CoChat
Sign Up
Cited by 1Cryptographic Implementations and SecuritySecurity in Wireless Sensor NetworksArduino and IoT Applications

Comparative performance analysis of hardware and software AES encryption implementation in WSN

Open access
Vracar, Ljubomir, Stojanovic, Milan·2025·Facta universitatis - series: Electronics and Energetics
DOI 10.2298/fuee2502221v
Open this paper in CoChat — free accountOpen PDF

Abstract

Wireless Sensor Networks (WSNs) have been among the most researched areas in the last ten or twenty years. Different sensor technologies and wireless communication contributed to the progress in this field, but two features of these systems are still insufficiently developed. These are data security and energy consumption. Data is particularly vulnerable during transmission, whether it is communication between nodes or between a base station and a server. One of the most promising solutions for this problem is data encryption. However, data encryption negatively affects sensor nodes' energy consumption, so there is a compromise between data security and energy consumption. Symmetric encryption algorithms are more effective than asymmetric for these applications. Advanced Encryption Standard (AES) is the most frequently used symmetric algorithm. This paper analyzes the performance of hardware and software implementation of the AES encryption algorithm on the sensor node. The hardware implementation is based on the integrated circuit ATECC608A, while the microcontroller firmware includes the software implementation. The analyzed parameters include execution time, energy consumption, and memory usage, for four working frequencies (8 MHz, 16 MHz, 32 MHz, and 64 MHz), and different supply voltages (2.6 V, 3.3 V, and 5 V). The results show that the hardware implementation works faster for the lower working frequency (up to 12 MHz). The execution time of software implementation is reduced approximately 50 % with doubled frequency, while the reduction is less, and inconstant in hardware implementation. The hardware implementation consumes more energy that software, and the difference rises with the working frequency. These differences are 20 %, 40 %, 60 %, and 75 % at the same frequencies considered for execution time analyzing. The hardware implementation occupies 4.58 % of the memory of the microcontroller (PIC18F45K22), while 10.13 % is needed for the software implementation. These results can provide a good starting point for choosing an appropriate encryption implementation method.

AI snapshot

Written from the paper’s full text.

TL;DR

This paper compares hardware and software AES encryption on a WSN sensor node using a PIC18F45K22 microcontroller and the ATECC608A crypto IC. Hardware AES is faster only at lower operating frequencies (up to about 12 MHz), while software AES uses less energy but more program memory.

Key findings

  • Hardware AES has lower execution time at lower frequencies, approximately up to 12 MHz; software AES is faster at higher frequencies. Doubling the microcontroller frequency reduces software execution time by almost 50%, while hardware execution time reductions are smaller and inconstant: 25%, 15%, and 13% for 8→16 MHz, 16→32 MHz, and 32→64 MHz, respectively.
  • Hardware AES consumes more energy than software AES at every tested frequency and supply voltage. Table 2 reports hardware-vs-software energy differences of +16%, +17%, +27% at 8 MHz for 2.6 V, 3.3 V, and 5 V; +40%, +39%, +45% at 16 MHz; +60%, +61%, +61% at 32 MHz; and +75%, +75%, +78% at 64 MHz.
  • Hardware AES occupies 4.58% of the microcontroller’s programmable memory, while software AES occupies 10.13%. The selected PIC18F45K22 has 4 kB of user program memory.
  • The hardware encryption process includes ATECC608A wake-up, which takes about 1500 µs, and encryption, which takes about 2030 µs; these fixed times do not depend on the microcontroller working frequency.

Methods

  • The sensor node uses a PIC18F45K22 microcontroller. Hardware AES is implemented with an external ATECC608A integrated circuit over I²C; software AES is implemented inside the microcontroller firmware.
  • Execution time was measured with an oscilloscope using GPIO pulses marking operation start/end. Current was measured with a 100 mΩ shunt resistor and INA212, and energy was calculated from current, voltage, and time.
  • Measurements were taken at supply voltages of 2.6 V, 3.3 V, and 5 V and microcontroller working frequencies of 8 MHz, 16 MHz, 32 MHz, and 64 MHz. The microcontroller was in sleep mode before and after encryption.

Limitations

  • The paper does not explicitly state limitations.
  • Evident from the text: the results are tied to one specific hardware/software setup—PIC18F45K22 plus ATECC608A—and to AES only; the authors state future work will test different microcontrollers and integrated circuits.
  • Evident from the text: the study measures encryption execution time, energy, and memory only, not radio transmission energy; the authors state future research will analyze various radio modules and transmission energy.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.2298/fuee2502221v

Cites

42 of 42 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

2 of 6 cited sources checked · 1 supported · 0 flagged · accuracy not yet measured

  • Findingreasoning

    The execution time of AES encryption decreases as the microcontroller working frequency increases.

    “The execution time decreases with the increase of the microcontroller working frequency.”
    Quote in the paperSubstantial claim
  • Findingown data

    The energy consumption differences between the hardware and software AES implementations are 20%, 40%, 60%, and 75% at the frequencies considered for execution time analysis.

    “These differences are 20 %, 40 %, 60 %, and 75 % at the same frequencies considered for execution time analyzing.”
    Quote in the paperSubstantial claim
  • Findingown data

    The reduction in execution time with increasing frequency is greater for software AES encryption than for hardware AES encryption.

    “The reduction of the execution time is more intensive for software than for hardware encryption.”
    Quote in the paperSubstantial claim
  • Findingown data

    For the software AES implementation, doubling the working frequency reduces execution time by approximately 50%.

    “The execution time of software implementation is reduced approximately 50 % with doubled frequency, while the reduction is less, and inconstant in hardware implementation.”
    Quote in the paperSubstantial claim
  • Findingown data

    The hardware AES implementation occupies 4.58% of the memory of the microcontroller PIC18F45K22.

    “The hardware implementation occupies 4.58 % of the memory of the microcontroller (PIC18F45K22), while 10.13 % is needed for the software implementation.”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim. 11
  • May combine two claims. 12
  • The quote only partly states this claim. 2

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Symmetric encryption algorithms are low-cost and high-efficiency.

    “Symmetric encryption algorithms are low-cost and high-efficiency.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [9] P. P. Santoso, E. Rilvani, A. B. Trisnawan, K. Adiyarta, D. Napitupulu, T. Sutabri and R. Rahim, "Systematic Literature Review: Comparison Study of Symmetric Key and Asymmetric Key Algorithm", In Proceedings of the 2nd Nommensen International Conference on Technology and Engineering, Medan, Indonesia, 2018, vol. 420, p. 012111. source

      The Figure 1 passage says symmetric algorithms operate faster than asymmetric algorithms, and Results and Discussion says they perform well on speed and power consumption. These support high efficiency, but the passages do not establish that symmetric algorithms are low-cost. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Asymmetric encryption algorithms are complex and take more execution time than symmetric algorithms, but they deliver a higher level of security.

    “Asymmetric algorithms are complex and take more execution time, but they deliver a higher level of security.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [9] P. P. Santoso, E. Rilvani, A. B. Trisnawan, K. Adiyarta, D. Napitupulu, T. Sutabri and R. Rahim, "Systematic Literature Review: Comparison Study of Symmetric Key and Asymmetric Key Algorithm", In Proceedings of the 2nd Nommensen International Conference on Technology and Engineering, Medan, Indonesia, 2018, vol. 420, p. 012111. source

  • May combine two claims.

    Findingnone given

    For the hardware AES implementation, the reduction in execution time with doubled frequency is less and inconstant.

    “The execution time of software implementation is reduced approximately 50 % with doubled frequency, while the reduction is less, and inconstant in hardware implementation.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingreasoning

    For hardware AES encryption, the execution time reduction is 25%, 15%, and 13% for frequency changes from 8 to 16 MHz, 16 to 32 MHz, and 32 to 64 MHz, respectively.

    “A similar trend is noted for hardware encryption, but the reduction of execution time is smaller: 25 %, 15 %, and 13 % for frequency changes 8 MHz to 16 MHz, 16 MHz to 32 MHz, and 32 MHz to 64 MHz, respectively.”
    Quote in the paperSubstantial claim

  • The quote only partly states this claim.

    Findingreasoning

    Software AES encryption energy consumption decreases slightly as working frequency increases.

    “Therefore, the energy consumption decreases slightly if the software encryption is applied.”
    Quote in the paperSubstantial claim

  • The quote only partly states this claim.

    Findingown data

    Software AES encryption is faster than hardware AES encryption at working frequencies above 12 MHz.

    “while the software encryption is faster at higher frequencies.”
    • Also stated in: § 5. RESULTS · plain
      “while the software encryption operates faster at higher values of working frequencies.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Hardware AES implementation occupies 4.58% of programmable memory, while software AES implementation occupies 10.13%.

    “For the hardware and software implementation, 4.58 % and 10.13 % of programmable memory are occupied, respectively.”
    • Also stated in: § 5. RESULTS · plain
      “the hardware implementation occupies less programmable memory”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    After the National Security Agency approved it, AES has become a recommended security standard of data encryption.

    “After the National Security Agency approved it, AES has become a recommended security standard of data encryption [37].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [37] M. Dworkin, E. Barker, J. Nechvatal, J. Foti, L. Bassham, E. Roback and J. Dray, "Advanced Encryption Standard (AES)", Federal Inf. Process. Stds. (NIST FIPS), Nov. 2001.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisereasoning

    Even though symmetric encryption algorithms provide a lower security level than asymmetric algorithms, they are usually used in WSNs because of resource limitations.

    “Even symmetric encryption algorithms provide a lower security level than asymmetric they are usually used in WSNs because of resource limitations.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    AES is a high-speed encryption algorithm and provides more security than other symmetric encryption algorithms.

    “AES is high speed encryption algorithm, and it provides more security that others symmetric encryption algorithms.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [10] C. Zhang, Y. Jia, L. Zhu and Z. Zhang, "Research on Simple Power Consumption Based on AES Algorithm", In Proceedings of the 2023 IEEE 2nd International Conference on Electrical Engineering, Big Data and Algorithms (EEBDA), Changchun, China, 2023, pp. 1883-1886. source
    • cites [11] E. S. Selvapriya and L. Suganthi, "Design and Implementation of Low Power Advanced Encryption Standard Cryptocore Utilizing Dynamic Pipelined Asynchronous Model", Integration, vol. 93, p. 102057, 2023. source

  • May combine two claims.

    Interpretationown data

    Hardware AES's lower memory usage is significant for sensor nodes built on microcontrollers with limited memory space, or sensor nodes that contain computationally demanding algorithms for data analysis.

    “On the other hand, the hardware implementation occupies less programmable memory, which is significant for sensor nodes built on microcontrollers with limited memory space, or sensor nodes that contain computationally demanding algorithms for data analysis.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationreasoning

    The experimental setup for current measurement based on INA212 operates at the same voltage level as a sensor node, so it can be made using a microcontroller instead of an oscilloscope and installed in a sensor node, providing real-time consumption measurement.

    “The experimental setup for current measurement based on INA212 operates at the same voltage level as a sensor node.So, it can be made using a microcontroller instead of an oscilloscope and installed in a sensor node, providing real-time consumption measurement.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationown data

    Software AES encryption should be applied inside the sensor node if energy saving is the primary goal, such as sensor nodes powered by battery or energy-harvesting power supply.

    “This encryption process should be applied inside the sensor node if energy saving is the primary goal, such as sensor nodes powered by battery or energy-harvesting power supply.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationnone given

    The only constraint for hardware AES implementation is a microcontroller inside a sensor node equipped with an I2C interface, and enough programmable memory for software implementation.

    “The only constraint for hardware implementation is a microcontroller inside a sensor node equipped with an I 2 C interface, and enough programmable memory for software implementation.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    The microcontroller is in sleep mode before and after the encryption process to minimize energy consumption during the inactive state.

    “The microcontroller is in sleep mode before and after the encryption process to minimize energy consumption during the inactive state (no measurement or sending data).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    The current measurement circuit based on shunt resistor and INA212 provides high precision and has a low impact on load operating voltage.

    “This methods provide the high precision and it has a low impact on load operating voltage.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The increase in power consumption for hardware AES encryption is mainly due to increased current intensity, which is more significant at higher frequencies than the reduction of execution time.

    “The main contribution to the increased power consumption is the increase of the current intensity, which is more significant at higher frequencies than the reduction of the execution time.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    All unused microcontroller pins were configured as output and set to low voltage levels to prevent unnecessary energy consumption.

    “All microcontroller unused pins are configured as output, and set to low voltage levels to prevent unnecessary energy consumption [41].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [41] B. Ivey, Low-Power Design Guide, Microchip Technology Inc., 2011.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

18 of 18

  • The paper relies on this without support.

    Traditional security solutions cannot be applied to WSNs because sensor nodes have significant resource constraints.

    “Traditional security solutions require a lot of computing energy and large memory.Therefore, they cannot be applied to WSNs because of significant resource constraints.”

  • The paper relies on this without support.

    Sensor nodes are usually deployed unattended and battery-powered, and battery recharging or replacement is not easy.

    “Sensor nodes are usually deployed in unattended environments without access to an electrical grid and powered by batteries.The battery recharging or replacement is not easy, so the sensor node's energy consumption should be reduced maximally.”

  • The paper relies on this without support.

    The performance results obtained on the PIC18F45K22 and ATECC608A platform are representative enough to guide selection of AES implementation in WSNs.

    “These results can provide a good starting point for choosing an appropriate encryption implementation method.”

  • The paper relies on this without support.

    AES is not energy-demanding and provides a high level of security.

    “It is not energy-demanding and provides a high level of security.”

  • The paper relies on this without support.

    The power supply voltages 3.3 V and 5 V are the most common in battery-powered sensor nodes, and most radio transceiver modules operate at these voltages.

    “Two other power supply voltages considered in this paper are 3.3 V and 5 V because they are the most common in battery-powered sensor nodes, and most radio transceiver modules operate at these voltages.”
5 of 5

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Current-measurement step specifies the GPIO oscilloscope channel

    Minor

    The setup connects the current-sensing output to CH1 and a GPIO pin to CH2, but step 6 says current is measured on channel 2.

1 of 1

Cites

Flagged first, then by citations

Research works printed but not linked: 2

  1. A. B. Karuppiah, Dr. S. Rajaram, "Energy Efficient Encryption Algorithm for Wireless Sensor Network", Int. J. Eng. Res. Technol. (IJERT), vol. 1, pp. 1-7, May 2012.
  2. Z. Nakutis, "Embedded Systems Power Consumption Measurement Methods Overview", Matavimai, vol. 2, no. 44, pp. 29-35, 200

Cited by

Sorted by citations

Authors
Vracar, L., & Stojanovic, M.
Affiliations
University of Nis (RS)
ORCID
Venue
Facta universitatis - series: Electronics and Energetics
Publisher
National Library of Serbia
Published
2025
Volume
38
Issue
2
Pages
221-237
ISSN
  • 0353-3670 (ISSN-L)
  • 2217-5997
DOI
10.2298/fuee2502221v
OpenAlex
W4415430425
Semantic Scholar
21b474cf4e1d647fae6a0d2947bc641e0c8c313e
Cited by
1
References
39
Citations by year
2026: 1
FWCI
0.43
Type
journal article
Language
en
Open access
Licence
Locations
Facta universitatis - series Electronics and Energetics (publishedVersion, cc-by-nc-nd)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Cryptographic Implementations and Security
  • Physical Sciences › Computer Science › Computer Networks and Communications › Security in Wireless Sensor Networks
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Arduino and IoT Applications
Keywords
Advanced Encryption Standard, AES, wireless sensor networks, execution time, energy consumption, memory usage, symmetric encryption, data security, supply voltage

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-23.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-23.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review