An oldie but as there appear to be so many new posters thought...

An oldie but as there appear to be so many new posters thought it might help to revisit the first introduction BrainChip shareholders had to AERO the electronic nose based on AKIDA technology. This first implementation was on the Akida Development Environment which is in software and does not have as a result the full performance expected from the AKIDA1000 system on a chip once it has been tested successfully and released. I have extracted the opening and the concluding paragraphs only:

A Hardware-Deployable Neuromorphic Solution for Encoding and Classification of Electronic Nose DataArticle (PDF Available) in Sensors

Anup Vanarse11.7Edith Cowan University
Adam Osseiran21.17Edith Cowan University
Alexander Rassau18.28Edith Cowan University
Peter van der Made2.81BrainChip

Abstract
In several application domains, electronic nose systems employing conventional data processing approaches incur substantial power and computational costs and limitations, such as significant latency and poor accuracy for classification. Recent developments in spike-based bio-inspired approaches have delivered solutions for the highly accurate classification of multivariate sensor data with minimized computational and power requirements. Although these methods have addressed issues related to efficient data processing and classification accuracy, other areas, such as reducing the processing latency to support real-time application and deploying spike-based solutions on supported hardware, have yet to be studied in detail. Through this investigation, we proposed a spiking neural network (SNN)-based classifier, implemented in a chip-emulation-based development environment, that can be seamlessly deployed on a neuromorphic system-on-a-chip (NSoC). Under three different scenarios of increasing complexity, the SNN was determined to be able to classify real-valued sensor data with greater than 90% accuracy and with a maximum latency of 3 s on the software-based platform. Highlights of this work included the design and implementation of a novel encoder for artificial olfactory systems, implementation of unsupervised spike-timing-dependent plasticity (STDP) for learning, and a foundational study on early classification capability using the SNN-based classifier.

5. Conclusions
In this paper, we presented the development of an SNN-based solution for real-time classificationof electronic olfactory data. The SNN was implemented using Brainchip’s Akida DevelopmentEnvironment, which provides an emulation of the Akida NSoC’s functionalities on a Python-basedSensors 2019,19, 4831 11 of 13software platform. The highlights of this implementation included the development of a novelAER-based encoder for olfaction data, implementation of unsupervised STDP for training the SNN,highly accurate real-time classification results, and preliminary results that lay a foundation forapplying the Akida SNN for reliable early classification results.One of the most significant contributions of this study is the development of AERO, an AER-baseddata-to-spike encoder for olfaction data. The operating principle of AERO is based on discretizing thesensor responses and encoding their activation levels along with sensor ID and temporal data. Forthis implementation, normalized relative resistance features were extracted from sensor responsesand provided as an input to the AERO encoder. Parameters, such as frequency of event-generation,the number of time points for encoding, and discretization levels, can be configured based on the inputdata and processing requirements. The development of AERO has opened several avenues for futureresearch, such as encoding multi-dimensional data using different features and interfacing of electronicnose systems with AER-based neuromorphic hardware for processing.The SNN was tested using the benchmark dataset [20] under four different scenarios of increasingcomplexity. In general, under each scenario, the classification performance of the SNN was between 90%and 100%, and the processing latency was between 2.5 and 3 s, which includes data-to-spike encoding,learning, classification, and other software-based latencies introduced due to looping and conditionalstatements. This processing latency would, of course, be dramatically reduced once the classifier isimplemented on the Akida NSoC hardware without the overhead of software emulation. Taken together,these classification results show that the SNN-based classifier can deliver highly accurate results withminimal processing latency. Moreover, the ability to transfer the SNN implementation to the AkidaNSoC can be leveraged to develop low-power electronic nose systems with minimal computationalcost and memory requirements. Intrinsically, in most cases, neuromorphic approaches have proven tooutperform traditional processing methods that suffer from limited accuracy, high computational andpower requirements, and substantial latency to provide classification results [32]. When evaluatedagainst other neuromorphic and traditional approaches based on the same dataset [21,24,33,34],the results revealed that the SNN classifier developed in this study achieved comparable and, inmost cases, better classification performance with minimal computation requirements and latencyfor both learning and processing. More importantly, the Akida SNN was able to identify patternsfrom a highly multi-dimensional dataset and classify the dataset based on the four chemical groups ofthe compounds.Future research based on these results will focus on the development of a robust SNN-basedclassifier on the Akida NSoC and its implementation in a real-world application. The efficacy of AEROwhen combined with rate coding methods, such as [35] and rank-order encoding [2,31,36], will alsobe investigated. This implementation also lays the foundation for the application of both the AEROencoder and an SNN to study neuromorphic gustation and the fusion of olfaction and gustation for thedevelopment of a comprehensive analytical tool for chemical sensing.