Research article Special Issues

Enhancing the robustness of block ciphers through a graphical S-box evolution scheme for secure multimedia applications

  • Received: 01 September 2024 Revised: 30 November 2024 Accepted: 03 December 2024 Published: 18 December 2024
  • MSC : 20D35, 94A60

  • Block ciphers are essential for the secure exchange of data and communication, as they are one of the primary components of network security systems. Modern-day block ciphers are most significantly reliant on substitution-boxes (S-boxes). In essence, the security of these cryptosystems is contingent upon the quality of the S-box that is implemented. Robustness and assurance of the security competency necessary to block ciphers are provided by the cryptographically strong S-boxes. A novel coset graph-based algebraic method was proposed to evolve a robust and efficient S-box in order to address the challenges of strong S-box generation. To begin, the vertices of coset graphs for two Galois fields and a bijective function were employed to generate an initial S-box of sufficient cryptographic strength. Afterwards, a permutation group of large order enhances the robustness of the initial S-box, ensuring its resistance against various cryptanalytic attacks. The proposed method's efficacy was verified by comparing the attributes of our S-box with those of S-boxes that have been recently investigated. Furthermore, the proposed S-box was used for image encryption. The outcome of the majority logic criterion (MLC) criteria, differential analysis, and histogram test demonstrates the suitability of the proposed S-box for secure multimedia applications in the results.

    Citation: Abdul Razaq, Muhammad Mahboob Ahsan, Hanan Alolaiyan, Musheer Ahmad, Qin Xin. Enhancing the robustness of block ciphers through a graphical S-box evolution scheme for secure multimedia applications[J]. AIMS Mathematics, 2024, 9(12): 35377-35400. doi: 10.3934/math.20241681

    Related Papers:

    [1] Dina Abuzaid, Samer Al-Ghour . Supra soft Omega-open sets and supra soft Omega-regularity. AIMS Mathematics, 2025, 10(3): 6636-6651. doi: 10.3934/math.2025303
    [2] Dina Abuzaid, Samer Al-Ghour . Correction: Supra soft Omega-open sets and supra soft Omega-regularity. AIMS Mathematics, 2025, 10(5): 10624-10625. doi: 10.3934/math.2025483
    [3] Tareq M. Al-shami, Salem Saleh, Alaa M. Abd El-latif, Abdelwaheb Mhemdi . Novel categories of spaces in the frame of fuzzy soft topologies. AIMS Mathematics, 2024, 9(3): 6305-6320. doi: 10.3934/math.2024307
    [4] Tareq M. Al-shami, El-Sayed A. Abo-Tabl . Soft α-separation axioms and α-fixed soft points. AIMS Mathematics, 2021, 6(6): 5675-5694. doi: 10.3934/math.2021335
    [5] Baravan A. Asaad, Tareq M. Al-shami, Abdelwaheb Mhemdi . Bioperators on soft topological spaces. AIMS Mathematics, 2021, 6(11): 12471-12490. doi: 10.3934/math.2021720
    [6] Orhan Göçür . Amply soft set and its topologies: AS and PAS topologies. AIMS Mathematics, 2021, 6(4): 3121-3141. doi: 10.3934/math.2021189
    [7] Fahad Alsharari, Ahmed O. M. Abubaker, Islam M. Taha . On r-fuzzy soft γ-open sets and fuzzy soft γ-continuous functions with some applications. AIMS Mathematics, 2025, 10(3): 5285-5306. doi: 10.3934/math.2025244
    [8] Ibtesam Alshammari, Islam M. Taha . On fuzzy soft β-continuity and β-irresoluteness: some new results. AIMS Mathematics, 2024, 9(5): 11304-11319. doi: 10.3934/math.2024554
    [9] Tareq M. Al-shami, Zanyar A. Ameen, A. A. Azzam, Mohammed E. El-Shafei . Soft separation axioms via soft topological operators. AIMS Mathematics, 2022, 7(8): 15107-15119. doi: 10.3934/math.2022828
    [10] Rehab Alharbi, S. E. Abbas, E. El-Sanowsy, H. M. Khiamy, Ismail Ibedou . Soft closure spaces via soft ideals. AIMS Mathematics, 2024, 9(3): 6379-6410. doi: 10.3934/math.2024311
  • Block ciphers are essential for the secure exchange of data and communication, as they are one of the primary components of network security systems. Modern-day block ciphers are most significantly reliant on substitution-boxes (S-boxes). In essence, the security of these cryptosystems is contingent upon the quality of the S-box that is implemented. Robustness and assurance of the security competency necessary to block ciphers are provided by the cryptographically strong S-boxes. A novel coset graph-based algebraic method was proposed to evolve a robust and efficient S-box in order to address the challenges of strong S-box generation. To begin, the vertices of coset graphs for two Galois fields and a bijective function were employed to generate an initial S-box of sufficient cryptographic strength. Afterwards, a permutation group of large order enhances the robustness of the initial S-box, ensuring its resistance against various cryptanalytic attacks. The proposed method's efficacy was verified by comparing the attributes of our S-box with those of S-boxes that have been recently investigated. Furthermore, the proposed S-box was used for image encryption. The outcome of the majority logic criterion (MLC) criteria, differential analysis, and histogram test demonstrates the suitability of the proposed S-box for secure multimedia applications in the results.



    The need for theories that cope with uncertainty emerges from daily experiences with complicated challenges requiring ambiguous facts. Molodstov's [1] soft set is a contemporary mathematical approach to coping with these difficulties. Soft collection logic is founded on the parameterization principle, which argues that complex things must be seen from several perspectives, with each aspect providing only a partial and approximate representation of the full item. Molodstov [1] was a pioneer in the application of soft sets in a variety of domains, emphasizing their advantages over probability theory and fuzzy set theory, which deal with ambiguity or uncertainty.

    Following that, Maji et al. [2] began researching soft set operations such as soft unions and soft intersections. To overcome the shortcomings of these operations, Ali et al. [3] created and showed new operations such as limited union, intersection, and complement of a soft set. Babitha and Sunil [4] investigated numerous aspects of linkages and functions in a soft setting. Qin and Hong [5] developed novel kinds of soft equal relations and showed some algebraic properties of them. Their pioneering work paved the way for subsequent papers (for more detail, see [6,7] and the references listed therein). Soft set theory has lately been a popular method among academics for dealing with uncertainty in a wide range of fields, including information theory [8], computer sciences [9], engineering [10], and medical sciences [11].

    Soft topology was introduced by Shabir and Naz in [12]. Since then, many soft topological notions, including soft separation axioms [13,14,15,16], soft covering axioms [17,18,19,20,21,22], soft connectedness [23,24,25,26], and different weak and strong types of soft continuity, have been developed and investigated in recent years. The equivalence between the enriched and extended soft topologies was discussed in [27].

    Separation axioms provide a way to study certain properties of compact and Lindelof spaces, as well as a way to categorize spaces and mappings into distinct families. As a result, topological scholars who presented various kinds of soft separation axioms became interested in soft separation axioms. Generally speaking, they can be separated into two classes: Soft points and ordinary points, based on the subjects being studied. While the authors in [14,15,16,28] examined soft separation axioms using ordinary points, the authors in [13,29,30,31,32,33] and others have applied the concept of soft points. In the present work, we introduce soft ω-almost-regularity, soft ω-semi-regularity, and soft ω-T212 as three novel soft separation axioms.

    This article is organized as follows:

    In Section 1, after the introduction, we provide a few definitions that are relevant to this paper.

    In Section 2, we define soft ω-almost-regularity as a new soft separation axiom that lies between soft regularity and soft almost-regularity. We introduce many characterizations of this type of soft separation axiom. Also, we provide several sufficient conditions establishing the equivalence between this newly introduced axiom and its relevant counterparts. Moreover, we establish that soft ω -almost-regularity is heritable for specific types of soft subspaces. Furthermore, we show that soft ω-almost-regularity is a productive soft property. In addition, we investigated the links between this class of soft topological spaces and its analogs in general topology.

    In Section 3, we define soft ω-semi-regularity and soft ω-T212 as two new soft separation axioms. We show that soft ω-semi-regularity is a weaker form of both soft semi-regularity and soft ω-regularity, and soft ω-T212 lies strictly between soft T212 and soft T2. Also, we provide several sufficient conditions establishing the equivalence between these newly introduced axioms and their relevant counterparts. Moreover, a decomposition theorem for soft regularity through the interplay of soft ω-semi-regularity and soft ω-almost-regularity is obtained. In addition, we investigated the links between these classes of soft topological spaces and their analogs in general topology.

    This paper follows the notions and terminologies as appear in [34,35,36]. Topological spaces and soft topological spaces, respectively, shall be abbreviated as TS and STS.

    The following definitions will be used in the remainder of the paper:

    Definition 1.1. A TS (H,β) is called

    (a) [37] almost-regular (A-R, for simplicity) if for every zH and every NSC(H,β) such that zHN, we find U,Vβ such that zU, NV, and UV=;

    (b) [38] semi-regular (S-R, for simplicity) if RO(H,β) forms a base for β;

    (c) [39] ω-almost-regular (ω-A-R, for simplicity) if for every zH and every NSωC(H,β) such that zHN, we find U,Vβ such that zU, NV, and UV=;

    (d) [39] ω-semi-regular (ω-S-R, for simplicity) if RωO(H,β) forms a base for β.

    Definition 1.2. A STS (H,φ,Σ) is called

    (a) [13] soft T2 if for every two soft points as,btSP(H,Σ), we find K,Wφ such that as˜K, by˜W, and K˜W=0Σ;

    (b) [13] soft regular if for every azSP(H,Σ) and every Kφ such that az˜K, we find Gφ such that az˜G˜Clφ(G)˜K;

    (c) [32] soft T212 if for every two soft points as,btSP(H,Σ), we find K,Wφ such that as˜K, by˜W, and Clφ(K)˜Clφ(W)=0Σ;

    (d) [31] soft almost-regular (soft A-R, for simplicity) if for every rzSP(H,Σ) and every GSC(H,φ,Σ) such that rz˜1ΣG, we find S,Tφ such that rz˜S, G˜T, and S˜T=0Σ.

    (d) [33] soft ω-regular for every azSP(H,Σ) and every Kφ such that az˜K, we find Gφ such that az˜G˜Clφω(G)˜K.

    (e) [22] fully if G(r) for every Gφ{0Σ} and rΣ.

    In this section, we define soft ω-almost-regularity as a new soft separation axiom that lies between soft regularity and soft almost-regularity. We introduce many characterizations of this type of soft separation axiom. Also, we provide several sufficient conditions establishing the equivalence between this newly introduced axiom and its relevant counterparts. Moreover, we establish that soft ω -almost-regularity is heritable for specific types of soft subspaces. Furthermore, we show that soft ω-almost-regularity is a productive soft property. In addition, we investigated the links between this class of soft topological spaces and its analogs in general topology.

    Definition 2.1. An STS (H,φ,Σ) is called soft ω-almost-regular (soft ω-A-R, for simplicity) if for every rzSP(H,Σ) and every GSωC(H,φ,Σ) such that rz˜1ΣG, we find S,Tφ such that rz˜S, G˜T, and S˜T=0Σ.

    Several characterizations of soft ω-almost-regularity are listed in the following theorem.

    Theorem 2.2. The following are equivalent for any STS (H,φ,Σ):

    (1) (H,φ,Σ) is soft ω-A-R.

    (2) For every rzSP(H,Σ) and every KSωO(H,φ,Σ) such that rz˜K, we find Lφ such that rz˜L˜Clφ(L)˜K.

    (3) For every rzSP(H,Σ) and every KSωO(H,φ,Σ) such that rz˜K, we find LSO(H,φ,Σ) such that rz˜L˜Clφ(L)˜K.

    (4) For every rzSP(H,Σ) and every KSωO(H,φ,Σ) such that rz˜K, we find LSωO(H,φ,Σ) such that rz˜L˜Clφ(L)˜K.

    (5) For every rzSP(H,Σ) and every Kφ such that rz˜K, there is LSωO(H,φ,Σ) such that rz˜L˜Clφ(L)˜Intφ(Clφω(K)).

    (6) For every rzSP(H,Σ) and every Kφ such that rz˜K, there is Lφ such that rz˜L˜Clφ(L)˜Intφ(Clφω(K)).

    (7) For every rzSP(H,Σ) and every GSωC(H,φ,Σ) such that rz˜1ΣG, there are S,Tφ such that rz˜S, G˜T, and Clφ(S)˜Clφ(T)=0Σ.

    (8) For every GSωC(H,φ,Σ), G=˜{Clφ(K):Kφ and G˜K}.

    (9) For every GSωC(H,φ,Σ), G=˜{Y:Yφc and G˜Intφ(Y)}.

    (10) For every LSS(H,Σ) and every MSωO(H,φ,Σ) such that L˜M0Σ, there is Kφ such that L˜K0Σ and Clφ(K)˜M.

    (11) For every LSS(H,Σ){0Σ} and every MSωC(H,φ,Σ) such that L˜M=0Σ, there are S,Tφ such that L˜S0Σ and M˜T.

    Proof. (1) (2): Let rzSP(H,Σ) and KSωO(H,φ,Σ) such that rz˜K. Then, rz˜1ΣKSωC(H,φ,Σ) and by (a) there exist L,Tφ such that rz˜L, 1ΣK˜T, and L˜T=0Σ. Thus, rz˜L˜1ΣT˜K with 1ΣTφc, and so rz˜L˜Clφ(L)˜1ΣT˜K. This ends the proof.

    (2) (3): Let rzSP(H,Σ) and KSωO(H,φ,Σ) such that rz˜K. By (2) we find Mφ such that rz˜M˜Clφ(M)˜K. Set L=Intφ(Clφ(M)). Then, LSO(H,φ,Σ). Since L˜Clφ(M)˜K, Clφ(L)˜Clφ(M)˜K. This completes the proof.

    (3) (4): Let rzSP(H,Σ) and KSωO(H,φ,Σ) such that rz˜K. By (3) we find LSO(H,φ,Σ) such that rz˜L˜Clφ(L)˜K. Since LSO(H,φ,Σ), and by Theorem 3 of [36], we have RO(H,φ,Σ)˜RωO(H,φ,Σ), LSωO(H,φ,Σ). This completes the proof.

    (4) (5): Let rzSP(H,Σ) and Kφ such that rz˜K. Since by Theorem 9 of [36] Intφ(Clφω(K))SωO(H,φ,Σ), by (4) there is LSωO(H,φ,Σ) such that rz˜L˜Clφ(L)˜Intφ(Clφω(K)). This completes the proof.

    (5) (6): Let rzSP(H,Σ) and Kφ such that rz˜K. Then, by (5) we find LSωO(H,φ,Σ) such that rz˜L˜Clφ(L)˜Intφ(Clφω(K)). Since by Theorem 3 of [36] we have RωO(H,φ,Σ)˜φ, then Lφ. This completes the proof.

    (6) (7): Let rzSP(H,Σ) and GSωC(H,φ,Σ) such that rz˜1ΣG. Since by Theorem 3 of [36] RωO(H,φ,Σ)˜φ, we have rz˜1ΣGφ. So, by (6) we find Nφ such that rz˜N˜Clφ(N)˜Intφ(Clφω(1ΣG))=1ΣG. Again, by (6) we find Sφ such that rz˜S˜Clφ(S)˜Intφ(Clφω(N))˜Clφ(N)˜1ΣG. Let T=1ΣClφ(N). Then, S,Tφ and rz˜S. Since Clφ(N)˜1ΣG, then G˜1ΣClφ(N)=T.

    Claim. Clφ(S)˜Clφ(T)=0Σ.

    Proof of Claim. Suppose to the contrary that there is ax˜Clφ(S)˜Clφ(T). Since ax˜Clφ(T) and ax˜Clφ(S)˜Intφ(Clφω(N))φ, then Intφ(Clφω(N))˜T0Σ. Since Intφ(Clφω(N))˜Clφ(N), then Clφ(N)˜T=Clφ(N)˜(1ΣClφ(N))0Σ, a contradiction.

    This completes the proof.

    (7) (8): Let GSωC(H,φ,Σ). Then, for each rz˜1ΣG, there exist Srz,Trzφ such that rz˜Srz, G˜Trz, and Clφ(Srz)˜Clφ(Trz)=0Σ. Thus, G˜Trz and rz˜Clφ(Trz).

    Claim. G=˜{Clφ(Trz):rz˜1ΣG}.

    Proof of Claim. For every rz˜1ΣG, we have G˜Trz˜Clφ(Trz), and so G˜˜{Clφ(Trz):rz˜1ΣG}. To show that ˜{Clφ(Trz):rz˜1ΣG}˜G, let rz˜1ΣG. Then, rz˜Clφ(Trz), and thus rz˜˜{Clφ(Trz):rz˜1ΣG}.

    By the above claim, we conclude that G˜˜{Clφ(T):Tφ with G˜T}˜˜{Clφ(Trz):rz˜1ΣG}=G. This completes the proof.

    (8) (9): Obvious.

    (9) (10): Let LSS(H,Σ) and MSωO(H,φ,Σ) such that L˜M0Σ. Pick ax˜L˜M. Since MSωO(H,φ,Σ), 1ΣMSωC(H,φ,Σ), and by (9) 1ΣM=˜{Y:Yφc with 1ΣM˜Intφ(Y)}. Since ax˜M, then ax˜˜{Y:Yφc with 1ΣM˜Intφ(Y)}, and thus we find Yφc such that 1ΣM˜Intφ(Y) and ax˜Y. Let S=1ΣY. Then, Sφ, S˜1ΣIntφ(Y)˜M, and ax˜S˜L. Since 1ΣIntφ(Y)φc and S˜1ΣIntφ(Y)˜M, then Clφ(S)˜M. This completes the proof.

    (10) (11): Let LSS(H,Σ){0Σ} and MSωC(H,φ,Σ) such that L˜M=0Σ. Then, 1ΣMSωO(H,φ,Σ) such that L˜(1ΣM)=L0Σ. Thus, by (10) we find Sφ such that L˜S0Σ and Clφ(S)˜1ΣM. Let T=1ΣClφ(S). Then, Tφ, M˜T, and S˜T=S˜(1ΣClφ(S))=0Σ.

    (11) (1): rzSP(H,Σ) and every GSωC(H,φ,Σ) such that rz˜1ΣG. Then, rz˜G=0Σ, and by (11) there exist S,Tφ such that rz˜S0Σ, G˜T, and S˜T=0Σ. Since rz˜S0Σ, then rz˜S. This ends the proof.

    In Theorems 2.3, 2.4, 2.7, and Corollary 2.8, we discuss the connections between soft almost-regularity and its analog in traditional topological spaces. Also, in Theorems 2.5, 2.6, 2.9, and Corollary 2.10, we discuss the connections between soft ω-almost-regularity and its analog in traditional topological spaces.

    Theorem 2.3. If (H,φ,Σ) is full and soft A-R, then (H,φr) is A-R for all rΣ.

    Proof. Let (H,φ,Σ) be full and soft A-R. Let rΣ. Let zH and let Wφr such that zW. Choose Kφ such that K(r)=W. Since rz˜Kφ, by Theorem 3.4 (ⅳ) of [31], we find Lφ such that rz˜L˜Clφ(L)˜Intφ(Clφ(K)). By Proposition 7 of [12], Clφr(L(r))(Clφ(L))(r). Also, by Theorem 12 (c) of [36], (Intφ(Clφ(K)))(r) =Intφ(Clφ(K(r))). Therefore, we have

    zL(r)Clφr(L(r))(Clφ(L))(r)(Intφ(Clφ(K)))(r)=Intφ(Clφ(K(r)))=Intφr(Clφr(W)).

    Hence, by Theorem 2.2 (d) of [37], it follows that (H,φr) is A-R.

    Theorem 2.4. Let (D,L) be a TS. Then, for any set Σ, (D,C(L),Σ) is soft A-R iff (D,L) is A-R.

    Proof. Necessity. Let (D,C(L),Σ) be soft A-R. Pick rΣ. Since it is clear that (D,C(L),Σ) is full, then by Theorem 2.3, (D,(C(L))r)=(D,L) is A-R.

    Sufficiency. Let (D,L) be A-R. Let rzSP(D,Σ) and let CUC(L) such that rz˜CU. Then, we have zUL. So, by Theorem 2.2 (d) of [37], we find VL such that zVClL(V)IntL(ClL(U)). Thus, we have CVC(L) and rz˜CV˜ClC(L)(CV)= CClL(V)˜CIntL(ClL(U))=IntC(L)(ClC(L)(CU)). Therefore, by Theorem 3.4 (ⅳ) of [39], (D,C(L),Σ) is soft A-R.

    Theorem 2.5. Let (D,L) be a TS. Then, for any set Σ, (D,C(L),Σ) is soft ω-A-R iff (D,L) is ω-A-R.

    Proof. Necessity. Let (D,C(L),Σ) be soft ω-A-R. Let zD and UL such that zU. Pick rΣ. Then, we have rz˜CUC(L). Since (D,C(L),Σ) is soft ω-A-R, by Theorem 2.2 (5) we find VL such that rz˜CV˜ClC(L)(CV)=CClL(V)˜IntC(L)(Cl(C(L))ω(CV))=CIntL(ClLω(U)). Therefore, zVClL(V)IntL(ClLω(U)). This shows that (D,L) is ω-A-R.

    Sufficiency. Let (D,L) be ω-A-R. Let rzSP(D,Σ) and let CUC(L) such that rz˜CU. Then, we have zUL. So, by Theorem 2.1 (e) of [39], we find VL such that zVClL(V)IntL(ClLω(U)). Thus, we have CVC(L) and rz˜CV˜ClC(L)(CV)=CClL(V)˜CIntL(ClLω(U))=IntC(L)(Cl(C(L))ω(CV)). Therefore, (D,C(L),Σ) is soft ω-A-R.

    Theorem 2.6. Let (D,L) be a TS. Then, for any set Σ, (D,C(L),Σ) is soft regular iff (D,L) is regular.

    Proof. Necessity. Let (D,C(L),Σ) be soft regular. Let zD and UL such that zU. Pick rΣ. Then, we have rz˜CUC(L). Since (D,C(L),Σ) is soft regular, we find VL such that rz˜CV˜ClC(L)(CV)=CClL(V)˜CU. Therefore, zVClL(V)U. This shows that (D,L) is regular.

    Sufficiency. Let (D,L) be regular. Let rzSP(D,Σ) and let CUC(L) such that rz˜CU. Then, we have zUL. So, we find VL such that zVClL(V)U. Thus, we have CVC(L) and rz˜CV˜CClL(V)=ClC(L)(CV)˜CU. Therefore, (D,C(L),Σ) is soft regular.

    Theorem 2.7. Let {(H,Lr):rΣ} be a collection of TSs. Then, (H,rΣLr,Σ) is soft A-R iff (H,Lr) is A-R for every rΣ.

    Proof. Necessity. Let (H,rΣLr,Σ) be soft A-R and let rΣ. Let zH and let ULr such that zU. Then, rz˜rUrΣLr. So, by Theorem 3.4 (ⅳ) of [31], we find LrΣLr such that rz˜L˜ClrΣLr(L)˜IntrΣLr(ClrΣLr(rU)). Thus, we have zL(r)Lr and zL(r)(ClrΣLr(L))(r)(IntrΣLr(ClrΣLr(rU)))(r).

    In contrast, by Lemma 4.9 of [40], (ClrΣLr(L))(r)=ClLr(L(r)) and (IntrΣLr(ClrΣLr(rU)))(r) =IntLr((ClLr(rU))(r))=IntLr(ClLr(U)). Thus, by Theorem 3.4 (ⅳ) of [31], (H,Lr) is A-R.

    Sufficiency. Let (H,Lr) be A-R for every rΣ. Let rzSP(H,Σ) and let KrΣLr such that rz˜K. By Theorem 3.5 of [34], we find ULr such that rz˜rU˜K. Then, we have zULr. So, by Theorem 2.1 (d) of [39], we find VLr such that zVClLr(V)IntLr(ClLr(U)). Thus, we have rVrΣLr and

    rz˜rV˜rClLr(V)=ClrΣLr(rV)˜rIntLr(ClLr(U))=IntrΣLr(ClrΣLr(rU))˜IntrΣLr(ClrΣLr(K)).

    Corollary 2.8. Let (D,L) be a TS. Then, for any set Σ, (D,τ(L),Σ) is soft A-R iff (D,L) is A-R.

    Proof. For each rΣ, set Lr=L. Then, τ(L)=rΣLr, and by Theorem 2.7 we get the result.

    Theorem 2.9. Let {(H,Lr):rΣ} be a collection of TSs. Then, (H,rΣLr,Σ) is soft ω-A-R iff (H,Lr) is ω-A-R for every rΣ.

    Proof. Necessity. Let (H,rΣLr,Σ) be soft ω-A-R and let rΣ. Let zH and let ULr such that zU. Then, rz˜rUrΣLr. So, by Theorem 2.2 (e), we find LrΣLr such that rz˜L˜ClrΣLr(L)˜IntrΣLr(Cl(rΣLr)ω(rU)). Thus, we have zL(r)Lr and zL(r)(ClrΣLr(L))(r)(IntrΣLr(ClrΣLr(rU)))(r).

    In contrast, by Lemma 4.9 of [40] and Theorem 8 of [35], (ClrΣLr(L))(r)=ClLr(L(r)) and IntrΣLr(Cl(rΣLr)ω(rU))(r) =IntLr((Cl(Lr)ω(rU))(r)) =IntLr(Cl(Lr)ω(U)). Thus, by Theorem 2.1 (e) of [39] (H,Lr) is ω-A-R.

    Sufficiency. Let (H,Lr) be ω-A-R for every rΣ. Let rzSP(H,Σ) and let KrΣLr such that rz˜K. By Theorem 3.5 of [34] we find ULr such that rz˜rU˜K. Then, we have zULr. So, by Theorem 2.1 (e) of [39] we find VLr such that zVClLr(V)IntLr(Cl(Lr)ω(U)). Thus, we have rVrΣLr and

    rz˜rV˜rClLr(V)=ClrΣLr(rV)˜rIntLr(Cl(Lr)ω(U))=IntrΣLr(Cl(rΣ(Lr)ω)(rU))˜IntrΣLr(ClrΣ(Lr)ω(K))=IntrΣLr(Cl(rΣLr)ω(K)).

    Corollary 2.10. Let (D,L) be a TS. Then, for any set Σ, (D,τ(L),Σ) is soft ω-A-R iff (D,L) is ω-A-R.

    Proof. For each rΣ, set Lr=L. Then, τ(L)=rΣLr and by Theorem 2.9 we get the result.

    Theorem 2.11. Soft regular STSs are soft ω-A-R.

    Proof. Let (H,φ,Σ) be soft regular. Let rzSP(H,Σ) and KSωO(H,φ,Σ) such that rz˜K. Since by Theorem 3 of [36] we have RωO(H,φ,Σ)φ, then Kφ. Since (H,φ,Σ) is soft regular, then we find Gφ such that rz˜G˜Clφ(G)˜K. Thus, by Theorem 2.2 (2) (H,φ,Σ) is soft ω-A-R.

    Theorem 2.12. Soft ω-A-R STSs are soft A-R.

    Proof. Let (H,φ,Σ) be soft ω-A-R. Let rzSP(H,Σ) and KSO(H,φ,Σ) such that rz˜K. Since by Theorem 3 of [36] we have RO(H,φ,Σ)RωO(H,φ,Σ), then KSωO(H,φ,Σ). Since (H,φ,Σ) is soft ω-A-R, then by Theorem 2.2 (b) there is Gφ such that rz˜G˜Clφ(G)˜K. Thus, by Theorem 2.2 (b) of [31], (H,φ,Σ) is soft A-R.

    Theorem 2.13. Soft L-C soft ω-A-R STSs are soft regular.

    Proof. Let (H,φ,Σ) be soft L-C and soft ω -A-R. Let rzSP(H,Σ) and Kφ such that rz˜K. Since (H,φ,Σ) is soft L-C, then by Theorem 5 of [36] KSωO(H,φ,Σ). Since (H,φ,Σ) is soft ω-A-R, then by Theorem 2.2 (2) there is Gφ such that rz˜G˜Clφ(G)˜K. Therefore, (H,φ,Σ) is soft regular.

    Theorem 2.14. Soft anti-L-C soft A-R STSs are soft ω-A-R.

    Proof. Let (H,φ,Σ) be soft anti-L-C and soft A-R. Let rzSP(H,Σ) and KSωO(H,φ,Σ) such that rz˜K. Since (H,φ,Σ) is anti-L-C, then by Theorem 6 of [36] KSO(H,φ,Σ). Since (H,φ,Σ) is soft A-R, then by Theorem 3.4 (ⅱ) of [31] there is Gφ such that rz˜G˜Clφ(G)˜K. Therefore, by Theorem 2.2 (b), (H,φ,Σ) is soft ω-A-R.

    Theorem 2.15. For any STS (H,φ,Σ), (H,φω,Σ) is soft A-R iff (H,φ,Σ) is soft ω -A-R.

    Proof. Necessity. Let (H,φω,Σ) be soft A-R. Let rzSP(H,Σ) and KSωO(H,φ,Σ) such that rz˜K. By Theorem 7 of [36] KSO(H,φω,Σ). Since (H,φω,Σ) is soft A-R, then by Theorem 3.4 (ⅱ) of [31] there is Gφ such that rz˜G˜Clφ(G)˜K. Therefore, by Theorem 2.2 (b) (H,φ,Σ) is soft ω-A-R.

    Sufficiency. Let (H,φ,Σ) be soft ω-A-R. Let rzSP(H,Σ) and KSO(H,φω,Σ) such that rz˜K. By Theorem 7 of [36] KSωO(H,φ,Σ). Since (H,φ,Σ) is soft ω-A-R, then by Theorem 2.2 (2) there is Gφ such that rz˜G˜Clφ(G)˜K. Therefore, by Theorem 3.4 (ⅱ) of [31] (H,φω,Σ) is soft A-R.

    The previously mentioned theorems lead to the following implications, yet Examples 2.16 and 2.17 that follow demonstrate that the opposite of these implications is false.

    Soft regular Soft ω-A-R Soft A-R.

    The following two examples show that any of the conditions soft L-C and soft anti-L-C in Theorems 2.13 and 2.14 cannot be dropped:

    Example 2.16. Consider (R,C(Θ),Z), where Θ is the cofinite topology on R. Since (R,Θ) is not regular, by Theorem 2.6 (R,C(Θ),Z) is not soft regular. In contrast, since (R,C(Θ),Z) is anti-L-C, then by Theorem 6 of [36] RωO(R,C(Θ),Z)=RO(R,C(Θ),Z)={0Z,1Z}, and thus (R,C(Θ),Z) is soft ω-A-R.

    Example 2.17. Consider (N,C(Θ),{a,b}), where Θ is the cofinite topology on N. Since (N,Θ) is not regular, by Theorem 2.6 (N,C(Θ),{a,b}) is not soft regular. Since (N,C(Θ),{a,b}) is soft L-C, then by Theorem 2.13 (N,C(Θ),{a,b})is not soft ω-A-R. In contrast, since RO(N,C(Θ),{a,b})={0{a,b},1{a,b}}, then (N,C(Θ),{a,b}) is soft A-R.

    The following lemma will be used in the next main result:

    Lemma 2.18. Let (H,φ,Σ) be an STS. If CY is a soft dense subset of (H,φω,Σ), then for any soft subset HSS(Y,Σ) IntφY(Cl(φω)Y(H))=Intφ(Clφω(H))˜CY.

    Proof. Suppose that CY is a soft dense subset of (H,φω,Σ) and let HSS(Y,Σ). To see that IntφY(Cl(φω)Y(H))˜Intφ(Clφω(H))˜CY, let ax˜IntφY(Cl(φω)Y(H)). Since IntφY(Cl(φω)Y(H))φY, then there is Mφ such that IntφY(Cl(φω)Y(H))=M˜CY. Thus, we have ax˜M˜CY˜Cl(φω)Y(H)=(Clφω(H))˜CY.

    Claim. M˜Clφω(H).

    Proof of Claim. Suppose to the contrary that M˜(1ΣClφω(H))0Σ. Since 1ΣClφω(H)φω and Mφφω, then M˜(1ΣClφω(H))φω. Since CY is soft dense in (H,φω,Σ), then M˜(1ΣClφω(H))˜CY0Σ. Choose by˜M˜(1ΣClφω(H))˜CY. Thus, we have by˜1ΣClφω(H) and by˜M˜CY˜(Clφω(H))˜CY˜Clφω(H), a contradiction.

    Therefore, by the above Claim, we must have ax˜M˜Clφω(H), and hence ax˜Intφ(Clφω(H). Hence, ax˜Intφ(Clφω(H))˜CY.

    To see that Intφ(Clφω(H))˜CY˜IntφY(Cl(φω)Y(H)), let ax˜Intφ(Clφω(H))˜CY. Since ax˜Intφ(Clφω(H))φ, then there is Mφ such that ax˜M˜Clφω(H) and so ax˜M˜CY˜Clφω(H)˜CY=Cl(φω)Y(H). Since M˜CYφω, then ax˜IntφY(Cl(φω)Y(H)).

    Theorems 2.19 and 2.21 establish that soft ω-almost-regularity is heritable for specific types of soft subspaces.

    Theorem 2.19. If (H,φ,Σ) is a soft ω-A-R STS and CY is a soft dense subspace of (H,φω,Σ), then (Y,φY,Σ) is soft ω-A-R.

    Proof. Let axSP(Y,Σ) and let HSωO(Y,φY,Σ) such that ax˜H. Since HSωO(Y,φY,Σ), then IntφY(Cl(φY)ω(H))=H. Since by Theorem 15 of [35] (φω)Y=(φY)ω, then IntφY(Cl(φω)Y(H))=H. So, by Lemma 2.18 H=Intφ(Clφω(H))˜CY. Thus, we have ax˜Intφ(Clφω(H))SωO(H,φ,Σ). Since (H,φ,Σ) is soft ω-A-R, then by Theorem 2.2 (2) there is Lφ such that ax˜L˜Clφ(L)˜Intφ(Clφω(H)). Therefore, we have ax˜L˜CYφY and ClφY(L˜CY)=Clφ(L˜CY)˜CY˜Intφ(Clφω(H))˜CY=H. This shows that (Y,φY,Σ) is soft ω -A-R.

    The following lemma will be used in the next main result:

    Lemma 2.20. Let (H,φ,Σ) be an STS and let CYSωO(H,φ,Σ){0Σ}, then RωO(Y,φY,Σ)RωO(H,φ,Σ).

    Proof. Let CYSωO(H,φ,Σ){0Σ} and let HSωO(Y,φY,Σ). Then, H=IntφY(Cl(φY)ω(H)). Since by Theorem 15 of [35], (φω)Y=(φY)ω, then Cl(φY)ω(H)=Cl(φω)Y(H)=Clφω(H)˜CY. Since by Theorem 3 of [36] RωO(H,φ,Σ)φ, then CYφ and so IntφY(Cl(φY)ω(H))=Intφ((Cl(φY)ω(H))). Thus, H=Intφ(Clφω(H)˜CY) =Intφ(Clφω(H))˜Intφ(CY)=Intφ(Clφω(H))˜CY. Since H˜CY, then Intφ(Clφω(H))˜Intφ(Clφω(CY))=CY and thus, Intφ(Clφω(H))˜CY=Intφ(Clφω(H)). Therefore, H=Intφ(Clφω(H)). Hence, HSωO(H,φ,Σ).

    Theorem 2.21. If (H,φ,Σ) is a soft ω-A-R STS and CYSωO(H,φ,Σ){0Σ}, then (Y,φY,Σ) is soft ω-A-R.

    Proof. Let axSP(Y,R) and let HSωO(Y,φY,Σ) such that ax˜H. By Lemma 2.20, HSωO(H,φ,Σ). Since (H,φ,Σ) is soft ω-A-R, then by Theorem 2.2 (2) there is Lφ such that ax˜L˜Clφ(L)˜H. Therefore, we have ax˜L˜CYφY and ClφY(L)=Clφ(L)˜CY˜H. Hence, (Y,φY,Σ) is soft ω-A-R.

    The following lemma will be used in Theorems 2.23 and 3.33:

    Lemma 2.22. For any two STSs (Z,δ,Σ) and (W,ρ,Ψ), (δ×ρ)δωδδω×ρδω.

    Proof. Let T(δ×ρ)δω and (e,f)(z,w)˜T. Then, by Theorem 20 of [36] we find SSωO(Z×W,δ×ρ,Σ×Ψ) such that (e,f)(z,w)˜S=Intδ×ρ(Cl(δ×ρ)ω(S))˜T. Choose Lδ and Mρ such that (e,f)(z,w)˜L×M˜S˜T. By Proposition 3 (b) of [33] we have Clδω(L)×Clρω(M)˜Cl(δ×ρ)ω(L×M), and so

    L×M˜Intδ(Clδω(L))×Intρ(Clρω(M))˜Intδ×ρ(Clδω(L)×Clρω(M))˜Intδ×ρ(Cl(δ×ρ)ω(L×M))˜T.

    By Theorem 9 and Corollary 7 of [36], Intδ(Clδω(L))δδω and Intρ(Clρω(M))ρδω. It follows that Tδδω×ρδω.

    The following result shows that soft ω-almost-regularity is a productive soft property:

    Theorem 2.23. The soft product of two soft ω-A-R STSs is soft ω-A-R.

    Proof. Let (Z,δ,Σ) and (W,ρ,Ψ) be two soft ω-A-R STSs. Let (e,f)(z,w)˜SP(Z×W,Σ×Ψ) and let KSωO(Z×W,δ×ρ,Σ×Ψ) such that (e,f)(z,w)˜K. Then, by Corollary 7 of [36] G(δ×ρ)δω. So, by Lemma 2.22 Kδδω×ρδω. Thus, there are Lδδω and Mρδω such that (e,f)(z,w)˜L×M˜K. By Corollary 7 of [36] we find SSωO(Z,δ,Σ) and TSωO(W,ρ,Ψ) such that (e,f)(z,w)˜S×T˜L×M˜G. So, by Theorem 2.2 (2) there are Mδ and Nρ such that ez˜M˜Clδ(M)˜S and fw˜N˜Clρ(N)˜T. Therefore, we have M×Nδ×ρ and (e,f)(z,w)˜M×N˜Clδ×ρ(M×N)=Clδ(M) ×Clρ(N)˜S×T˜L×M˜K. Again, by Theorem 2.2 (2) (Z×W,δ×ρ,Σ×Ψ) is soft ω-A-R.

    The following result shows that soft almost-regularity is a productive soft property:

    Theorem 2.24. Let (Z,δ,Σ) and (W,ρ,Ψ) be two STSs. Then (Z×W,δ×ρ,Σ×Ψ) is soft A-R iff (Z,δ,Σ) and (W,ρ,Ψ) are both soft A-R.

    Proof. Necessity. Let (Z×W,δ×ρ,Σ×Ψ) be soft A-R. To see that (Z,δ,Σ) is soft A-R, let ezSP(Z,Σ) and GSO(Z,δ,Σ) such that ez˜G. Choose fw˜SP(W,F). Then, (e,f)(z,w)˜G×1ΨSO(Z×W,δ×ρ,Σ×Ψ). Thus, by Theorem 3.4 (ⅱ) of [31] we find Hδ×ρ such that (e,f)(z,w)˜H˜Clδ×ρ(H)˜G×1Ψ. Choose Mδ and Nρ such that (e,f)(z,w)˜M×N˜H. Thus,

    (e,f)(z,w)˜M×N˜Clδ(M)×Clρ(N)=Clδ×ρ(M×N)˜Clδ×ρ(H)˜G×1Ψ.

    Therefore, we have ez˜M˜Clδ(M)˜G. Hence, by Theorem 3.4 (ⅱ) of [31] (Z,δ,Σ) is soft A-R. Similarly, we can show that (W,ρ,Ψ) is soft A-R.

    Sufficiency. Let (Z,δ,Σ) and (W,ρ,Ψ) be soft A-R. Let (e,f)(z,w)˜SP(Z×W,Σ×Ψ) and let KSO(Z×W,δ×ρ,Σ×Ψ) such that (e,f)(z,w)˜K. Choose Mδ and Nρ such that (e,f)(z,w)˜M×N˜K. Since ez˜M and fw˜N, by Theorem 3.4 (ⅳ) of [31] there are SSO(Z,δ,Σ) and TSO(W,ρ,Ψ) such that ez˜S˜Clδ(S)˜Intδ(Clδ(M)) and fw˜T˜Clρ(T)˜Intρ(Clρ(N)). Thus, we have S×TSO(Z×W,δ×ρ,Σ×Ψ) and

    (e,f)(z,w)˜S×T˜Clδ(S)×Clρ(T)=Clδ×ρ(S×T)˜Intδ(Clδ(M))×Intρ(Clρ(N))=Intδ×ρ(Clδ×ρ(M×N))˜Intφ×ρ(Clδ×ρ(M×N))=K.

    Therefore, by Theorem 3.4 (ⅳ) of [31] (Z×W,δ×ρ,Σ×Ψ) is soft A-R.

    In this section, we define soft ω-semi-regularity and soft ω- T212 as two new soft separation axioms. We show that soft ω-semi-regularity is a weaker form of both soft semi-regularity and soft ω-regularity, and soft ω-T212 lies strictly between soft T212 and soft T2. Also, we provide several sufficient conditions establishing the equivalence between these newly introduced axioms and their relevant counterparts. Moreover, a decomposition theorem for soft regularity through the interplay of soft ω-semi-regularity and soft ω-almost-regularity is obtained. In addition, we investigated the links between these classes of soft topological spaces and their analogs in general topology.

    Definition 3.1. An STS (H,φ,Σ) is called soft ω-semi-regular (soft ω-S-R, for simplicity) if RωO(H,φ,Σ) forms a soft base for φ.

    Two characterizations of soft ω-semi-regularity are listed in the following theorem.

    Theorem 3.2. For any STS (H,φ,Σ), T.F.A.E:

    (a) (H,φ,Σ) is soft ω-S-R.

    (b) For every Hφ{0Σ} and every rz˜H, we find Kφ such that rz˜K˜Intφ(Clφω(K))˜H.

    (c) φδω=φ.

    Proof. (a) (b): Let Hφ{0Σ} and let rz˜H. By (a) we find KSωO(H,φ,Σ) such that rz˜K=Intφ(Clφω(K))˜H.

    (b) (c): By Theorem 21 of [36] we have φδω˜φ. To show that φφδω, let Hφ{0Σ}, then for every rz˜H we find Krzφ such that rz˜Krz˜Intφ(Clφω(Krz))˜H. Let K=˜rz˜HIntφ(Clφω(Krz)). Since for every rz˜H Intφ(Clφω(Krz))SωO(H,φ,Σ)φδω, then Kφδω.

    (c) (a): Since RωO(H,φ,Σ) is a soft base for φδω, and by (c) φδω=φ, then RωO(H,φ,Σ) is a soft base for φ. Therefore, (H,φ,Σ) is soft ω-S-R.

    Corollary 3.3. Every soft ω-regular STS is soft ω -S-R.

    Proof. The proof follows from Theorem 25 of [36] and Theorem 3.2.

    Theorem 3.4. Every soft S-R STS is soft ω-S-R.

    Proof. Let (H,φ,Σ) be soft S-R. Then φδ=φ. So, by Theorem 21 of [36] φ=φδφδωφ, and thus φδω=φ. Therefore, by Theorem 3.2, (H,φ,Σ) is soft ω-S-R.

    Theorem 3.5. Every soft ω-S-R soft anti-L-C STS is soft S-R.

    Proof. Let (H,φ,Σ) be soft ω-S-R soft anti-L-C. Since (H,φ,Σ) is soft ω-S-R, then RωO(H,φ,Σ) is a soft base for φ. Since (H,φ,Σ) is soft anti-L-C, then by Theorem 6 of [36], RO(H,φ,Σ)=RωO(H,φ,Σ). So, RO(H,φ,Σ) is a soft base for φ. Hence, (H,φ,Σ) is soft S-R.

    The following implications come from the previous theorems; nevertheless, Examples 3.15 and 3.16 show that the converses of these implications are not true.

     soft S-Rsoft ω-S-Rsoft ω-regular.

    Theorem 3.6. Soft L-C STSs are soft ω-S-R.

    Proof. Let (D,φ,Σ) be soft L-C. Then, by Theorem 5 of [36] RωO(D,φ,Σ)=φ. So, RωO(D,φ,Σ) is a soft base for φ. Hence, (D,φ,Σ) is soft ω-S-R.

    Theorem 3.7. Let (D,φ,Σ) be an STS. If (D,φω,Σ) is soft ω-S-R, then (D,φω,Σ) is soft S-R.

    Proof. Let (D,φω,Σ) be soft ω-S-R. Then, RωO(D,φω,Σ) is a soft base for φω. Since by Theorem 7 of [36] RωO(D,φω,Σ)=RO(D,φω,Σ), then RO(D,φω,Σ) is a soft base for φω. Hence, (D,φω,Σ) is soft S-R.

    In Theorems 3.8, 3.9, 3.11, and Corollary 3.12, we discuss the connections between soft semi-regularity and its analog in traditional topological spaces. Also, in Theorems 3.10, 3.13, and Corollary 3.14, we discuss the connections between soft ω-semi-regularity and its analog in traditional topological spaces.

    Theorem 3.8. If (H,φ,Σ) is full and soft S-R, then (H,φr) is S-R for all rΣ.

    Proof. Let (H,φ,Σ) be full and soft S-R. Let rΣ. Let zH and let Wφr such that zW. Choose Kφ such that K(r)=W. Since (H,φ,Σ) is soft S-R and rz˜Kφ, we find LSO(H,φ,Σ) such that rz˜L˜K and so zL(r)K(r)=W. In contrast, by Theorem 13 of [36], L(r)SO(H,φr). This shows that (H,φr) is S-R for all rΣ.

    Theorem 3.9. Let (D,L) be a TS. Then, for any set Σ, (D,C(L),Σ) is soft S-R iff (D,L) is S-R.

    Proof. Necessity. Let (D,C(L),Σ) be soft S-R. Pick rΣ. Since it is clear that (D,C(L),Σ) is full, then by Theorem 3.8 (D,(C(L))r)=(D,L) is S-R.

    Sufficiency. Let (D,L) be S-R. Let rzSP(D,Σ) and let CUC(L) such that rz˜CU. Then, we have zUL. So, we find VSO(D,L) such that zIntL(ClL(V))U. Thus, we have CVC(L) and rz˜CIntL(ClL(V))=IntC(L)(ClC(L)(CV))˜CU. This shows that (D,C(L),Σ) is soft S-R.

    Theorem 3.10. Let (D,L) be a TS. Then, for any set Σ, (D,C(L),Σ) is soft ω-S-R iff (D,L) is ω-S-R.

    Proof. Necessity. Let (D,C(L),Σ) be soft ω-S-R. Let zD and UL such that zU. Pick rΣ. Then, we have rz˜CUC(L). Since (D,C(L),Σ) is soft ω-S-R, we find CVSωO(D,C(L),Σ) such that rz˜CV˜CU. Therefore, we have zVSωO(D,L) and VU. This shows that (D,L) is ω-S-R.

    Sufficiency. Let (D,L) be ω-S-R. Let rzSP(H,Σ) and let CUC(L) such that rz˜CU. Then, we have zUL. So, we find VSωO(D,L) such that zVU. Thus, we have RωO(D,C(L),Σ) and rz˜CV˜CU. This shows that (D,C(L),Σ) is soft ω-S-R.

    Theorem 3.11. Let {(H,Lr):rΣ} be a collection of TSs. Then (H,rΣLr,Σ) is soft S-R iff (H,Lr) is S-R for every rΣ.

    Proof. Necessity. Let (H,rΣLr,Σ) be soft S-R and let rΣ. Let zH and let ULr such that zU. Then, rz˜rUrΣLr. So, we find LSO(H,rΣLr,Σ) such that rz˜L˜rU and thus, zL(r)(rU)(r)=U. In contrast, by Theorem 14 of [36] L(r)SO(H,Lr). This shows that (H,Lr) is S-R.

    Sufficiency. Let (H,Lr) be S-R for every rΣ. Let rzSP(H,Σ) and let KrΣLr such that rz˜K. Then, we have zK(r)Lr and so we find VSO(H,Lr) such that zVU. Now, we have rz˜rV˜rU˜K, and by Theorem 14 of [36] rVSO(H,rΣLr,Σ). This shows that (H,rΣLr,Σ) is soft S-R.

    Corollary 3.12. Let (D,L) be a TS. Then, for any set Σ, (D,τ(L),Σ) is soft S-R iff (D,L) is S-R.

    Proof. For each rΣ, set Lr=L. Then, τ(L)=rΣLr and by Theorem 3.11 we get the result.

    Theorem 3.13. Let {(H,Lr):rΣ} be a collection of TSs. Then, (H,rΣLr,Σ) is soft ω-S-R iff (H,Lr) is ω-S-R for every rΣ.

    Proof. Necessity. Let (H,rΣLr,Σ) be soft ω-S-R and let rΣ. Let zH and let ULr such that zU. Then, rz˜rUrΣLr. So, we find LSωO(H,rΣLr,Σ) such that rz˜L˜rU, and thus zL(r)(rU)(r)=U. In contrast, by Theorem 15 of [36] L(r)SωO(H,Lr). This shows that (H,Lr) is ω-S-R.

    Sufficiency. Let (H,Lr) be ω-S-R for every rΣ. Let rzSP(H,Σ) and let KrΣLr such that rz˜K. Then, we have zK(r)Lr and so we find VSωO(H,Lr) such that zVU. Now, we have rz˜rV˜rU˜K, and by Theorem 15 of [36] rVSωO(H,rΣLr,Σ). This shows that (H,rΣLr,Σ) is soft ω-S-R.

    Corollary 3.14. Let (D,L) be a TS. Then, for any set Σ, (D,τ(L),Σ) is soft ω-S-R iff (D,L) is ω-S-R.

    Proof. For each rΣ, set Lr=L. Then, τ(L)=rΣLr and by Theorem 3.13 we get the result.

    The following two examples show, respectively, that each of Theorem 3.4 and Corollary 3.3 does not have to be true in all cases:

    Example 3.15. Consider (H,φ,Σ) in Example 2.17. RO(H,φ,Σ)={0Σ,1Σ} is not a soft base for φ and thus (H,φ,Σ) is not soft S-R. In contrast, by Theorem 3.6 (H,φ,Σ) is soft ω-S-R.

    Example 3.16. Let (D,L) be as in Example 3.9 of [39]. It is proved in [39] that (D,L) is ω-S-R but not ω-regular. Therefore, by Corollaries 3.14 and 19 of [33], (D,τ(L),Σ) is soft ω-S-R but not soft ω-regular.

    The following main result introduces a decomposition of soft regularity in terms of soft ω-semi-regularity and soft ω-almost-regularity:

    Theorem 3.17. An STS (H,φ,Σ) is soft regular iff it is soft ω-S-R and soft ω-A-R.

    Proof. Necessity. Let (H,φ,Σ) be soft regular. Then, by Theorem 15 of [33] and Corollary 3.3 (H,φ,Σ) is soft ω-S-R. In contrast, by Theorem 2.11 (H,φ,Σ) is soft ω-A-R.

    Sufficiency. Let (H,φ,Σ) be soft ω -S-R and soft ω-A-R. Let Hφ{0Σ} and let rz˜H. Since (H,φ,Σ) is soft ω-S-R, then there is GSωO(H,φ,Σ) such that rz˜G˜H. Since (H,φ,Σ) is soft ω-A-R, then by Theorem 2.2 (2) there is Tφ such that rz˜T˜Clφ(T)˜G˜H. Hence, (H,φ,Σ) is soft regular.

    Definition 3.18. An STS (H,φ,Σ) is called soft ω-T212 if for every rx,sySP(H,Σ) such that rxsy, we find K,Gφ such that rx˜K, sy˜G, and Clφω(K)˜Clφω(G)=0Σ.

    In Theorems 3.19, 3.21, and Corollary 3.12, we discuss the connections between soft T212 spaces and their analogs in traditional topological spaces. Also, in Theorems 3.20, 3.23, and Corollary 3.24, we discuss the connections between soft ω-T212 spaces and their analogs in traditional topological spaces.

    Theorem 3.19. If (H,φ,Σ) is soft T212, then (H,φr) is T212 for every rΣ.

    Proof. Suppose that (H,φ,Σ) is soft T212 and let rΣ. Let x,yZ such that xy. Then, rx,rySP(S,D) such that rxry. Since (H,φ,Σ) is soft T212, we find K,Gφ such that rx˜K, ry˜G, and Clφ(K)˜Clφ(G)=0Σ. Thus, we have xK(r)φr, yG(r)φr, and by Proposition 7 of [12] Clφr(K(r))Clφr(G(r))(Clφ(K))(r)(Clφ(G))(r) =(Clφ(K)˜Clφ(G))(r)=. This shows that (H,φr) is T212.

    Theorem 3.20. If (H,φ,Σ) is soft ω-T212, then (H,φr) is ω-T212 for every rΣ.

    Proof. Suppose that (H,φ,Σ) is soft ω-T212 and let rΣ. Let x,yZ such that xy. Then, rx,rySP(S,D) such that rxry. Since (H,φ,Σ) is soft ω-T212, we find K,Gφ such that rx˜K, ry˜G, and Clφω(K)˜Clφω(G)=0Σ. Thus, we have xK(r)φr, yG(r)φr, and by Proposition 7 of [12] Cl(φω)r(K(r))Cl(φω)r(G(r))(Clφω(K))(r) (Clφω(G))(r) =(Clφω(K)˜Clφω(G))(r)=. But by Theorem 7 of [35], (φω)r=(φr)ω. This shows that (H,φr) is ω-T212.

    Theorem 3.21. Let {(H,LR):rΣ} be a collection of TSs. Then, (H,rΣLr,Σ) is soft T212 iff (H,Lr) is T212 for every rΣ.

    Proof. Necessity. Suppose that (H,rΣLr,Σ) is soft T212 and let rΣ. Then, by Theorem 3.19 (H,(rΣLr)r) is T212. On the other hand, by Theorem 3.7 of [34] (rΣLr)r=Lr.

    Sufficiency. Suppose that (H,Lr) is T212 for every rΣ. Let rx,sySP(H,Σ) such that rxsy.

    Case 1. rs. Then, rx˜rZrΣLr, sy˜sZrΣLr, and ClrΣLr(rZ)˜ClrΣLr(sZ)=0Σ.

    Case 2. r=s. Then, xy. Since (H,Lr) is T212, we find U,VLr such that xU, yV, and ClLr(U)ClLr(V)=. Then, we have rx˜rUrΣLr, sy˜sVrΣLr and ClrΣLr(rU)ClrΣLr(sV)=0Σ.

    Corollary 3.22. Let (D,L) be a TS. Then, for any set Σ, (D,τ(L),Σ) is soft T212 iff (D,L) is T212.

    Proof. For each rΣ, put Lr=L. Then, τ(L)=rΣLr. We get the result as a consequence of Theorem 3.21.

    Theorem 3.23. Let {(H,Lr):rΣ} be a collection of TSs. Then, (H,rΣLr,Σ) is soft ω-T212 iff (H,Lr) is ω-T212 for every rΣ.

    Proof. Necessity. Suppose that (H,rΣLr,Σ) is soft ω-T212 and let rΣ. Then, by Theorem 3.20 (H,(rΣLr)r) is ω-T212. In contrast, by Theorem 3.7 of [35], (rΣLr)r=Lr.

    Sufficiency. Suppose that (H,Lr) is ω-T212 for every rΣ. Let rx,sySP(H,Σ) such that rxsy.

    Case 1. rs. Then, rx˜rZrΣLr, sy˜sZrΣLr, and Cl(rΣLr)ω(rZ)˜Cl(rΣLr)ω(sZ)=0Σ.

    Case 2. r=s. Then, xy. Since (D,L) is ω-T212, we find A,BLr such that xA, yB, and AB=. Then, we have rx˜rArΣLr, sy˜sBrΣLr and Cl(rΣLr)ω(rA)˜Cl(rΣLr)ω(sB)=0Σ.

    Corollary 3.24. Let (D,L) be a TS. Then, for any set Σ, (D,τ(L),Σ) is soft ω-T212 iff (D,L) is ω-T212.

    Proof. For each rΣ, put Lr=L. Then, τ(L)=rΣLr. The result follows from Theorem 3.23.

    Theorem 3.25. If (H,φ,Σ) is soft T212 , then (H,φ,Σ) is soft ω-T212.

    Proof. Let (H,φ,Σ) be soft T212 and let rx,sySP(H,Σ) such that rxsy. Then, we find K,Gφ such that rx˜K, sy˜G, and Clφ(K)˜Clφ(G)=0Σ. Since Clφω(K)˜Clφω(G)˜Clφ(K)˜Clφ(G)=0Σ, then Clφω(K)˜Clφω(G)=0Σ. This shows that (H,φ,Σ) is soft ω-T212.

    Theorem 3.26. If (H,φ,Σ) is soft anti-L-C and soft ω-T212, then (H,φ,Σ) is soft T212.

    Proof. Let (H,φ,Σ) be soft anti-L-C and soft ω-T212. Let rx,sySP(H,Σ) such that rxsy. Since (H,φ,Σ) is soft ω-T212, then we find K,Gφ such that rx˜K, sy˜G, and Clφω(K)˜Clφω(G)=0Σ. Since (H,φ,Σ) is anti-L-C, then by Theorem 14 of [35] Clφ(K)˜Clφ(G)=Clφω(K)˜Clφω(G)=0Σ. Hence, (H,φ,Σ) is soft T212.

    Theorem 3.27. Every soft ω-T212 STS is soft T2.

    Proof. Let (H,φ,Σ) be ω-T212 and let rx,sySP(H,Σ) such that rxsy. Then, we find K,Gφ such that rx˜K, sy˜G, and Clφω(K)˜Clφω(G)=0Σ. Since K˜G˜Clφω(K)˜Clφω(G)=0Σ, then K˜G=0Σ. Hence, (H,φ,Σ) is soft T2.

    Theorem 3.28. If (H,φ,Σ) is soft L-C and soft T2, then (H,φ,Σ) is soft ω-T212.

    Proof. Let (H,φ,Σ) be soft L-C and soft T2. Let rx,sySP(H,Σ) such that rxsy. Since (H,φ,Σ) is soft T2, then we find K,Gφ such that rx˜K, sy˜G, and K˜G=0Σ. Since (H,φ,Σ) is soft L-C, then by Corollary 5 of [35] Clφω(K)˜Clφω(G)=K˜G=0Σ. Hence, (H,φ,Σ) is soft ω-T212.

    The following example demonstrates that Theorem 3.25's converse does not have to be true in general:

    Example 3.29. Let (D,L) be the TS in Example 75 of [41]. Then (D,L) is T2 but not T212. Since (D,τ(L),N) is soft L-C, by Corollary 5 of [35] it is soft T2. Thus, by Theorem 3.28 (D,τ(L),N) is soft ω-T212. On the other hand, by Corollary 2.22 (D,τ(L),N) is not soft T212.

    The following example demonstrates why Theorem 3.27 does not have to be true in general:

    Example 3.30. Let (D,L) be the TS in Example 81 of [41]. It is known that (D,L) is T2 but not T212. Then, by Corollary 7 of [33] and Corollary 2.22 (D,τ(L),[0,1]) is soft T2 but not soft T212. Since (D,τ(L),[0,1]) is soft anti-L-C, then by Theorem 3.26 (D,τ(L),[0,1]) is not ω-T212.

    Theorem 3.31. Every soft ω-regular T2 STS is soft ω-T212.

    Proof. Let (H,φ,Σ) be soft ω-regular and soft T2. Let rx,sySP(H,Σ) such that rxsy. Since (H,φ,Σ) is soft T2, then we find K,Gφ such that rx˜K, sy˜G, and K˜G=0Σ. Since (H,φ,Σ) is soft ω-regular, then we find L,Mφ such that rx˜L˜Clφω(L)˜K and sy˜M˜Clφω(M)˜G. Therefore, we have rx˜L, sy˜M, and Clφω(L)˜Clφω(M)˜K˜G=0Σ. This proves that (H,φ,Σ) is soft ω-T212.

    Question 3.32. Is it true that every soft ω-T212 STS is soft ω-regular?

    Theorem 3.33. If (Z,β,Σ) and (W,ρ,Ψ) are two soft ω-S-R STSs such that the soft product (Z×W,β×ρ,Σ×Ψ) is soft ω-S-R, then both of (Z,β,Σ) and (W,ρ,Ψ) are soft ω-S-R.

    Proof. Since (Z×W,β×ρ,Σ×Ψ) is soft ω-S-R, then by Theorem 3.2 (β×ρ)δω=β×ρ. So, by Lemma 2.22, β×ρ˜ βδω×ρδω. Hence, β = βδω and ρ=ρδω. Therefore, again by Theorem 3.2 (Z,β,Σ) and (W,ρ,Ψ) are soft ω-S-R.

    Soft separation axioms are a collection of requirements for categorizing a system of STSs based on certain soft topological features. These axioms are often expressed in terms of classes of soft sets.

    In this work, "soft ω-almost-regular", "soft ω -semi-regular", and "soft ω-T212" are defined as three new notions of soft separation axioms (Definitions 2.1, 3.1, 3.18). Several characterizations of soft ω-almost-regularity (Theorems 2.2) and soft ω-semi-regularity (Theorem 3.2) are given. It is proved that soft ω-almost-regularity lies strictly between regularity and almost-regularity (Theorems 2.11, 2.12 and Examples 2.16, 2.17); soft ω-semi-regularity is a weaker form of both soft semi-regularity and soft ω-regularity (Corollary 3.3, Theorem 3.4 and Examples 3.15, 3.16); soft ω-T212 lies strictly between soft T212 and soft T2 (Theorems 3.25, 3.27 and Examples 3.29, 3.30). Several sufficient conditions for the equivalence between these new three notions and some of their relevant ones are given (Theorems 2.13, 2.14, 3.5, 3.6, 3.26, 3.28). A decomposition theorem of soft regularity by means of soft ω-semi-regularity and soft ω -almost-regularity is given (Theorem 3.17). It is shown that soft ω -almost-regularity is heritable for specific kinds of soft subspaces (Theorems 2.19, 2.21). Soft product theorems regarding soft almost regular spaces (Theorem 2.23), soft ω-almost regular spaces (Theorem 2.24), and soft ω-semi-regular spaces (Theorem 3.33). Finally, the article delves into the connections between the newly proposed as well as some known soft axioms and their counterparts in traditional topological spaces, facilitating a bridging of concepts between the soft and classical realms (Theorems 2.3–2.7, 3.8–3.11, 3.13, 3.19–3.21, 3.23, and Corollaries 2.8, 2.10, 3.12, 3.14, 3.22, 3.24).

    In the next work, we intend to: 1) Define and investigate soft ω -almost-normality; 2) investigate the behavior of these new soft separation ideas under various kinds of soft mappings; and 3) find an application for our new two conceptions in the "decision-making problem", "information systems", or "expert systems".

    The authors declare that they have not used Artificial Intelligence tools in the creation of this article.

    The authors declare that they have no conflicts of interest.



    [1] M. Zhang, Y. Zhang, Q. Cen, S. Wu, Deep learning-based resource allocation for secure transmission in a non-orthogonal multiple access network, Int. J. Distr. Sensor Net., 18 (2022), 1975857866. https://doi.org/10.1177/15501329221104330 doi: 10.1177/15501329221104330
    [2] B. Bi, D. Huang, B. Mi, Z. Deng, H. Pan. Efficient LBS security-preserving based on NTRU oblivious transfer, Wireless Pers. Commun., 108 (2019), 2663–2674. https://doi.org/10.1007/s11277-019-06544-2 doi: 10.1007/s11277-019-06544-2
    [3] R. Bhanot, R. Hans, A review and comparative analysis of various encryption algorithms, Int. J. Secur. Its Appl., 9 (2015), 289–306. http://dx.doi.org/10.14257/ijsia.2015.9.4.27 doi: 10.14257/ijsia.2015.9.4.27
    [4] G. Sun, Y. Li, D. Liao, V. Chang, Service function chain orchestration across multiple domains: A full mesh aggregation approach, IEEE Tran. Network Service Manag., 15 (2018), 1175–1191. https://doi.org/10.1109/TNSM.2018.2861717 doi: 10.1109/TNSM.2018.2861717
    [5] J. Daemen, V. Rijmen, The design of Rijndael, New York: Springer-verlag, 2002. https://doi.org/10.1007/978-3-662-04722-4
    [6] E. Biham, A. Shamir, Differential cryptanalysis of DES-like cryptosystems, J. CRYPTOLOGY, 4 (1991), 3–72. https://doi.org/10.1007/BF00630563 doi: 10.1007/BF00630563
    [7] C. E. Shannon, Communication theory of secrecy systems, Bell Syst. Technical J., 28 (1949), 656–715. https://doi.org/10.1002/j.1538-7305.1949.tb00928.x doi: 10.1002/j.1538-7305.1949.tb00928.x
    [8] N. Siddiqui, A. Naseer, M. Ehatisham-ul-Haq, A novel scheme of substitution-box design based on modified Pascal's triangle and elliptic curve, Wirel. Personal Commun., 116 (2021), 3015–3030. https://doi.org/10.1007/s11277-020-07832-y
    [9] H. A. Ahmed, M. F. Zolkipli, M. Ahmad, A novel efficient substitution-box design based on firefly algorithm and discrete chaotic map, Neural Comput. Appl., 31 (2019), 7201–7210. https://doi.org/10.1007/s00521-018-3557-3 doi: 10.1007/s00521-018-3557-3
    [10] F. Masood, J. Masood, L. Zhang, S. S. Jamal, W. Boulila, S. U. Rehman, et al., A new color image encryption technique using DNA computing and Chaos-based substitution box, Soft Comput., 26 (2022), 7461–7477. https://doi.org/10.1007/s00500-021-06459-w
    [11] A. Razaq, M. Ahmad, A. Yousaf, M. Alawida, A. Ullah, U. Shuaib, A group theoretic construction of large number of AES-like substitution-boxes, Wirel. Personal Commun., 122 (2022), 2057–2080. https://doi.org/10.1007/s11277-021-08981-4 doi: 10.1007/s11277-021-08981-4
    [12] A. Razaq, S. Akhter, A. Yousaf, U. Shuaib, M. Ahmad, A group theoretic construction of highly nonlinear substitution box and its applications in image encryption, Multi. Tools Appl., 81 (2022), 4163–4184. https://doi.org/10.1007/s11042-021-11635-z doi: 10.1007/s11042-021-11635-z
    [13] F. Gonzalez, R. Soto, B. Crawford, Stochastic fractal search algorithm improved with opposition-based learning for solving the substitution box design problem, Mathematics, 10 (2022), 2172. https://doi.org/10.3390/math10132172 doi: 10.3390/math10132172
    [14] F. Artuğer, F. Özkaynak, SBOX-CGA: Substitution box generator based on chaos and genetic algorithm, Neural. Comput. App., 34 (2022), 20203–20211. https://doi.org/10.1007/s00521-022-07589-4
    [15] M. S. Fadhil, A. K. Farhan, M. N. Fadhil, Designing substitution box based on the 1D logistic map chaotic system, IOP Conf. Series: Mater. Sci. Eng., 1076 (2021), 012041. https://doi.org/10.1088/1757-899X/1076/1/012041 doi: 10.1088/1757-899X/1076/1/012041
    [16] A. Razaq, Iqra, M. Ahmad, M. A. Yousaf, S. Masood, A novel finite rings based algebraic scheme of evolving secure S-boxes for images encryption, Mult. Tools Appl., 80 (2021), 20191–20215. https://doi.org/10.1007/s11042-021-10587-8 doi: 10.1007/s11042-021-10587-8
    [17] I. Ullah, N. A. Azam, U. Hayat, Efficient and secure substitution box and random number generators over Mordell elliptic curves, J. Infor. Security Appl., 56 (2021), 102619. https://doi.org/10.1016/j.jisa.2020.102619 doi: 10.1016/j.jisa.2020.102619
    [18] Z. Hua, J. Li, Y. Chen, S. Yi, Design and application of an S-box using complete Latin square, Nonlinear Dyn., 104 (2021), 807–825. https://doi.org/10.1007/s11071-021-06308-3 doi: 10.1007/s11071-021-06308-3
    [19] A. A. A. El-Latif, J. Ramadoss, B. Abd-El-Atty, H. S. Khalifa, F. Nazarimehr, A novel chaos-based cryptography algorithm and its performance analysis, Mathematics, 10 (2022), 2434. https://doi.org/10.3390/math10142434 doi: 10.3390/math10142434
    [20] A. Razaq, G. Alhamzi, S. Abbas, M. Ahmad, A. Razzaque, Secure communication through reliable S-box design: A proposed approach using coset graphs and matrix operations, Heliyon, 9 (2023). https://doi.org/10.1016/j.heliyon.2023.e15902
    [21] M. A. Khan, A. Ali, V. Jeoti, S. Manzoor, A chaos-based substitution box (S-Box) design with improved differential approximation probability (DP), Iran. J. Sci. Technol. Trans. Electr. Eng., 42 (2018), 219–238. https://doi.org/10.1007/s40998-018-0061-9 doi: 10.1007/s40998-018-0061-9
    [22] F. Artuğer, F. Özkaynak, A novel method for performance improvement of chaos-based substitution boxes, Symmetry, 12 (2020), 571. https://doi.org/10.3390/sym12040571
    [23] A. Freyre-Echevarría, A. Alanezi, I. Martínez-Díaz, M. Ahmad, A. A. A. Abd El-Latif, H. Kolivand, et al., An external parameter independent novel cost function for evolving bijective substitution-boxes, Symmetry, 12 (2020), 1896. https://doi.org/10.3390/sym12111896
    [24] L. Chu, Y. Su, X. Zan, W. Lin, X. Yao, P. Xu, et al, A deniable encryption method for modulation-based DNA storage, Interdisciplinary Sciences: Comput. Life Sci., 16 (2024), 872–881. https://doi.org/10.1007/s12539-024-00648-5
    [25] X. Yao, R. Xie, X. Zan, Y. Su, P. Xu, W. Liu, A novel image encryption scheme for DNA storage systems based on DNA hybridization and gene mutation, Inter. Sci. Comput. Life Sci., 15 (2023), 419–432. https://doi.org/10.1007/s12539-023-00565-z doi: 10.1007/s12539-023-00565-z
    [26] S. Gao, R. Wu, X. Wang, J. Liu, Q. Li, X. Tang, EFR-CSTP: Encryption for face recognition based on the chaos and semi-tensor product theory, Inf. Sci., 621 (2023), 766–781. https://doi.org/10.1016/j.ins.2022.11.121 doi: 10.1016/j.ins.2022.11.121
    [27] S. Gao, H. H. C. Iu, J. Mou, U. Erkan, J. Liu, R. Wu, et al., Temporal action segmentation for video encryption, Chaos Solitons Fract., 183 (2024), 114958. https://doi.org/10.1016/j.chaos.2024.114958
    [28] S. Gao, H. H. C. Iu, M. Wang, D. Jiang, A. A. Abd El-Latif, R. Wu, et al., Design, hardware implementation, and application in video encryption of the 2-D memristive cubic map, IEEE Int. Things J., 11 (2024), 21807–21815. https://doi.org/10.1109/JIOT.2024.3376572
    [29] C. Fan, Q. Ding, A universal method for constructing non-degenerate hyperchaotic systems with any desired number of positive Lyapunov exponents, Chaos Solitons Fract., 161 (2022), 112323. https://doi.org/10.1016/j.chaos.2022.112323 doi: 10.1016/j.chaos.2022.112323
    [30] S. Gao, J. Liu, H. H. C. Iu, U. Erkan, S. Zhou, R. Wu, et al., Development of a video encryption algorithm for critical areas using 2D extended Schaffer function map and neural networks, Appl. Math. Model., 134 (2024), 520–537. https://doi.org/10.1016/j.apm.2024.06.016
    [31] M. Wang, X. Fu, L. Teng, X. Yan, Z. Xia, P. Liu, A new 2D-HELS hyperchaotic map and its application on image encryption using RNA operation and dynamic confusion, Chaos Solitons Fract., 183 (2024), 114959. https://doi.org/10.1016/j.chaos.2024.114959 doi: 10.1016/j.chaos.2024.114959
    [32] J. A. Gallian, Contemporary abstract algebra, Chapman and Hall/CRC, 2021. https://doi.org/10.1201/9781003142331
    [33] Q. Mushtaq, A. Razaq, Homomorphic images of circuits in PSL(2, Z)-space, Bull. Malaysian Math. Sci. Society, 40 (2017), 1115–1133. https://doi.org/10.1007/s40840-016-0357-8
    [34] Q. Mushtaq, A. Razaq, A. Yousaf, On contraction of vertices of the circuits in coset diagrams for PSL(2, Z), Proc. Math. Sci., 129 (2019), 1–26. https://doi.org/10.1007/s12044-018-0450-z
    [35] M. Conder, Three-relator quotients of the modular group, Quart. J. Math., 38 (1987), 427–447. https://doi.org/10.1093/qmath/38.4.427 doi: 10.1093/qmath/38.4.427
    [36] G. A. Jones, Maximal subgroups of the modular and other groups, J. Group Theory, 22 (2019), 277–296. https://doi.org/10.1515/jgth-2018-0144 doi: 10.1515/jgth-2018-0144
    [37] A. Razaq, Q. Mushtaq, A. Yousaf, The number of circuits of length 4 in PSL(2, ℤ)-space, Commun. Algebra, 46 (2018), 5136–5145. https://doi.org/10.1080/00927872.2018.1461880
    [38] I. Hussain, T. Shah, H. Mahmood, A projective general linear group based algorithm for the construction of substitution box for block ciphers, Neural. Comput. Appl., 22 (2013), 1085–1093. https://doi.org/10.1007/s00521-012-0870-0 doi: 10.1007/s00521-012-0870-0
    [39] A. Altaleb, M. S. Saeed, I. Hussain, M. Aslam, An algorithm for the construction of substitution box for block ciphers based on projective general linear group, AIP Adv., 7 (2017), 035116. https://doi.org/10.1063/1.4978264 doi: 10.1063/1.4978264
    [40] S. Farwa, T. Shah, L. Idrees, A highly nonlinear S-box based on a fractional linear transformation, Spr. Plus, 5 (2016), 1658. https://doi.org/10.1186/s40064-016-3298-7 doi: 10.1186/s40064-016-3298-7
    [41] J. Pieprzyk, G. Finkelstein, Towards effective nonlinear cryptosystem design, IEE Proc. E-Comput. Digital Tech., 135 (1988), 325–335. https://doi.org/10.1049/ip-e.1988.0044 doi: 10.1049/ip-e.1988.0044
    [42] A. F. Webster, S. E. Tavares, On the Design of S-boxes, Adv. Crypt.-CRYPTO'85 Proc., 218 (1986). https://doi.org/10.1007/3-540-39799-X_41
    [43] M. Matsui, Linear Cryptanalysis Method for DES Cipher, Adv. Crypt.-EUROCRYPT'93, 765 (1994). https://doi.org/10.1007/3-540-48285-7_33
    [44] U. Hayat, N. A. Azam, H. R. Gallegos-Ruiz, S. Naz, L. Batool, A truly dynamic substitution box generator for block ciphers based on elliptic curves over finite rings, Arabian J. Sci. Engineer., 46 (2021), 1–13. https://doi.org/10.1007/s13369-021-05666-9 doi: 10.1007/s13369-021-05666-9
    [45] S. Ibrahim, A. M. Abbas, Efficient key-dependent dynamic S-boxes based on permutated elliptic curves, Inf. Sci., 558 (2021), 246–264. https://doi.org/10.1016/j.ins.2021.01.014 doi: 10.1016/j.ins.2021.01.014
    [46] B. M. Alshammari, R. Guesmi, T. Guesmi, H. Alsaif, A. Alzamil, Implementing a symmetric lightweight cryptosystem in highly constrained IoT devices by using a chaotic S-box, Symmetry, 13 (2021), 129. https://doi.org/10.3390/sym13010129 doi: 10.3390/sym13010129
    [47] H. S. Alhadawi, M. A. Majid, D. Lambić, M. A. Ahmad, A novel method of S-box design based on discrete chaotic maps and cuckoo search algorithm, Multimed. Tools Appl., 80 (2021), 7333–7350. https://doi.org/10.1007/s11042-020-10048-8 doi: 10.1007/s11042-020-10048-8
    [48] M. Long, L. Wang, S-box design based on discrete chaotic map and improved artificial bee colony algorithm, IEEE Access, 9 (2021), 86144–86154. https://doi.org/10.1109/ACCESS.2021.3069965 doi: 10.1109/ACCESS.2021.3069965
    [49] R. Soto, B. Crawford, F. G. Molina, R. Olivares, Human Behaviour based optimization supported with self-organizing maps for solving the S-box design problem, IEEE Access, 9 (2021), 84605–84618. https://doi.org/10.1109/ACCESS.2021.3087139 doi: 10.1109/ACCESS.2021.3087139
    [50] W. Yan, Q. Ding, A novel S-box dynamic design based on nonlinear-transform of 1D chaotic maps, Electronics, 10 (2021), 1313. https://doi.org/10.3390/electronics10111313 doi: 10.3390/electronics10111313
    [51] P. Zhou, J. Du, K. Zhou, S. Wei, 2D mixed pseudo-random coupling PS map lattice and its application in S-box generation, Nonlinear Dyn., 103 (2021), 1151–1166. https://doi.org/10.1007/s11071-020-06098-0 doi: 10.1007/s11071-020-06098-0
    [52] S. S. Jamal, M. M. Hazzazi, M. F. Khan, Z. Bassfar, A. Aljaedi, Z. ul Islam, Region of interest-based medical image encryption technique based on chaotic S-boxes, Expert Syst. Appl., 238 (2024), 122030. https://doi.org/10.1016/j.eswa.2023.122030
    [53] M. Wang, H. Liu, M. Zhao, Construction of a non-degeneracy 3D chaotic map and application to image encryption with keyed S-box, Mult. Tools Appl., 82 (2023), 34541–34563. https://doi.org/10.1007/s11042-023-14988-9 doi: 10.1007/s11042-023-14988-9
    [54] A. Razaq, L. A. Maghrabi, M. Ahmad, Q. H. Naith, Novel substitution-box generation using group theory for secure medical image encryption in E-healthcare, AIMS Math., 9 (2024), 6207–6237. https://doi.org/10.3934/math.2024303 doi: 10.3934/math.2024303
    [55] K. Z. Zamli, F. Din, H. S. Alhadawi, Exploring a Q-learning-based chaotic naked mole rat algorithm for S-box construction and optimization, Neural Comput. Appl., 35 (2023), 10449–10471. https://doi.org/10.1007/s00521-023-08243-3 doi: 10.1007/s00521-023-08243-3
    [56] A. A. Alzaidi, M. Ahmad, H. S. Ahmed, E. A. Solami, Sine-cosine optimization-based bijective substitution-boxes construction using enhanced dynamics of chaotic map, Complexity, 2018 (2018), 1–16. https://doi.org/10.1155/2018/9389065 doi: 10.1155/2018/9389065
    [57] T. Farah, R. Rhouma, S. Belghith, A novel method for designing S-box based on chaotic map and teaching-learning-based optimization, Nonlinear Dyn., 88 (2017), 1059–1074. https://doi.org/10.1007/s11071-016-3295-y doi: 10.1007/s11071-016-3295-y
    [58] H.S. Alhadawi, D. Lambic, M.F. Zolkipli, M. Ahmad, Globalized firefly algorithm and chaos for designing substitution box, J. Inf. Security Appl., 55 (2020), 1–13. https://doi.org/10.1016/j.jisa.2020.102671 doi: 10.1016/j.jisa.2020.102671
    [59] F. Özkanak, A.B. Özer, A method for designing strong S-boxes based on chaotic Lorenz system, Phys. Lett. A, 374 (2010), 3733–3738. https://doi.org/10.1016/j.physleta.2010.07.019
    [60] Y. Aydin, A. M. Garipcan, F. Özkaynak, A novel secure S-box design methodology based on FPGA and SHA-256 hash algorithm for block cipher algorithms, Arab. J. Sci. Eng., 2024, 1–14. https://doi.org/10.1007/s13369-024-09251-8
    [61] I. Hussain, T. Shah, M.A. Gondal, H. Mahmood, Generalized majority logic criterion to analyze the statistical strength of S-boxes, Z Nat. A, 67 (2012), 282–288. https://doi.org/10.5560/zna.2012-0022 doi: 10.5560/zna.2012-0022
  • This article has been cited by:

    1. Tareq M. Al-shami, Abdelwaheb Mhemdi, Alaa M. Abd El-latif, Fuad A. Abu Shaheen, Finite soft-open sets: characterizations, operators and continuity, 2024, 9, 2473-6988, 10363, 10.3934/math.2024507
    2. Gözde Yaylalı Umul, A multi-attribute group decision-making algorithm based on soft intervals that considers the priority rankings of group members on attributes of objects, along with some applications, 2025, 10, 2473-6988, 4709, 10.3934/math.2025217
    3. Dina Abuzaid, Samer Al-Ghour, Supra soft Omega-open sets and supra soft Omega-regularity, 2025, 10, 2473-6988, 6636, 10.3934/math.2025303
  • Reader Comments
  • © 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(671) PDF downloads(32) Cited by(1)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog