{"title":"Exploration of Edge of Chaos in Bio-Inspired Devices, Circuits, and Systems","authors":"A. Ascoli, A. S. Demirkol, R. Tetzlaff, L. Chua","doi":"10.1109/CNNA49188.2021.9610770","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610770","url":null,"abstract":"With Moore's era approaching an inevitable end, much research is currently focused on multi-purpose nano-devices, which may endow traditional purely-CMOS circuits with add-on functionalities, allowing to boost the performance of integrated circuits, despite further reductions in transistor dimensions shall no longer be viable. One of the most promising nanotechnologies for fostering progress in electronics in the years to come allows to fabricate memristors, which, depending upon their constitutive materials, may pave the way toward the circuit realisation of disruptive in-memory computing or mem-computing paradigms of great appeal to the Internet-of- Things industry. Moreover, with their extraordinary capability to capture the rich dynamics of neural structures, memristors shall play a key role in the development of miniaturised hardware systems operating according to principles similar to the mechanisms underlying the functionalities of the human brain. In this regard this work sheds light into the conditions, under which electronic systems, leveraging the locally-active behaviour of a NbO device stack, may enter a particular operating regime, hosting the seed for complexity, and referred to as Edge of Chaos, similarly as axon membranes on the verge to generate all-or-none spikes, which, travelling across neural structures, enable the development of intelligence in human beings.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131773696","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Towards 3D Cave Mapping with UAVs","authors":"Hunor Laczkó, Bálint Jánossy, T. Zsedrovits","doi":"10.1109/CNNA49188.2021.9610810","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610810","url":null,"abstract":"Cave mapping is far beyond surface mapping technologies because of the rugged terrain and the limited availability of sensors (for example, no GPS). Most of the existing maps were hand-drawn and are inaccurate. The mapping technologies are outdated or require expensive tools, or they are too complicated to use by anyone. For this, we propose a drone-based system where with the use of its basics sensors and laser range finders, and monocular cameras, we will be able to create 3D maps in real-time.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122647207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Irene Mazzilli, Gianmario Mirabile, P. Lino, G. Maione, A. Rybakov, N. Svishchev, Ileana Blanco, L. De Bellis, A. Luvisi
{"title":"UAV Inspection of Olive Trees for the Detection of Xylella Fastidiosa Disease Using Neural Networks","authors":"Irene Mazzilli, Gianmario Mirabile, P. Lino, G. Maione, A. Rybakov, N. Svishchev, Ileana Blanco, L. De Bellis, A. Luvisi","doi":"10.1109/CNNA49188.2021.9610752","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610752","url":null,"abstract":"This paper presents a fully automated procedure for the detection of trees affected by Xylella Fastidiosa using UAVs and convolutional neural networks. Drones are able to collect an adequate number of olive leaf images to detect the presence of disease symptoms. Several neural networks are trained to compare results and determine the best solution.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131306188","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Making Real Memristive Processing-in-Memory Faster and Reliable","authors":"Shahar Kvatinsky","doi":"10.1109/CNNA49188.2021.9610786","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610786","url":null,"abstract":"Memristive technologies are attractive candidates to replace conventional memory technologies, and can also be used to perform logic and arithmetic operations using a technique called ‘statefullogic.’ Combining data storage and computation in the memory array enables a novel non-von Neumann architecture, where both the operations are performed within a memristive Memory Processing Unit (mMPU). The mMPU relies on adding computing capabilities to the memristive memory cells without changing the basic memory array structure. The use of an mMPU alleviates the primary restriction on performance and energy in a von Neumann machine, which is the data transfer between CPU and memory. Here, the various aspects of mMPU are discussed, including its architecture and implications on the computing system and software, as well as examining the micro architectural aspects. We show how mMPU can be improved to accelerate different applications and how the poor reliability of memristors can be improved as part of the mMPU operation.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114573136","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Theodoros Panagiotis Chatzinikolaou, Iosif-Angelos Fyrigos, V. Ntinas, Stavros Kitsios, P. Bousoulas, Michail-Antisthenis I. Tsompanas, D. Tsoukalas, G. Sirakoulis
{"title":"Memristive Oscillatory Networks for Computing: The Chemical Wave Propagation Paradigm","authors":"Theodoros Panagiotis Chatzinikolaou, Iosif-Angelos Fyrigos, V. Ntinas, Stavros Kitsios, P. Bousoulas, Michail-Antisthenis I. Tsompanas, D. Tsoukalas, G. Sirakoulis","doi":"10.1109/CNNA49188.2021.9610785","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610785","url":null,"abstract":"During the last decade, there is an ever-growing concern regarding the future of CMOS technology, as well as the emerging difficulties on handling upcoming technological issues related with silicon transistors' dimensions, electrical power, energy consumption, and last but not least reaching the physical limits of this technology. At the same time, new computing alternatives beyond the classical computing systems, namely von Neumman architectures, are heavily sought after to tackle energy and memory-wall problems. In this talk, we focus on a hybrid analogue computational circuit-level system with unipolar memristor nanodevices connected in oscillatory networks and based on wave-like propagation of information. These methods are inspired by biochemical processes occurring in nature. The proposed insightful electrochemical wave propagation is apparent in many natural and biological systems and is modelled with powerful, inherently parallel computational tools, like Cellular Automata (CAs). This framework enables us to further proceed into realising alternative types of computations executed on the designed, modelled and fabricated memristor nanodevices, which are finally employed for the design and development of wave based electronic computational units. The proposed nanoelectronic memristive oscillatory networks will be in the advantageous position to perform both classical and unconventional calculations, like multi-digit, in memory and neuromorphic, to name a few of them. Thus, we will have a powerful tool targeting beyond the existing von Neumann information processing techniques and alleviating the aforementioned disadvantages associated with them.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133795463","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Defect-Resilient Technique of Memristor Crossbar with Large On-Off Ratio for Implementing HTM Spatial Pooler in Near-IoT-Sensor Cognition","authors":"Tien Van Nguyen, Jiyong An, Seokjin Oh, K. Min","doi":"10.1109/CNNA49188.2021.9610755","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610755","url":null,"abstract":"In this paper, we discuss memristor crossbars with large On-Off ratio are better for improving the read voltage margin, but, unfortunately, more vulnerable to memristor defects. For overcoming this problem using a defect-resilient technique, we explain and review the self-repairable boost-factor circuit presented previously. Based on the technique, we expect HTM spatial pooler with defect-resilient memristor crossbar can be suitable for realizing near-IoT-sensor cognitive hardware in future.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"112 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121131358","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Synthesis of Three-Layer Dynamic Binary Neural Networks for Control of Hexapod Walking Robots","authors":"Takumi Suzuki, Toshimichi Saito","doi":"10.1109/CNNA49188.2021.9610809","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610809","url":null,"abstract":"Three-layer dynamic binary neural networks are characterized by binary connection parameters and signum activation function. Depending on the parameters, the networks can generate various periodic orbits. First, as a basic theory, we give a synthesis method that guarantees storage and global stability of desired periodic orbits. Second, as an engineering application, we synthesize networks that generate periodic orbits corresponding to typical walking patterns of hexapod robots. Presenting an FPGA based hardware, switching of the walking patterns is confirmed experimentally.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128281665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A bio- inspired locomotion control approach through synchronization of embodied neural oscillators","authors":"P. Arena, A. Spinosa, Giuseppe Sutera, L. Patané","doi":"10.1109/CNNA49188.2021.9610804","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610804","url":null,"abstract":"The realization of a bio-inspired central pattern generator for legged robots passes through the definition of the basic neural units and their interconnections, designed to generate stable locomotion gaits. In this work, the interplay between the neural units and the actuators is investigated with the aim to integrate the two components realizing an embodied neuron. A nullcline-based control strategy is applied to modulate the neuron oscillatory behaviour for the generation of adaptive locomotion gaits.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121409180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"2D Material Memristor Devices for Neuromorphic Computing","authors":"Dafydd Ravenscroft, L. Occhipinti","doi":"10.1109/CNNA49188.2021.9610802","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610802","url":null,"abstract":"Neuromorphic computing offers to overcome the shortcomings associated with traditional von Neumann architectures. These suffer from bottlenecks caused by the separation of memory and processing, leading to large latency and high power consumption. One of the leading methods of achieving this is using memristors. The mass development of memristors presents a number of challenging problems which may be addressed through the use of 2D materials which offer a range of unique properties. In this work we explore the different 2D materials which demonstrate potential to be used as memristors. We explore some applications of such memristors and assess the challenges and future directions this area could develop.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126443952","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Novel Hardware-Efficient Cochlea Model based on Asynchronous Cellular Automaton Dynamics: Two-Tone Suppression and FPGA Implementation","authors":"I. Kubota, H. Torikai","doi":"10.1109/CNNA49188.2021.9610787","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610787","url":null,"abstract":"In this paper, a novel cochlea model based on the nonlinear dynamics of an asynchronous cellular automaton is presented. It is shown that the presented model has frequency selectivity with respect to a single-tone sound stimulation. Furthermore, it is shown that the presented model exhibits two-tone suppression, which is one of the most typical nonlinear responses of biological cochleae. In addition, the presented model is implemented in a field programmable gate array. It is shown that the presented model can be implemented by fewer hardware resources and lower power compared to a typical differential equation cochlea model.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132510394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}