Savoir faire et faire savoir: what I learned about innovation while I thought I was doing a Ph.D.

This short text is a reflection of lessons learned about innovation during my studies at the signal processing laboratory of the Swiss Federal Institute of Technology, while I thought I was studying video segmentation based on multiple features for interactive multimedia applications.     

A self-immune to 3D masks attacks face recognition system

Due to their advantages, facial recognition systems are among the most widely used biometric systems in recent years. However, their drawback is that they can easily be deceived by using 3D masks, which are replicas of real faces. To confirm this fact, we have tested the vulnerability to 3D masks attacks of the already approved Legendre moments invariants (LMI)-based face recognition method. This has been achieved by using the 3D mask attack database (3DMAD), which consists of real faces and faces with 3D masks. The obtained spoof false acceptance rate (SFAR) was close to 65%, which proves that this recognition system is vulnerable to 3D masks attacks. This is generally the case of other face recognition systems, with no anti-spoofing provision. In this paper, a face recognition method is proposed to prevent hackers from deceiving face recognition systems by using 3D masks of people belonging to the system database. This method combines the LMI and the linear discriminant analysis for characteristic features extraction, and the maximum likelihood for classification. With a 97.6% recognition rate and a SFAR equal to 0.83%, the results obtained may be considered as very satisfactory. These results, while obtained with a lower computational time, compare favorably with those of the state-of-the-art method that uses the same 3DMAD database.     

Optical waveguides in fluoride lead silicate glasses fabricated by carbon ion implantation

The carbon ion implantation with energy of 4.0 MeV and a dose of 4.0×1014 ions/cm2 is employed for fabricating the optical waveguide in fluoride lead silicate glasses. The optical modes as well as the effective refractive indices are measured by the prism coupling method. The refractive index distribution in the fluoride lead silicate glass waveguide is simulated by the reflectivity calculation method (RCM). The light intensity profile and the energy losses are calculated by the ?nite-difference beam propagation method (FD-BPM) and the program of stopping and range of ions in matter (SRIM), respectively. The propagation properties indicate that the C2+ ion-implanted fluoride lead silicate glass waveguide is a candidate for fabricating optical devices.     

Completeness for Symmetric Two-Party Functionalities: Revisited

Understanding the minimal assumptions required for carrying out cryptographic tasks is one of the fundamental goals of theoretic cryptography. A rich body of work has been dedicated to understanding the complexity of cryptographic tasks in the context of (semi-honest) secure two-party computation. Much of this work has focused on the characterization of trivial and complete functionalities (resp., functionalities that can be securely implemented unconditionally, and functionalities that can be used to securely compute all functionalities). Most previous works define reductions via an ideal implementation of the functionality; i.e., f reduces to g if one can implement f using a black-box (or oracle) that computes the function g and returns the output to both parties. Such a reduction models the computation of f as an atomic operation. However, in the real world, protocols proceed in rounds, and the output is not learned by the parties simultaneously. In this paper, we show that this distinction is significant. Specifically, we show that there exist symmetric functionalities (where both parties receive the same outcome) that are neither trivial nor complete under “black-box reductions,” and yet the existence of a constant-round protocol for securely computing such a functionality implies infinitely often oblivious transfer (meaning that it is secure for infinitely many values of the security parameter). In light of the above, we propose an alternative definitional infrastructure for studying the triviality and completeness of functionalities.     

A novel proposal for all-optical Galois field adder based on photonic crystals

In this paper we are going to propose an all-optical structure for implementing Galois field adder. To do so, we will use four optical XOR gates. The working principle of the proposed structure is based on destructive interference of optical waves. By choosing different lengths for the input waveguides, 180\(^\circ \) of phase difference will be generated between the optical waves. In the final structure, the normalized power for logic 0 and 1 at the output ports was 1 and 45%. Time delay of the proposed structure is about 1.5 ps.     

Visible Light Communications (VLC) for Ambient Assisted Living

With the advent of high efficacy light emitting diode directional lamps as a key component in focal lighting, new possibilities emerge for re-designing the smart home scenario. A smart home scenario is characterized by enabled intelligent interworking of various wireless and wired technologies to provide inhabitants with ease of use of appliances, while creating a personalized and safe ambience space. More and more high and low data rate circulates within the indoor ambient space (e.g., home, hospitals, offices). Although, unlicensed technologies, such as wireless local area networks can take upon part of the indoor traffic, the ever increasing demand for such data, and users, calls for either use of licensed or novel unlicensed wireless communication technologies as part of the smart home enablers. This paper focuses on the potentials of visible light communications (VLCs), jointly with radio and fiber communications, to support very dense low and high data rate connectivity, while enabling deployment of secure-sensitive indoor applications, including indoor tracking and localization. The paper proposes a scenario for integrating VLC into the smart home scenario and a conceptual supporting architecture for its deployment. Further, the technical challenges and possible roadmap for the actual deployment are analyzed for the particular case of an eHealth scenario where the utilization of VLC technology is the enabler of the cost-efficient rollout of the required infrastructure and thus the game-changer in a multi-billion eHealth niche that is seeking for cost affordable solutions.     

DTN routing using explicit and probabilistic routing table states

Routing in communication networks involves the indirection from a persistent name (ID) to a locator. The locator specifies how packets are delivered to a destination with a particular ID. Such a mapping is provided by a routing table entry, i.e. state. In a DTN, it is hard to maintain routing state because intermittent connectivity prevents protocols from refreshing states when they become inaccurate. In prior work, per-destination state mostly corresponds to utilities, where a high utility value about a destination implies that the probability to encounter the destination for the node maintaining the state is high. This approach depends on a particular mobility pattern in which nodes that met frequently in the past are likely to encounter in the future. In this paper, we use the concept of weak state that does not rely on external messages to remain valid (Acer et al. in MobiCom ’07: proceedings of the 13th annual ACM international conference on mobile computing and networking, pp 290–301, 2007). Our weak state realization provides probabilistic yet explicit information about where the destination is located. We build Weak State Routing protocol for Delay Tolerant Networks (WSR-D) that exploits the direction of node mobility in forwarding. It provides an osmosis mechanism to disseminate the state information to the network. With osmosis, a node has consistent information about a portion of the nodes that are located in regions relevant to its direction of mobility. Through simulations, we show that WSR-D achieves a higher delivery ratio with smaller average delay, and reduces the number of message transfers in comparison to Spray & Wait (Spyropoulos et al. in Proceedings of ACM SIGCOMM 2005 workshops: conference on computer communications, pp 252–259, 2005) and Spray & Focus (Spyropoulos et al. in IEEE/ACM Trans Netw, 16(1):77–90, 2008), a stateless and a utility based protocol, respectively.     

Cooperative Diversity Based Multi-Carrier CDMA System with Decode and Forward Relays in Rayleigh Fading channels

Cooperative communication is one of the major diversity techniques which exploit spatial diversity through a virtual antenna array. In addition to cooperation, transmitting the same symbols through different sub carriers (Multi-Carriers) introduces frequency diversity as well. The objective of this paper is to investigate cooperative diversity performance in Multi-Carrier Code Division Multiple Access (MC-CDMA) systems with orthogonal and non-orthogonal spreading codes. We evaluate the Symbol Error Rate (SER) of cooperative transmission in frequency-selective Rayleigh faded uplink channel. We focus on the use of despreading-combining (DC) receiver and coherent detection with Maximal Ratio Combining for the above selected criteria respectively. It is shown that the orthogonal spreading code cancels Multiple-Access Interference (MAI) and the performance is completely independent of the length of the spreading code. The approximate SER derived here for the the transmission scheme is well matched with the increasing number of users in the system. Hence the Gaussian approximation is acceptable for systems which are operating with their maximum user capacities at lower SNR values.     

Electromigration studies of flip chip Sn95/Sb5 solder bumps on Cr/Cr-Cu/Cu under-bump metallization

The electromigration-induced failure of Sn95/Sb5 flip chip solder bumps was investigated. The failure of the joints was found at the cathode/chip side after current stressing with a density of 1×10⁴ A/cm² at 150°C for 13 sec. The growth of intermetallic compounds (IMCs) was observed at the anode side after current stressing. Voids were found near the current crowding area in the cathode/chip side, and the (Cu,Ni)₆Sn₅ IMC at the cathode/chip end was transformed into the Sn phase. The failure mechanism for Sn95/Sb5 flip chip solder joint is proposed in this paper.     

On the path-loss attenuation regime for positive cost and linear scaling of transport capacity in wireless networks

Wireless networks with a minimum inter-node separation distance are studied where the signal attenuation grows in magnitude as 1/ρ/sup δ/ with distance ρ. Two performance measures of wireless networks are analyzed. The transport capacity is the supremum of the total distance-rate products that can be supported by the network. The energy cost of information transport is the infimum of the ratio of the transmission energies used by all the nodes to the number of bit-meters of information thereby transported. If the phases of the attenuations between node pairs are uniformly and independently distributed, it is shown that the expected transport capacity is upper-bounded by a multiple of the total of the transmission powers of all the nodes, whenever δ>2 for two-dimensional networks or δ>5/4 for one-dimensional networks, even if all the nodes have full knowledge of all the phases, i.e., full channel state information. If all nodes have an individual power constraint, the expected transport capacity grows at most linearly in the number of nodes due to the linear growth of the total power. This establishes the best case order of expected transport capacity for these ranges of path-loss exponents since linear scaling is also feasible. If the phases of the attenuations are arbitrary, it is shown that the transport capacity is upper-bounded by a multiple of the total transmission power whenever δ>5/2 for two-dimensional networks or δ>3/2 for one-dimensional networks, even if all the nodes have full channel state information. This shows that there is indeed a positive energy cost which is no less than the reciprocal of the above multiplicative constant. It narrows the transition regime where the behavior is still open, since it is known that when δ<3/2 for two-dimensional networks, or δ<1 for one-dimensional networks, the transport capacity cannot generally be bounded by any multiple of the