A lightweight data security model is of much importance in view of security and privacy of data in several networks (e.g., fog networks) where available computing units at edge nodes are often constrained with low computing capacity and limited storage/availability of energy. To facilitate lightweight data security at such constrained scenarios, cellular automata (CA)-based lightweight data security model is presented in this chapter to enable low-cost physical implementation. For this reason, a detailed investigation is presented in this chapter to explore the potential capabilities of CA-based scheme towards the design of lightweight data security model. Further, a comparison among several existing lightweight data security models ensure the effectiveness for proposed CA-based lightweight data security model. Thus, application suitability in view of fog networks is explored for the proposed CA-based model which has further potential for easy training of a reservoir of computers towards uses in IoT (internet of things)-based multiple industry applications.
Top1. Introduction
With the advances in computing technologies, the simple computing by a single computing unit has been transformed into today's advanced computing technologies, which are often being performed by several computing devices in a co-operation and may be situated over different locations. Among several advanced computing paradigms, Fog Computing has recently gained attention and popularity among researchers and practitioners for its potential advantages towards modelling of several IoT (Internet of Things)-based real life and industry applications (Mitra et al., 2019). The concept of Fog Computing describes it as a distributed computing where “resources and application services are used in logically efficient places from data source to cloud” (Saha et al., 2019a). As presented in (Saha et al., 2019b), it is found that the MEC (Mobile Edge computing, is used to process task over cloud by running a cloud server), MCC (Mobile Cloud computing, where mobile Cloud Applications store data in cloud (i.e., outer side with reference to the mobile device) after data processing) etc are similar concepts with reference to the concept of Fog Computing. A typical workflow between the Fog, Cloud and Computing Devices is presented in Figure 1 (Figure 1 is inspired from (Saha et al., 2019a)).
Figure 1. Typical workflow between the Fog, Cloud and Computing Devices
As mentioned earlier, several researchers at present days have focused on the enhanced design of Fog computing network to enhance the efficiency and service quality. In our studies we found research have focused on several aspects in Fog networks e.g., on the design of an enhanced Fog network architecture (Bonomi et al., 2012; Khakimov et al., 2012; and Saha et al., 2019b), data security (Suo et al., 2012; Ni et al., 2017; and Mitra et al., 2019), design of energy optimized algorithm (Saha et al., 2019a) etc. Further the role of Fog computing in IoT-based applications was examined in (Bonomi et al., 2012) by Bonomi et al. Though several research areas are present, we focused on data security in Fog network.
In our studies we found that conventional data security (authentication and cryptography) approaches are not well suited for IoT/ Fog computing-based applications. Conventional data security approaches require a high power of computation and hence requires a high amount of power and high volume of memory (Mitra et al., 2019). Previously with the same focus, uses of lightweight cryptography in lieu of existing conventional cryptography were described in (Eisenbarth et al., 2007; and Buchanan et al., 2017), several lightweight authentications were presented in (Vajda et al., 2003; Gilbart et al., 2005) towards IoT/Fog computing-based applications.
At the same time, Green Computing and sustainability is one of major concern among researchers, which primarily focus on the minimization of energy consumption. Several approaches towards Green Computing in Cloud may be found in (Mitra et al., 2017b; Mitra 2019). To facilitate, energy efficient physical modelling researchers have presented Cellular Automata (CA) (Chaudhuri et al., 1997) based design (Nandi et al., 1994; Chaudhuri et al., 1997; Vajda et al. 2003; Gilbart et al., 2005; Mitra et al., 2017b; and Mitra, 2019) towards several engineering applications in Distributed Computing environment. In our studies, we found that CA-based several lightweight cryptosystems were investigated in (Nandi et al., 1994; Ojha et al., 2009; Tripathy et al., 2009; and Mitra et al. 2017a), further several lightweight CA-based authentications were investigated and presented in (Shemaili et al., 2014; and Mariot et al., 2019). A brief discussion on CA is presented next.
CA is a dynamic modelling tool which advance over discrete time and discrete space. Elementary CA (ECA) are known as simple CA configuration consisting three cells in single-dimension, three neighbourhood (left and right neighbour and self-cell) at fixed / periodic boundary condition (Chaudhuri et al., 1997). The next state of a cell in ECA configuration at time
is determined by a function (known as CA rule, also known as Wolfram CA rules, total 256 rules),
where at time
,
indicates value at left cell,
indicates value at self-cell,
value at right cell and
indicates the value of a cell at time
(Chaudhuri et al., 1997). A typical ECA configuration is presented in Figure 2.