DOAJ is a unique and extensive index of diverse open access journals from around the world, driven by a growing community, committed to ensuring quality content is freely available online for everyone.

RFID and wireless sensor networks exemplify computationally constrained environments, where the compact nature of the components cannot support complex computations or high communication overhead. On the other hand, such components should support security applications such as message integrity, authentication, and time stamping. The latter are efficiently implemented by hash message authentication codes (HMAC). As clearly stated in the literature, current approved implementations of HMAC require resources that cannot be supported in constrained components. An approach to implement a compact HMAC by the use of stream ciphering is presented in this paper.

Certification in public key cryptographic applications concerns the involvement of a CA (Certifying Agent) in approving the validity of users' public keys. Distributed certification pertains to the case where several CAs are involved in issuing certificates. This also includes the case of multi-attribute certification, where different CAs approve different user's attributes. In delegated certification, agents transfer certificate issuing authority in hierarchical chain. Distributed, multi-attribute and delegated certification techniques having a low computational complexity are presented in this paper. It is shown how multiplicity aspects of the various applications are mapped into a multiplied exponents mathematical implementation of the form Pii=1mAibimod p, which is essentially equivalent to a single exponentiation for a moderate m. A fundamental feature of the presented procedures pertains to implementing distributed/multi-attribute certification by referring to any desired subset of participating CAs.

Based on its definition, it appears that primality testing should inherently involve a division operation. Even the popular primality testing procedure which involves modular exponentiations has a division essence. Montgomery arithmetic facilitates modular exponentiation without divisions, provided that certain precalculations which involve proper divisions are first performed. In primality testing these precalculations have to be re-done for each number whose candidacy for primality is tested. It is shown in this paper how primality testing can still be based on Montgomery arithmetic, such that no operation, including the precalculation, involves proper divisions. That is, it is shown that a complete primality testing process can be based on divisionless operations only.

The author considers the case where double-precision modular multiplications need to be performed by a device containing a custom IC that provides for a single-precision modular multiplication. The method presented is obviously extended to multiplying (k*n)-bit operands using an n-bit modular multiplier, for the case where k is a power of 2. Whereas multiprecision modular multiplication is simply executed using an available single-precision modular multiplier which provides external division services, the existence of these services is not assumed in the process presented, as the operation carried by the available modular multiplier is not necessarily based on proper division. The process is controlled by a standard CPU which also performs some simple calculations.>

Two exponentiation circuits are proposed. Using the fact that squaring is a linear operation over GF(2/sup n/), a time-space tradeoff in smartcard-based circuitry is presented. It is shown how multiplication is performed by a single shift, based on replacing the public key alpha /sup a/ in GF(2/sup n/) by its minimal polynomial. Other considerations, related to structure regularity and the possible use of dynamic shift registers, are also treated.>

In a standard binary search, the binary representation of the index of an element in an ordered linear array is recovered serially bit by bit. For an array of N elements, the index of an element is recovered, in principle, by assigning to each element one value out of log/sub 2/ N possibilities. It is shown here that by arranging 2/sup n/-1 elements in a circular array, the bits of the binary representation of the index of an element are all recovered simultaneously based n assigning to each element one value out of two possibilities. The main theoretical result shows that the parity of an integer X is trivially recovered from the parity of the Hamming weight of the binary representation of X, X+1, X+2, and X+3, whereas, on the other hand, the parity of the Hamming weight of the binary representation of an integer is consistent with modular arithmetic considerations.>

The flat concurrent Prolog (FCP) language, which enables the implementation of synchronization through data flow, communication through shared variables, parallelism through the reduction of the AND components in the clause's body, and indeterminacy through guarded commands, is discussed. It is shown that FCP, used in conjunction with Petri net methods, forms a powerful tool in the modeling of parallel systems that involve concurrency. The simulation of systems represented by various types of Petri nets is achieved using the reduction process of FCP. AND parallelism supports free competition among conflicting enabled transitions. A structural analysis of systems is provided using the metaprogramming technique.>

The possibility of developing hardware description languages (HDLs) based on the principles of logic programming is discussed. The specific logic programming language used to demonstrate this possibility is Flat Concurrent Prolog (FCP). It is shown explicitly how FCP naturally satisfies the commonly accepted fundamental requirements of a hardware description language. It is then demonstrated how FCP overcomes known disadvantages of the highly acclaimed VHDL. Some other parallel logic programming languages beside FCP are also presented briefly, and the possibility of using them for hardware description is discussed.>

The possibility of integrating human visual intelligence into the process of encrypting sensitive information by presenting certain visual information to the recipient's eye is discussed. This adds a new dimension to the cryptocomplexity of such a process. Two implementations are based on this principle are described. The first shows how keys used for encryption can be randomly generated by the transmitter, without the necessity of exchanging them with the legitimate recipient. The keys are 'embedded' in a master key and are recovered from it by the intelligence of the legitimate recipient after he or she uses the master key. No human intelligence can be helpful to a user who does not possess the master key. The second implementation concerns the possibility of creating a secret connection between a numerical key and a specific image (e.g. a face). Such a scheme can be used, for example, in validating the identity of the users of credit cards.>

In this paper we consider the problem of assigning a numerical value of each of N elements which are arranged in a linear array, for the purpose of being able to recover the location of any specified element (with respect to the first element), with this aim being achieved using a small number of possible different assigned values. An obvious solution is using a binary search approach, characterized by the features: (a) At least $[\log _2 N] + 2$ different values are needed for implementing the process. (b) The bits forming the binary representation of the specified location are recovered sequentially (starting with the least significant bit). A new approach in assigning values to elements, for the purpose of recovering efficiently their location, is presented in this paper, for the case where N is of the form $2^n - 1$. Its main features are (a) less than $[\log _2 N] + 2$ different values are sufficient for achieving this task (even 2 are enough!). (b) The bits forming the binary representation of the specified location are recovered in parallel. This is achieved by very simple means, where the value of each bit in the binary representation is recovered by considering the numerical values assigned to 4 elements having fixed location with respect to the element whose location is to be found.

A prototype for a binary adder, in which the carry propagation is handled by light signals is suggested, thus enabling a fast execution of addition operation even for large operands.

A handwriting sample is identified by comparing its features to the corresponding features of a list of handwriting samples contained in a reference library. Some of the needed features are extracted from the structure of specific parts of the handwriting's sample which have to be identified first (e.g., specific letters). Other features are extracted from the external structure of the handwriting (e.g., margins). Due to their described property, these features are extracted automatically with relative ease. Techniques for extracting features of the described kind are presented in this work. Experimental results obtained when using the extracted features for identification purposes are then given.

A public key cryptosystem is suggested in which a message is enciphered by adding its numerical value to a sum of elements selected randomly from a publicly known knapsack, thus enabling multiple mapping of the message. Data expansion is moderate while reasonable security is maintained. Since the selected knapsack elements are message-independent, the encryption is very fast.

A method is presented for displaying the time-varying transfer characteristic of communication channels, and is illustrated by showing in detail the behavior of an HF-SSB radio channel during one minute periods, under the various fading characteristics classified by USAEL. It is also demonstrated how this method aided study of the causes of errors in data received by digital modems.

A known way of scrambling binary data for security purposes is by adding to it, bit by bit, a pseudorandom sequence A. Synchronization is established by providing the descrambler with the states of A's generator. It is shown here how for data transmitted in blocks, the states of the generator can be recovered at any instant which enables self-synchronization. If the block length iskand the generator of A consists of ann-stage register, there is a delay ofk(n - 1) + 1bits, between sync lost and sync recovery.

The repeat request codeword in an ARQ data transmission can be the detected error syndrome. In that case, there exists an optimal single-burst detecting scheme which improves the throughput of a Stop and Wait scheme, especially under high error rate conditions.

An optimal coding scheme is offered for binary messages with a limited number of successive runs of length 1 (i.e., strings of the form 010101..., in the messages, have a known limited length). A practical example is given of a case where such messages are found.

A simple connection is shown to exist between the Hadamard transforms of a discrete signalAand a signalBobtained fromAin three different ways: a) time scaling ofA; b) expandingAby means of zero insertion; and c) lettingArepeat itself. Both one-dimensional and two-dimensional cases are investigated.