Cryptographic Performance and Energy Efficiency on SimpleLink™ CC13x2/CC26x2 Wireless MCUs

This application report describes the benefits of cryptographic acceleration and provides performance and energy consumption measurements of on-chip cryptographic accelerators integrated in the SimpleLink CC13x2/CC26x2 family of wireless microcontrollers (MCUs). It also benchmarks these measurements against Arm®Cortex®-M4F software-based implementations of cryptographic operations. This document also describes device power management and TI driver concepts to consider for enabling efficient usage of SimpleLink cryptographic drivers.

Trademarks

SimpleLink, LaunchPad are registered trademarks of Texas Instruments.

Arm, Cortex are registered trademarks of Arm Limited.

Bluetooth is a registered trademark of Bluetooth SIG, Inc.

All other trademarks are the property of their respective owners.

1 Abbreviations and Acronyms

AES	Advanced Encryption Standard
CBC	Cipher Block Chaining
CCM	Counter with CBC-MAC
CPU	Central Processing Unit
CSPRNG	Cryptographically Secure Pseudo-Random Number Generator
CTR	Counter Mode of Operation
DRBG	Deterministic Random Bit Generator
ECB	Electronic Code Book
ECC	Elliptic Curve Cryptography
ECDH	Elliptic Curve Diffie-Helfman
ECDSA	Elliptic Curve Digital Signature Algorithm
ECJPAKE	Elliptic Curve Password Authenticated Key Exchange by Juggling
GCM	Gallois Counter Mode
HW	Hardware
HWI	Hardware Interrupt
ISR	Interrupt Service Routine
LFSR	Linear Feedback Shift Register
MAC	Message Authentication Code
MCU	Microcontroller Unit
MIPS	Millions of Instructions Per Second
NIST	National Institute of Standards and Technology
OS	Operating System
PKA	Public Key Accelerator
PRNG	Pseudo-Random Number Generator
RAM	Random Access Memory
SHA	Secure Hash Algorithm
SPI	Serial Peripheral Interface
SRAM	Static Random Access Memory
SW	Software
SWI	Software Interrupt
TLS	Transport Layer Security
TRNG	True Random Number Generator
UART	Universal Asynchronous Receiver/Transmitter

2 Introduction

Security in network connected systems has become increasingly critical. There is increased motivation for compromising network connected products over closed systems due to factors such as: a larger attack surface through remote and local access and the potential for successful attacks to leverage compromising nodes in the field into impacting large groups of persons or organizations in a significant way. Recent product exploits have demonstrated that security in network connected ecosystems is only as strong as the weakest link; thus, pushing for increased security, not just in gateways and servers but also in the end nodes, for strengthening the security of the entire system.

Security solutions are implemented across network components to mitigate security risks. Cryptography is a foundational component used in these security solutions that aid in protection of security assets (code, data, or keys) from adversaries. Cryptography is used to provide secrecy and integrity for data and enables both authentication and anonymity to entities involved in communication. Modern cryptography is heavily based on mathematical theory and computer science practice and can impact the performance and energy consumption of resource-constrained embedded systems. Resource-constrained embedded systems typically refer to microcontrollers with limited hardware resources like CPU MIPS (millions of instructions per second) or memory. They are often powered from batteries and have specific battery-life requirements to meet.

The following sections discuss the benefits of cryptographic acceleration, followed by basic concepts of device power management and TI drivers for SimpleLink™ MCUs in the context of using cryptographic APIs for the development of security-focused applications. Next, the document covers benchmarking results for various cryptographic accelerators integrated in the SimpleLink CC13x2/CC26x2 family of wireless microcontrollers (MCUs). The benchmarking results show cryptographic performance and energy consumption using on-chip cryptographic accelerators compared to an Arm®Cortex®-M4F software-based implementation. In this benchmarking effort, we have used Arm mbed TLS software cryptographic functions to compare performance with on-chip cryptographic accelerators.