Wireless operators spend millions of dollars every year paying for the electricity to power their cell sites and small cells. But there are new energy-saving features that are being developed that could make a dramatic difference in energy consumption. And these new features incorporate tools like artificial intelligence (AI) and machine learning. In a new Ericsson white paper called "Breaking the Energy Curve," the company said that machine learning can be used to make certain network features more autonomous. Two of those features, MIMO Sleep Mode and Cell Sleep Mode, are using machine learning to study data traffic patterns and save operators money.

Hexa Food's IoT team has deployed Huawei's ModelArts coupled with the intelligent device Atlas 500 to accurately identify the quality of the chilies it uses in spice blends. The AI can distinguish good chilies from bad, improving production efficiency and the quality of spices available to chefs and homes across Malaysia. Known as the "Kingdom of Spices", Malaysia is a multi-ethnic nation comprising Malays, Chinese, Indians, and the indigenous Orang Asli people. Its diverse culture is reflected in its cuisine, which draws from a multicultural heritage that sees hundreds of spices add flavor to the Malaysian diet. And of these, the colorful, aromatic, and spicy chili powder is a mainstay of many of the nation's signature dishes.

Connected vehicular network is one of the key enablers for next generation cloud/fog-supported autonomous driving vehicles. Most connected vehicular applications require frequent status updates and Age of Information (AoI) is a more relevant metric to evaluate the performance of wireless links between vehicles and cloud/fog servers. This paper introduces a novel proactive and data-driven approach to optimize the driving route with a main objective of guaranteeing the confidence of AoI. In particular, we report a study on three month measurements of a multi-vehicle campus shuttle system connected to cloud/fog servers via a commercial LTE network. We establish empirical models for AoI in connected vehicles and investigate the impact of major factors on the performance of AoI. We also propose a Deep Q-Learning Netwrok (DQN)-based algorithm to decide the optimal driving route for each connected vehicle with maximized confidence level. Numerical results show that the proposed approach can lead to a significant improvement on the AoI confidence for various types of services supported.

Google has rolled out a new technology to improve audio quality in Duo calls when the service can't maintain a steady connection called WaveNetEQ. It's based on technology from Google's DeepMind division that aims to replace audio jitter with artificial noise that sounds just like human speech, generated using machine learning. If you've ever made a call over the internet, chances are you've experienced audio jitter. It happens when packets of audio data sent as part of the call get lost along the way or otherwise arrive late or in the wrong order. Google says that 99 percent of Duo calls experience packet loss: 20 percent of these lose over 3 percent of their audio, and 10 percent lose over 8 percent.

Huawei has announced that its framework for AI app development MindSpore is now open source and available on GiHub and Gitee. The lightweight suite is similar to Google's TensorFlow and Facebook's PyTorch as it lowers the barrier to entry for developers looking to add AI to their apps. We implement AI Algorithms As Code through on-demand collaboration for easier model development, and provide cutting-edge technologies, and co-optimization with Huawei Ascend AI processors to improve runtime efficiency and computing performance. We also support other processors such as GPU and CPU." MindSpore already has the backing of a number of partners including the University of Edinburgh, Peking University, Imperial College London and the robotics startup Milvus. The framework is able to run on processors, graphics cards and dedicated neural processing units such as the one in Huawei's own Ascend AI chips.

Blind modulation classification is an important step to implement cognitive radio networks. The multiple-input multiple-output (MIMO) technique is widely used in military and civil communication systems. Due to the lack of prior information about channel parameters and the overlapping of signals in the MIMO systems, the traditional likelihood-based and feature-based approaches cannot be applied in these scenarios directly. Hence, in this paper, to resolve the problem of blind modulation classification in MIMO systems, the time-frequency analysis method based on the windowed short-time Fourier transform is used to analyse the time-frequency characteristics of time-domain modulated signals. Then the extracted time-frequency characteristics are converted into RGB spectrogram images, and the convolutional neural network based on transfer learning is applied to classify the modulation types according to the RGB spectrogram images. Finally, a decision fusion module is used to fuse the classification results of all the receive antennas. Through simulations, we analyse the classification performance at different signal-to-noise ratios (SNRs), the results indicate that, for the single-input single-output (SISO) network, our proposed scheme can achieve 92.37% and 99.12% average classification accuracy at SNRs of -4 dB and 10 dB, respectively. For the MIMO network, our scheme achieves 80.42% and 87.92% average classification accuracy at -4 dB and 10 dB, respectively. This outperforms the existing classification methods based on baseband signals.

Engineer.ai, which uses Artificial Intelligence to help small and mid-sized organisations build their own bespoke software (custom or tailor-made software), has raised a Series A investment of $29.5 Mn, led by Lakestar and Jungle Ventures. The funding round also saw participation from DeepCore -- Softbank's AI-focussed investment fund. Founded by Sachin Dev Duggal and Saurabh Dhoot in 2012, Engineer.ai is a global company with split headquarters in Los Angeles and London, supported by offices in Delhi and Tokyo. The startup was formerly known as SD Squared and was rebranded to Engineer.ai. in June 2018. With over $24M in gross revenue and customers that include BBC, Virgin Group and the San Francisco Giants, Engineer.ai

SoftBank Group Corp. fell as much as 10 percent after a provider of satellite-based internet service that it invested in filed for bankruptcy, ceding some gains from an unprecedented plan to sell assets and buy back shares. OneWeb made the filing late Friday U.S. time after raising about $3.3 billion in debt and equity financing from shareholders including SoftBank, Airbus SE and Qualcomm Inc. since its inception. At least $1 billion of that came from SoftBank, which said it first invested in December 2016 and declined to give a total amount. It is the latest blow to SoftBank founder Masayoshi Son, who last week unveiled a plan to raise $41 billion to buy back shares and slash debt. The announcement sent the shares soaring more than 50 percent in just a few days.

To accommodate diverse Quality-of-Service (QoS) requirements in 5th generation cellular networks, base stations need real-time optimization of radio resources in time-varying network conditions. This brings high computing overheads and long processing delays. In this work, we develop a deep learning framework to approximate the optimal resource allocation policy that minimizes the total power consumption of a base station by optimizing bandwidth and transmit power allocation. We find that a fully-connected neural network (NN) cannot fully guarantee the QoS requirements due to the approximation errors and quantization errors of the numbers of subcarriers. To tackle this problem, we propose a cascaded structure of NNs, where the first NN approximates the optimal bandwidth allocation, and the second NN outputs the transmit power required to satisfy the QoS requirement with given bandwidth allocation. Considering that the distribution of wireless channels and the types of services in the wireless networks are non-stationary, we apply deep transfer learning to update NNs in non-stationary wireless networks. Simulation results validate that the cascaded NNs outperform the fully connected NN in terms of QoS guarantee. In addition, deep transfer learning can reduce the number of training samples required to train the NNs remarkably. I. INTRODUCTION A. Background The 5th Generation (5G) cellular networks are expected to support various emerging applications with diverse Quality-of-Service (QoS) requirements, such as enhanced mobile broadband services, massive This paper has been presented in part at the IEEE Global Communications Conference 2019 [1]. The authors are with the School of Electrical and Information Engineering, University of Sydney, Sydney, NSW 2006, Australia (email: {rui.dong, To guarantee the QoS requirements of different types of services, existing optimization algorithms for radio resource allocation are designed to maximize spectrum efficiency or energy efficiency by optimizing scarce radio resources, such as time-frequency resource blocks and transmit power, subject to QoS constraints [3-9]. There are two major challenges for implementing existing optimization algorithms in practical 5G networks. First, QoS constraints of some services, such as delay-sensitive and URLLC services, may not have closed-form expressions. To execute an optimization algorithm, the system needs to evaluate the QoS achieved by a certain policy via extensive simulations or experiments, and thus suffers from long processing delay [9, 10]. Second, even if the closed-form expressions of QoS constraints can be obtained in some scenarios, the optimization problems are non-convex in general [8,10,11].

Edge intelligence refers to a set of connected systems and devices for data collection, caching, processing, and analysis in locations close to where data is captured based on artificial intelligence. The aim of edge intelligence is to enhance the quality and speed of data processing and protect the privacy and security of the data. Although recently emerged, spanning the period from 2011 to now, this field of research has shown explosive growth over the past five years. In this paper, we present a thorough and comprehensive survey on the literature surrounding edge intelligence. We first identify four fundamental components of edge intelligence, namely edge caching, edge training, edge inference, and edge offloading, based on theoretical and practical results pertaining to proposed and deployed systems. We then aim for a systematic classification of the state of the solutions by examining research results and observations for each of the four components and present a taxonomy that includes practical problems, adopted techniques, and application goals. For each category, we elaborate, compare and analyse the literature from the perspectives of adopted techniques, objectives, performance, advantages and drawbacks, etc. This survey article provides a comprehensive introduction to edge intelligence and its application areas. In addition, we summarise the development of the emerging research field and the current state-of-the-art and discuss the important open issues and possible theoretical and technical solutions.