When building a unified vision system or gradually adding new capabilities to a system, the usual assumption is that training data for all tasks is always available. However, as the number of tasks grows, storing and retraining on such data becomes infeasible. A new problem arises where we add new capabilities to a Convolutional Neural Network (CNN), but the training data for its existing capabilities are unavailable. We propose our Learning without Forgetting method, which uses only new task data to train the network while preserving the original capabilities. Our method performs favorably compared to commonly used feature extraction and fine-tuning adaption techniques and performs similarly to multitask learning that uses original task data we assume unavailable. A more surprising observation is that Learning without Forgetting may be able to replace fine-tuning with similar old and new task datasets for improved new task performance.

Learning without Forgetting

WiFi localization, the task of determining the physical location of a mobile device from wireless signal strengths, has been shown to be an accurate method of indoor and outdoor localization and a powerful building block for location-aware applications. However, most localization techniques require a training set of signal strength readings labeled against a ground truth location map, which is prohibitive to collect and maintain as maps grow large. In this paper we propose a novel technique for solving the WiFi SLAM problem using the Gaussian Process Latent Variable Model (GPLVM) to determine the latent-space locations of unlabeled signal strength data. We show how GPLVM, in combination with an appropriate motion dynamics model, can be used to reconstruct a topological connectivity graph from a signal strength sequence which, in combination with the learned Gaussian Process signal strength model, can be used to perform efficient localization.

WiFi-SLAM Using G aussian Process Latent Variable Models

The Wireless sensor network play a vital role in collecting a Real – Time data, monitoring environmental conditions based on technology adoption. These sensor network is the combination of sensing, computation, and communication through a single tiny device. Here many tiny nodes assemble and configure themselves. It also controls actuators that extend control from cyberspace into the physical world. Here the sensor nodes communicate with the local peers rather than the high – power control tower or base station. Instead, of relying on a predeployed infrastructure, each individual sensor or actuator become part of the overall infrastructure. Here we have three hardware platforms that addresses the needs of wireless sensor netwoks. The operating system here uses an event based execution to support concurrency. The platform serves as a baseline and does not contain any hardware accelerators. . First platform serves as a baseline and it produces Operating system concepts for refining concurrency mechanisms. The second node validates the architectural designs and improve the communicational rates. The third node represents the full realization of the general architecture. Keywords— node, platform, concurrency.

A Wireless Sensor Network.

The operational efficiency of remote environmental wireless sensor networks (EWSNs) has improved tremendously with the advent of Internet of Things (IoT) technologies over the past few years. EWSNs require elaborate device composition and advanced control to attain long-term operation with minimal maintenance. This article is focused on power supplies that provide energy to run the wireless sensor nodes in environmental applications. In this context, EWSNs have two distinct features that set them apart from monitoring systems in other application domains. They are often deployed in remote areas, preventing the use of mains power and precluding regular visits to exchange batteries. At the same time, their surroundings usually provide opportunities to harvest ambient energy and use it to (partially) power the sensor nodes. This review provides a comprehensive account of energy harvesting sources, energy storage devices, and corresponding topologies of energy harvesting systems, focusing on studies published within the last 10 years. Current trends and future directions in these areas are also covered.

Energy Harvesting Sources, Storage Devices and System Topologies for Environmental Wireless Sensor Networks: A Review.

Condition monitoring can reduce machine breakdown losses, increase productivity and operation safety, and therefore deliver significant benefits to many industries. The emergence of wireless sensor networks (WSNs) with smart processing ability play an ever-growing role in online condition monitoring of machines. WSNs are cost-effective networking systems for machine condition monitoring. It avoids cable usage and eases system deployment in industry, which leads to significant savings. Powering the nodes is one of the major challenges for a true WSN system, especially when positioned at inaccessible or dangerous locations and in harsh environments. Promising energy harvesting technologies have attracted the attention of engineers because they convert microwatt or milliwatt level power from the environment to implement maintenance-free machine condition monitoring systems with WSNs. The motivation of this review is to investigate the energy sources, stimulate the application of energy harvesting based WSNs, and evaluate the improvement of energy harvesting systems for mechanical condition monitoring. This paper overviews the principles of a number of energy harvesting technologies applicable to industrial machines by investigating the power consumption of WSNs and the potential energy sources in mechanical systems. Many models or prototypes with different features are reviewed, especially in the mechanical field. Energy harvesting technologies are evaluated for further development according to the comparison of their advantages and disadvantages. Finally, a discussion of the challenges and potential future research of energy harvesting systems powering WSNs for machine condition monitoring is made.

Energy Harvesting Technologies for Achieving Self-Powered Wireless Sensor Networks in Machine Condition Monitoring: A Review.

With the emergence of Internet-of-Things (IoT) technologies, the application of Wireless Sensor Network (WSN) has become pervasive in our physical environment. These technologies rely on the WSN to capture environmental data and transmit it through the network to the specified location or storage. Given the massive data that captured by the WSN and the data transmission that happen in the process, there is a challenge for any developers and researchers in WSN in terms of a continuous operation of the WSN device. This is because WSNs are battery-powered and often deployed in a remote location that is hard to access for a battery replacement. The purpose of this paper is to survey the energy harvesting for improving the lifetime of the wireless sensor networks powered by the environmental energy. Energy harvesting takes advantages of the stray energy from the environment as the sustainable source of power.

Energy harvesting in wireless sensor networks: A survey

Convolutional Neural Network (CNN) models are a type of deep learning architecture introduced to achieve the correct classification of breast cancer. This paper has a two-fold purpose. The first aim is to investigate the various deep learning models in classifying breast cancer histopathology images. This study identified the most accurate models in terms of the binary, four, and eight classifications of breast cancer histopathology image databases. The different accuracy scores obtained for the deep learning models on the same database showed that other factors such as pre-processing, data augmentation, and transfer learning methods can impact the ability of the models to achieve higher accuracy. The second purpose of our manuscript is to investigate the latest models that have no or limited examination done in previous studies. The models like ResNeXt, Dual Path Net, SENet, and NASNet had been identified with the most cutting-edge results for the ImageNet database. These models were examined for the binary, and eight classifications on BreakHis, a breast cancer histopathology image database. Furthermore, the BACH database was used to investigate these models for four classifications. Then, these models were compared with the previous studies to find and propose the most state-of-the-art models for each classification. Since the Inception-ResNet-V2 architecture achieved the best results for binary and eight classifications, we have examined this model in our study as well to provide a better comparison result. In short, this paper provides an extensive evaluation and discussion about the experimental settings for each study that had been conducted on the breast cancer histopathology images.

/pdf/breast-cancer-classification-using-deep-learning-approaches-z6mg35rdn9.pdf

Breast Cancer Classification Using Deep Learning Approaches and Histopathology Image: A Comparison Study

The Global Positioning System demonstrates the significance of Location Based Services but it cannot be used indoors due to the lack of line of sight between satellites and receivers. Indoor Positioning Systems are needed to provide indoor Location Based Services. Wireless LAN fingerprints are one of the best choices for Indoor Positioning Systems because of their low cost, and high accuracy, however they have many drawbacks: creating radio maps is time consuming, the radio maps will become outdated with any environmental change, different mobile devices read the received signal strength (RSS) differently, and peoples’ presence in LOS between access points and mobile device affects the RSS. This research proposes a new Adaptive Indoor Positioning System model (called DIPS) based on: a dynamic radio map generator, RSS certainty technique and peoples’ presence effect integration for dynamic and multi-floor environments. Dynamic in our context refers to the effects of people and device heterogeneity. DIPS can achieve 98% and 92% positioning accuracy for floor and room positioning, and it achieves 1.2 m for point positioning error. RSS certainty enhanced the positioning accuracy for floor and room for different mobile devices by 11% and 9%. Then by considering the peoples’ presence effect, the error is reduced by 0.2 m. In comparison with other works, DIPS achieves better positioning without extra devices.

https://www.mdpi.com/1424-8220/17/8/1789/pdf

Adaptive Indoor Positioning Model Based on WLAN-Fingerprinting for Dynamic and Multi-Floor Environments

Hash functions are considered key components of nearly all cryptographic protocols, as well as of many security applications such as message authentication codes, data integrity, password storage, and random number generation. Many hash function algorithms have been proposed in order to ensure authentication and integrity of the data, including MD5, SHA-1, SHA-2, SHA-3 and RIPEMD. This paper involves an overview of these standard algorithms, and also provides a focus on their limitations against common attacks. These study shows that these standard hash function algorithms suffer collision attacks and time inefficiency. Other types of hash functions are also highlighted in comparison with the standard hash function algorithm in performing the resistance against common attacks. It shows that these algorithms are still weak to resist against collision attacks.

/pdf/comparison-of-hash-function-algorithms-against-attacks-a-24jxbjc2oh.pdf

Comparison of hash function algorithms against attacks: A review

The first step in developing a ubiquitous environment, in which the user can interact with any available electronic device, is the existence of an accurate indoor positioning system. WLAN-based indoor positioning system is considered as one of the best choices for indoor positioning due to its low cost, simple configuration and high accuracy. Although the WLAN Received Signal Strength fingerprinting method is the most accurate positioning method, the offline phase of this method known as radio map creation is a time consuming process. On the other hand, in dynamically changing environments, this radio map will be outdated and this will reduce the positioning accuracy. In this paper the Multi-Wall signal path loss model will be used to automatically generate the radio map based on the knowledge of the environment layout. The results of the experiment show that the indoor positioning by using the generated radio map can achieve high accuracy with average distance error reaching up to 1.2m. This promising results means that an accurate indoor positioning system can be easily developed with time saving features.

Noor Azurati Ahmad

Papers

Energy harvesting in wireless sensor networks: A survey

Breast Cancer Classification Using Deep Learning Approaches and Histopathology Image: A Comparison Study

Adaptive Indoor Positioning Model Based on WLAN-Fingerprinting for Dynamic and Multi-Floor Environments

Comparison of hash function algorithms against attacks: A review

Automatic WLAN fingerprint radio map generation for accurate indoor positioning based on signal path loss model