components and managing them as independent applications. This automatic recompilation involves minimal runtime overhead, demonstrating efficiency in real-world scenarios. The paper reports the ability to manage around 183 events per second, which is sufficiently high permformance for typical usage scenarios on older Android hardware.

The implementation details reveal several design choices, such as maintaining distinct permission sets for applications and ads, and managing lifecycle events separately. This design ensures that advertisements only run concurrent with their host application to conserve resources. Furthermore, it securely verifies ad interactions, guarding against click fraud by exploiting mechanisms akin to the Quire architecture to certify that UI events are genuine.

Practical and Theoretical Implications

Practically, AdSplit promises a more secure and privacy-preserving environment for mobile users. By isolating the advertising code and minimizing the permissions required for app installation, developers can potentially improve user trust and satisfaction. AdSplit's mechanism for ensuring ad integrity could also incentivize advertisers by providing a verified platform that deters fraudulent activities.

Theoretically, AdSplit underscores a direction towards more modular and secure app architectures that can effectively decouple third-party services from core application functionalities. It opens avenues for further research into privacy-preserving advertisement techniques and architectures that align with emerging web standards.

Future Directions

The authors identify potential extensions of their work, such as further integration with HTML-driven advertisements and improvements in the handling of permissions for richer, more targeted advertising. There's also the proposition of addressing user experience issues related to separate ad installations and managing distinct permissions in a user-friendly manner.

Overall, AdSplit initiates a crucial dialogue in the field of secure mobile computing and is illustrative of the evolving intersection between application design and security. As the mobile ecosystem grows increasingly complex, approaches like AdSplit could pave the way for more refined solutions to persistent privacy and security challenges in mobile platforms.