Debug Port Protection Mechanism for Secure Embedded Devices

Abstract

Abstract-In this paper we propose a protection mechanism for the debug port. While debug ports are useful tools for embedded device development and maintenance, they can also become potential attack tools for device hacking in case their usage is permitted to hackers with malicious intentions. The proposed approach prevents illicit use of debug ports by controlling access through user authentication, where the device generates and issues authentication token only to the server-authenticated users. An authentication token includes user access information which represents the user's permitted level of access and the maximum number of authentications allowed using the token. The device authenticates the user with the token and grants limited access based on the user's access level. The proposed approach improves the degree of overall security by removing the need to expose the device's secret key. Availability is also enhanced by not requiring server connection after the initial token generation and further by supporting flexible token transfer among predefined device groups. Low implementation cost is another benefit of the proposed approach, enabling it to be adopted to a wide range of environments in demand of debug port protection.

Fine-Grained Access Management in Reconfigurable Scan Networks

“…Stronger authentication schemes, which do not reveal the secret during communication with the chip, are based on challenge-response protocols [21], [22], [23], [24], [25], [26], [27]: In these schemes, upon a request of an entity e ∈ E to access a protected instrument i ∈ I P , the chip provides a non-repeating challenge value and expects the entity to provide an expected response value. The response is calculated from the challenge, the secret k i , and possibly the secret k e , using a specified cryptographic algorithm, e.g.…”

“…To reduce the amount of online communication, the secure server can also issue credentials which can later be used by the entity to gain offline access to the chip [24]. To limit the number of allowed offline accesses, a similar approach based on authentication tokens was later proposed by the same authors [25]. For provably secure mutual authentication, an application of the Schnorr protocol was proposed recently in [26].…”

“…However, while it is desired that the access rights are managed at the level of individual on-chip instruments distributed over the system-wide scan infrastructure, the techniques presented in [21], [24], [25], [26] allow access management at the level of 1149.1 TAP instructions only. Finegrained access management for individual instruments can be achieved by connecting each protected instrument with the authorization controller, as proposed in [22], [27].…”

“…If required, the security of the proposed scheme can be further improved by replacing the underlying challengeresponse protocol with other authentication mechanisms, e.g. based on credentials or authentication tokens, as in [25], or by using the Schnorr protocol, as in [26].…”

Detection of Illegitimate Access to JTAG via Statistical Learning in Chip

“…Another risk is password leakage, especially when all fabricated instances share the same password. Protocol-based schemes use more complex techniques for authentication [18]- [21]. Specifically, a trusted server is used to manage the multi-stage authentication between the user and the device, but has the drawback of requiring network availability.…”