Abstract
Edge computing is an emerging paradigm in which speed is enhanced by deploying software programs on embedded systems in the vicinity of data collection. Edge computing is considered to be a vital concept for realizing the far-fetched dream of pervasively interconnecting millions of devices through Internet of things (IoT). In this paper, an approximate convolutional layer is proposed that is based on analysis of dataset before application to training and inference. The technique uses the similarity of the image pixels at the edges of the images in the dataset. At the same time, a novel approximate 8-bit fixed point multiplier is proposed that increases the energy efficiency without compromising much accuracy. The comparison results of exact and approximate CNN prove that the approximate CNN has 5.5% (about 8.8 million) less MAC operations as compared to the exact CNN with a minimal accuracy loss of 3.3%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Zanella A, Bui N, Castellani A, Vangelista L, Zorzi M (2014) Internet of things for smart cities. IEEE Internet Things J 1(1):22–32
Misra NN, Dixit Y, Al-Mallahi A, Bhullar MS, Upadhyay R, Martynenko A (2022) IoT, big data, and artificial intelligence in agriculture and food industry. IEEE Internet Things J 9(9):6305–6324
Han J, Orshansky M (2013) Approximate computing: an emerging paradigm for energy-efficient design. In: 2013 18th IEEE European test symposium (ETS). IEEE, pp 1–6
Irtija N, Anagnostopoulos I, Zervakis G, Tsiropoulou EE, Amrouch H, Henkel J (2021) Energy efficient edge computing enabled by satisfaction games and approximate computing. IEEE Trans Green Commun Network
Gao C, Rios-Navarro A, Chen X, Liu S, Delbruck T (2020) EdgeDRNN: recurrent neural network accelerator for edge inference. IEEE J Emerg Selected Top Circuits Syst 10(4):419–432
Yu F, Cui L, Wang P, Han C, Huang R, Huang X (2020) Easiedge: A novel global deep neural networks pruning method for efficient edge computing. IEEE Internet Things J 8(3):1259–1271
Castro-Godínez J, Hernández-Araya D, Shafique M, Henkel J (2020) Approximate acceleration for CNN-based applications on IoT edge devices. In: IEEE 11th Latin American symposium on circuits and systems (LASCAS). IEEE, pp 1–4
Lee J, Kang S, Lee J, Shin D, Han D, Yoo H (2020) The hardware and algorithm co-design for energy-efficient DNN processor on edge/mobile devices. IEEE Trans Circuits Syst I: Regular Papers 67(10):3458–3470
Guo K, Han S, Yao S, Wang Y, Xie Y, Yang H (2017) Software-hardware codesign for efficient neural network acceleration. IEEE Micro 37(2):18–25
Zhang W, Fu K, Zheng N, Chen Q, Li C, Zheng W, Guo M (2021) CHARM: collaborative host and accelerator resource management for GPU datacenters. In: 2021 IEEE 39th international conference on computer design (ICCD). IEEE, pp 307–315
Takahashi R, Matsubara T, Uehara K (2019) Data augmentation using random image cropping and patching for deep CNNs. IEEE Trans Circuits Syst Video Technol 30(9):2917–2931
Uchôa V, Aires K, Veras R, Paiva A, Britto L (2020) Data augmentation for face recognition with CNN transfer learning. In: International conference on systems, signals and image processing (IWSSIP). IEEE, pp 143–148
Guesmi A, Alouani I, Khasawneh K, Baklouti M, Frikha T, Abid M, Abu-Ghazaleh N (2021) Defensive approximation: securing CNNs using approximate computing. In: Proceedings of the 26th ACM international conference on architectural support for programming languages and operating systems, pp 990–1003
Mrazek V, Vasícek Z, Sekanina L, Hanif MA, Shafique M (2020) ALWANN: automatic layer-wise approximation of deep neural network accelerators without retraining. In: IEEE/ACM international conference on computer-aided design (ICCAD). IEEE, pp 1–8
Shi W, Dustdar S (2016) The promise of edge computing. Computer 49(5):78–81
Sulieman N, Celsi L, Li W, Zomaya A, Villari M (2022) Edge-oriented computing: a survey on research and use cases. Energies 15(2):452
Wu Y, Chen D, Fang J (2001) Better exploration of region-level value locality with integrated computation reuse and value prediction. In: Proceedings 28th annual international symposium on computer architecture. IEEE, pp 98–108
He X, Jiang S, Lu W, Yan G, Han Y, Li X (2016) Exploiting the potential of computation reuse through approximate computing. IEEE Trans Multi-Scale Comput Syst 3(3):152–165
Aizaz Z, Khare K (2022) Area and power efficient truncated booth multipliers using approximate carry based error compensation. IEEE Trans Circuits Syst II Express Briefs
Kim M, A Del Barrio A, Oliveira L, Hermida R, Bagherzadeh N (2018) Efficient Mitchell’s approximate log multipliers for convolutional neural networks. IEEE Trans Comput 68(5):660–675
Ha M, Lee S (2018) Multipliers with approximate 4–2 compressors and error recovery modules. IEEE Embed Syst Lett 10(1):6–9
Aizaz Z, Khare K (2022) State-of-art analysis of multiplier designs for image processing and convolutional neural network applications. In: International conference for advancement in technology (ICONAT), pp 1–11
Aizaz Z, Khare K, Tirmizi A (2022) Efficient approximate multipliers for neural network applications. In: Computational intelligence in data mining. Smart innovation, systems and technologies, vol 281. Springer, Singapore
Sarwar S, Venkataramani S, Ankit A, Raghunathan A, Roy K (2018) Energy-efficient neural computing with approximate multipliers. ACM J Emerg Technol Comput Syst (JETC) 14(2):1–23
Venkatachalam S, Ko S (2017) Design of power and area efficient approximate multipliers. IEEE Trans Very Large Scale Integr (VLSI) Syst 25(5):1782–1786
LeCun Y (1998) The MNIST database of handwritten digits. http://yann.lecun.com/exdb/mnist/
Golovko V, Egor M, Brich A, Sachenko A (2016) A shallow convolutional neural network for accurate handwritten digits classification. In: International conference on pattern recognition and information processing. Springer, pp 77–85
Aizaz Z, Khare K (2022) Energy efficient approximate booth multipliers using compact error compensation circuit for mitigation of truncation error. Int J Circ Theor Appl 50(6)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Aizaz, Z., Khare, K., Tirmizi, A. (2023). Exploiting Pixel Redundancy and Approximate Computing for Efficient Hardware–Software Co-design of CNN on IoT Edge Devices. In: Noor, A., Saroha, K., Pricop, E., Sen, A., Trivedi, G. (eds) Proceedings of Third Emerging Trends and Technologies on Intelligent Systems. ETTIS 2023. Lecture Notes in Networks and Systems, vol 730. Springer, Singapore. https://doi.org/10.1007/978-981-99-3963-3_43
Download citation
DOI: https://doi.org/10.1007/978-981-99-3963-3_43
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-3962-6
Online ISBN: 978-981-99-3963-3
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)