Aaeon is the first company to publicize it is applying the KL520, while Kneron has already revealed customer wins for its facial recognition model, like systems integrator TIIS, which has built it into a security system for the public banks of Taiwan. Company sources said that Kneron made around $5 million in revenue in 2019, a stark difference to various edge AI chip companies who are not yet market-ready.

 

“Aaeon has used our chip and put it into a form factor that’s easy to insert into a pre-exisiting design,” said Kneron chief commercial officer Adrian Ong.

 

“We have other customers in the commercialisation phase,” he added.

 

The company has a good focus on AI at the edge and is directed at inference in computer vision applications that require low latency and low power, such as smart door locks. The flagship chip KL520, unveiled in May 2019, is optimized for image processing models based on convolutional neural networks (CNNs), including but not limited to Kneron’s own models for facial recognition. The KL520 runs 0.3 TOPS at 0.5 W (equivalent to 0.6 TOPS/W).

 

Kneron also has its very own neural processing unit (NPU) IP, neural network models for image processing, and a corresponding toolchain. “We started out developing models for face recognition, face detection and object detection,” Ong said. “But because [a lot of us] had a silicon background, we decided to develop our own IP and our own chips as well.”

 

Kneron’s is using a reconfigurable chip architecture to customize its silicon to specific models, assisted by a compiler which allocates resources successfully so the hardware can be best utilised to run the models. The compiler also uses compression techniques which Ong claimed can tremendously reduce the size of the models to allow them to run in resource-constrained environments to save power and cost.

 

Kneron’s facial recognition model was earlier recognized by NIST (the National Institute of Standards and Technology) as the best performing model under 100MB. “In fact, the model is 57 MB,” Ong said. “It even outperformed competitors’ models that were bigger than 1 GB. For embedded applications, it can be compressed even further, down to 32 or even 16 MB.”

 

Kneron was established in 2015, with its 110 employees based at headquarters in San Diego, a development office in Taiwan and a sales office in Shenzhen. It is backed by Horizons Ventures, Alibaba Entrepreneurs Fund, Qualcomm and Sequoia Capital, amongst others. “The strength of the company is image processing, but we recognise that over time people are going to want to add audio capability as well, so we are developing IP for that,” Ong said.

 

Kneron’s second generation chip will be able to accelerate both CNNs and RNNs (recurrent neural networks), for vision and audio applications. Samples of the second generation chip are due in Q1-2020.

 

TRONSERVE

 

Make Full Backups of Essential Business Data

 

Facilitate a full, encrypted backup of your data on each computer and mobile device a minimum of once a month, shortly after a complete malware scan. Store these backups at a safeguarded, off-site location. Save your encryption password or key in a secure location separate from where your backups are stored. Multiple software applications will allow you to encrypt your backups.

 

With your backups in place, if a computer breaks, an employee makes an error, or a malicious program infects your system, you will be able to restore your data. Without backups, you will have to manually recreate your business information from paper records and employee memory.

 

It is mandatory to back up data such as:

- Word processing documents and electronic spreadsheets

- Databases, especially customer relationship management (CRM), financial, human resource (HR), and accounts receivable (AR)/payable (AP) files

- Product design and manufacturing data associated with or related to CAD/CAE/CAM, process plans, tooling and other inventory information, production scheduling, inspection, maintenance, bid data, work orders, scheduling

- Other operational technology (OT) data such as machine and process condition monitoring and analysis

- System logs and other information technology (IT) information

 

Don’t be afraid about the software applications; simply focus on the data. Store your backups on an external USB hard drive, other removable media, or a separate server. Use caution when picking a partner if you decide to store your data online and encrypt all data ahead of storing it in the cloud. Hard-drive backups should be large enough to hold all your monthly backups for one year. Create split folders for each computer so you can copy your data into the appropriate folder on the external drive. After your backups are complete, test them instantly to assure your efforts were successful.

 

Make Incremental Backups of Important Business Information

 

Plan automatic incremental or differential backups a minimum of once a week. Because they will only record information since your last backup, you may need to schedule them daily or once an hour, depending on the needs of your business. Think about how much information was changed or generated between each backup and the impact to your company if that information was lost. So many security software suites offer automated backup functions that will do this on a regular schedule for you.

 

Check always your storage capacity. You should be able to hold data for 52 weekly backups, so the capacity should be about 52 times the amount of data you want to store. Take caution to back up the data for every computer and mobile device.

 

For extra redundancy, store your backups in multiple locations, such as one in the office, one in a safety deposit box across town, and one in the cloud. Remember that incremental testing is just as important as incremental backups to make sure you can read your data and use that information in the event of a security breach.

 

Consider Cyber Insurance for Increased Recovery Capability

 

Like flood or fire insurance, you can obtain cyber insurance for your facility. These services can help you recover from an information security incident more promptly and effectively and may cover the cost of: 

- Cybersecurity expertise to assist in identifying the extent of damage caused

- Consultation to help investigate the incident and report it to the appropriate authorities

- Loss of revenue due to downtime

- Legal fees, fines, and penalties incurred

 

In regards to any partner, select a cyber insurance provider with care. Do your due diligence by researching the company, the services they provide, the type of events they cover, and their reputation for meeting their contractual agreements.

 

Assess & Improve Your Procedures & Technologies

 

Take an honest look at your processes, procedures, and technology solutions and assess what improvements you need to make to lessen your risks. Try conducting training or tabletop exercises. These scenario exercises can simulate a major event, which will allow you to identify potential weaknesses and readiness. Then you can make corrections as recommended.

 

TRONSERVE

 

The Galaxy S10’s Intelligent Wi-Fi uses machine learning to make sure that users’ Wi-Fi connections are both smart and safe. The innovation allows the device to recognize when a user has entered a space where Wi-Fi connections are usually to be unstable, and quickly switch to LTE connectivity to avoid network interruption. What’s more, Intelligent Wi-Fi allows the device to more promptly identify weaker Wi-Fi signals in locations which include elevators, allowing it to transition from Wi-Fi to LTE connectivity much more quickly. After connecting to a new network, the function even allows the device to examine traffic patterns and notify users if the network is suspicious, thus protecting them from dangers such as phishing and malware attacks.

 

The Galaxy S10’s Wi-Fi 6 capability produces smartphone support for next generation technology, boosting download and upload speeds by up to 20 percent in comparison with previous generations of Wi-Fi. When users are in crowded areas where Wi-Fi signals would usually be slow and unstable, Wi-Fi 6 enables for network connections that are about four times faster than those experienced by users of older Wi-Fi generations connecting in the same location. And shockingly, this added speed and reliability actually uses less battery power, not more, allowing users to conserve power for when they need it.

 

The WBA Industry Awards have been honoring great developments in Wi-Fi ecosystems globally for the past seven years. The awards acknowledge exceptional innovations in wireless networks, smart cities and remote wireless deployments in the interests of recognizing and encouraging excellence and promoting successful practices. Each shortlisted nomination is featured in a WBA Case Study so that the innovation may be shared with the wider industry.

 

The ‘Best Wi-Fi Innovations’ category acknowledges network operators and technology companies whose solutions are post-deployment, and “recognizes and celebrates innovations and deployments with Wi-Fi, which enable seamless and secure support, greater interoperability and/or transformational benefits to customers.”

 

Talking about the latest awards win, Jong-Mu Choi, Head of Convergence R&D Group at Samsung Electronics said, “Samsung is proud that its Wi-Fi innovations have been recognized in such a way. The company will keep working to enable its devices to make use of these new Wi-Fi capabilities, so that the improved usability therein reaches consumers as quickly as possible.  One example of this is the Galaxy Note10’s recent accreditation as Wi-Fi CERTIFIED 6™, which saw it join the Galaxy S10 as one of Samsung’s certified Wi-Fi 6-capable devices. Going forward, Samsung looks forward to utilizing faster, smarter Wi-Fi connections as the foundation that will allow it to continue bringing added convenience to users.”

 

TRONSERVE

 

The Clock Is Ticking When a Threat Is Detected

 

Don’t wait for a hacker to hit before developing your incident response plan. For smaller manufacturers, even a small security breach can have a significant impact on their operations. Acting upon instantly will empower you to better contain or lower the impact of a cyber attack.

 

Develop a Plan for Information Security Incidents

 

When developing your response strategy, look at the immediate actions you and your employees will need to take in case there is an incident. Your response plan should include:

 

Roles and Responsibilities

 

- Who: Make a list of who to call in the instance of an incident. It’s crucial you know who will make the decision to initiate recovery procedures and who will be the primary contact with appropriate law enforcement personnel.

- What: Be sure you have a plan for what to do with your data in the case of an incident. This may include shutting down or locking your computer systems, moving your information to a backup site, and/or physically getting rid of important documents and sensitive materials.

- When: Determine when to alert senior management, emergency personnel, cybersecurity professionals, legal council, service providers, or insurance providers. Be sure to include all relevant contact information.

- Type: Your response plan should describe the sorts of activities that constitute an information security incident. Include incidents such as your website being down for more than a specified length of time or evidence of information theft.

 

Know Your Notification Obligations

 

Be aware that many states and countries have notification laws that require businesses to alert customers if there is a chance their information was stolen, disclosed, or otherwise lost. Familiarise yourself with international, state, and local laws regarding notification obligations and include that information in your response plan.

 

Likewise include suggestions on when to notify appropriate authorities. You should contact your local police to file a report when there's a possibility that any personal information, intellectual property, or other sensitive information was forged. Most importantly, you and your employees should understand your role in your cybersecurity response plan. Develop procedures for each job role that describe exactly what the employee is required to do if there is a cybersecurity incident. When everyone understands their role in your response plan, you can act fast and mitigate the potential damage.

 

TRONSERVE

 

Richard Yu Chengdong, chief executive of Huawei's mobile business, told CNBC that the Shenzhen-based company has indefinitely put aside for now the launch of a new product in its Matebook series. The US blacklist, which prohibits American companies from selling products and services to Huawei, was behind the cancellation, as reported by Yu. The company did not immediately confirm the remarks by Yu.

 

The Post reported last month that Microsoft had stopped acknowledging new orders from Huawei after the Chinese telecoms gear and smartphone maker was added to the US Entity List. The two major areas of business between Huawei and Microsoft - Windows operating systems for laptops and other content-related services - have both been dangling by the US company as it drives to stick to US government restrictions, according to the Post report, which referred people familiar with the matter.

 

Huawei, the largest smartphone vendor in China and the second largest all over the world, has been pushing its laptop business not long ago, launching a series of personal computers that cover the mid- and high-end markets.

 

Huawei also confronts the prospect of not being able to use the Android operating system (OS) on its smartphones and Windows OS on its PC products. The company set out work on an alternative OS seven years earlier and yet isn't fully prepared to launch it yet as the US ban came on suddenly, sources told the Post in a separate report.

 

TRONSERVE

 

Amazon said its ''Climate Pledge'' initiative and said it would be its first signatory, as part of an attempt to cut down on emissions according to a 2050 goal for carbon neutrality set by the Paris agreement.

 

''We're done being in the middle of the herd on this issue,'' Mr Bezos said. ''If a company with as much physical infrastructure as Amazon - which delivers more than 10 billion items a year - can meet the Paris agreement 10 years early, then any company can.'' Mr Bezos remarked he had spoken with other CEOs of global companies, and noted ''I'm finding a lot of interest in joining the pledge.''

 

Together with the ramped-up effort, Mr Bezos said Amazon had agreed to get 100,000 electric delivery vans from vehicle startup Rivian, to help cut its carbon footprint. Amazon already revealed a US$440 million (S$605.7 million) investment in Rivian. The first vans will hit the road in 2021, with the fleet to be perfectly operating in 2030. Amazon also pledged to invest US$100 million on reforestation efforts in co-operation with the Nature Conservancy.

 

Amazon said its new ''sustainability'' initiative will deal with all of its business operations, with reduced carbon in packaging, delivery and its own energy use. Mr Bezos's announcement came the day before a global day of demonstrations to demand action on climate change, previous to a UN summit on zero-emissions on September 23. Amazon confronts mounting pressure to address its environmental impact, with more than 1,000 of its workers preparing to walk off the job Friday as part of the Global Climate Strike.

 

Asked about the Amazon employees set to take part in the strike, Mr Bezos called it ''totally understandable.'' ''People are passionate about this issue,'' he said. ''Everybody in this room should be passionate about this issue.''

 

Amazon staff welcomed the declaration but said they would maintain their protest. ''Climate Pledge is a huge win.. & we're thrilled at what workers have achieved in under a year. But we know it's not enough,'' said a Twitter message from Amazon Employees for Climate Justice.

 

The Paris Agreement, alone, will not get us to a livable world. Today, we celebrate. Tomorrow, we're going to be in the streets. Greenpeace USA senior campaigner Gary Cook said the Amazon news was optimistic but lacked details on exactly how it would be implemented.

 

''Amazon is known for speed, but if Jeff Bezos wants Amazon to be a leader on climate, he needs to spell out exactly how it is going to rapidly move the company off of fossil fuels to keep our planet within the 1.5 degrees temperature threshold in the Paris Agreement that Amazon has now committed to,'' Mr Cook said.

 

Seeking 'many more' companies 

 

Mr Bezos was accompanied by Christiana Figueres, the former UN climate chief and founder of the climate activist group Global Optimism, who expressed hope the pledge would stimulate more action by companies. ''If Amazon can set ambitious goals like this and make significant changes at their scale, we think many more companies should be able to do the same and will accept the challenge,'' she said.

 

Mr Bezos stated he hopes Amazon to reach 80 per cent renewable energy use by 2024, up from around 40 per cent today, and 100 per cent renewable energy by 2030 on a path to net-zero carbon by 2040. This will involve investments in wind and solar energy projects and initiatives to control carbon emissions at its facilities including the second headquarters, HQ2, just outside the US capital. Mr Bezos said that Amazon's step to speedier shipping, including one-day delivery on many items, would be a net positive in environmental terms even though it was ''counterintuitive.''

 

''The reason is, that once you get to one day and the same day, you can't really do it by air transportation anymore,'' he said. This means keeping warehouses and products closer to the consumer and as a result ''you're actually transporting the products in a very efficient way,'' he said.

 

TRONSERVE

 

A good autonomous checkout should enable the shopper to walk in, grab stuff, and leave. That is all you have to do to buy in an autonomous checkout store. Companies getting into this space have been trying to fix this problem using various kinds of approaches. It is astonishing how few of these companies have addressed the 5 hurdles of autonomous checkout necessary to make any idea a viable market solution: privacy, scalability, experience, flexibility, and insights (or “PSEFI” for short).

 

Privacy

 

Leave out regulations for a while. It is a company’s job to do everything it can to shield their customers’ privacy. As technology advances, it is the responsibility of those who bring the advancements, not the regulators, to guard the people they serve. That said, guidelines can be a practical guide and should be seen as the very least set of requirements a solution must satisfy. Take the example of the class action lawsuit against Walmart filed in April of this year.  It involves eradicating the use of facial recognition for checkout.  However this case goes, we believe biometric information, mostly facial recognition, is unnecessary for autonomous checkout. Shoppers deserve a solution that serve their needs without compromising their privacy.

 

Scalability

 

Autonomous checkout can not include closing a shop to deploy technology. Using in-shelf sensors and in-shelf cameras is not realistic for chain-wide deployment. A retailer would have to either take all the product off the shelves, change all the shelves, and restock the new shelves, OR they would have to retrofit all the current shelving with sensors on premise. None of these options are practical for retailers as they would require stores to close entirely for at least multiple weeks. Not only will a retail chain not accept that as a market solution, but it would mean a chain-wide rollout that could take years to complete.

 

The appropriate solution is a lightweight overhead camera installation that can be done as quickly as a few hours.  These installs can be done in a single day without disrupting customers or operations. Lower installation requirements indicate less labor and faster deployment. Less hardware and less labor translate to lower costs and fewer lost sales. Additionally, just overhead cameras leads to reduced maintenance calls and operational complexity.

 

Experience

 

After having spoken with 200+ retail chains in many countries, one thing is clear: The retail experience is changing, fast! The advent of ecommerce has forced retailers to rethink the in-store experience to bring back online shoppers.  Pressure from ecommerce isn't going to kill retail, shopping is too fun! It is, however, forcing retailers to compete by reimagining what shopping means. Two of the biggest complaints most US retailers receive from customers are (1) long wait times in line and (2) poor customer service. Bypassing the need for waiting in line allows retailers to put more resources toward enhancing the in-store customer experience.

 

Autonomous checkout is the first big modify to retail since barcode scanners and it’s only going to get better.  Soon, shoppers will be able to walk in, without having to take their phones out, shop and leave.  A number of the retailers I have spoken with already have specific experiences in mind for what they will do with their stores once they have autonomous checkout. In other words, autonomous checkout helps retailers a chance to better serve their customers and the flexibility to reimagine the original checkout space to delight their shoppers.

 

Flexibility

 

Retail environments are intricate and continually updating. Sales happen on the fly, items get moved around the store by shoppers, the Internet goes down, inventory and compliance issues are dynamic, and displays and presentation must be consistent with the retailer’s brand. Autonomous checkout should provide retailers the ability to amplify their brand with their customers. The fewer hardware types used in autonomous checkout, the more flexibility a store can be with its merchandising and its brand experience.

 

Flash sales are one example of this flexibility. Imagine baked goods that are about to expire within the day. ‘Let’s put them on a table over there and slap a 50% off sticker on them. Go!’

 

If a store uses shelf sensors, all of its products need to stay on that unique shelf, making this scenario extremely hard. There are so many examples when a store manager needs to make changes to adapt and meet customer needs on the fly. Payment flexibility is also important. A viable solution must accept cash, credit, and other popular payment methods.

 

Insights

 

Retail analytics are invaluable, but they must be seen through the lens of privacy. A lot of information can be gathered by retailers and brands, but they should have no connection to any particular individual. Nevertheless, data helps retailers and brands clarify their messaging to customers and create products that customers want. What products do people pick up, but not buy? How are shoppers reacting to brand campaigns and advertisements? What is the storefront inventory at any given moment? These questions  are currently only answered via ecommerce, but AI-based computer vision can level the playing for brick & mortar retailers.

 

Autonomous checkout will push numerous new experiences imagined by the world’s best retailers. The only way for the technology company to bring them to market, however, is to clear these 5 Hurdles.

 

TRONSERVE

 

NEC’s technology rated No. 1 in NIST testing for the fifth time, after its top placement in the face recognition testing for video in 2017. The high performance of NEC’s technology is shown in the test results which placed the company dramatically ahead of the runner-up.

 

In the past few years, the rising convenience of biometric authentication technology, improved security awareness, and the remarkable development of artificial intelligence (AI), have driven companies in many countries to begin adopting biometric authentication technology. In particular, the use of face recognition technology is rather quickly enlarging across an array of fields throughout the world. Face recognition technologies are now in use in areas that require high reliability, convenience and long-term use, for example identity verification and national infrastructure, transaction settlements, bank account establishment, and passport verification.

 

Forty-nine organizations, along with companies from the United States, China, Russia, Europe, and Japan, participated in the NIST’s FRVT 2018, where the evaluation of face recognition accuracy was practiced. These tests are the most strenuous and fair benchmarks implemented by the NIST as each organization is required to submit and be reviewed on programs that were developed during the same period. By performing multi-stage matching, an impressive search speed of 230 million matchings per second was achieved. On top of that, leveraging NEC's deep learning methods to considerably decrease the identification error rate, NEC exactly matched images of a subject taken over a 10-year interval with an error rate that was 4 times lower than the runner-up.

 

TRONSERVE

 

Install & Update Anti-virus & Other Cybersecurity Programs

 

If you have not already installed anti-virus, anti-malware, and anti-spyware software on every device at your manufacturing facility, now is the time. Install, use, and continuously update these cybersecurity measures on every computer, tablet, and smartphone.

 

These mechanisms can certainly help shield your company’s important data and information from malware, which is the catch-all term for malicious code. Written with the purpose to steal or cause harm to information systems, malware contains viruses, spyware, and ransomware. Destructive code can not just steal your computer memory; it can also facilitate a cyber criminal to record your computer actions and access sensitive information.

 

To get the most from your anti-malware programs, set the software to conveniently check for updates at a minimum once daily, or in real-time, if available. Set the settings to run a complete scan after daily updates.

 

A typical example of typical business anti-malware settings might include:

- Running anti-virus programs daily or nightly, such as at midnight

- Scheduling a virus scan to run about half an hour later (12:30 a.m.)

- Following up by running anti-spyware software a couple of hours later, such as at 2:30 a.m.

- Running a full system scan shortly afterward (3:00 a.m.)

 

This example is based on the assumption that a facility normally has a running, high-speed Internet connection for all devices. The time of your updates and scans may differ, but you need to do them daily. Do not forget to schedule them so that basically one activity takes place at any given time. For home-based employees or for employees’ personal devices, ensure that they have copies or access to the same anti-virus and anti-spyware software, and ask them to run frequent updates per the previous example.

 

It's vital that all employees understand why running anti-virus, anti-malware, and anti-spyware is vital to protecting company information and assets. Employees must also realize how early detection could potentially save the company from serious consequences associated with a cybersecurity incident or breach.

 

For redundant security, it's a good idea to use two different anti-virus solutions from different vendors. Using anti-malware protection from two different providers can increase your chances of detecting a virus. Routers, firewalls, or Intrusion Detection and Prevention Systems (IDPS) usually have some anti-virus capabilities; but you don’t want to rely on them exclusively to protect your network.

 

Understand that anti-virus solutions can only find known viruses. If a new virus is developed and deployed, your anti-virus may not be able to recognize it. It is essential to keep your anti-virus solutions up to date in order to detect the latest viruses.

 

Maintain & Monitor Detection Logs

 

Most malware protection and detection hardware or software is built with logging capability. Check your user manual for instructions on how to:

- Use your logs to identify suspicious activity

- Maintain regular log records that are valuable in an investigation

- Back up logs regularly and save them for at least a year (although some types of information may need to be stored for longer)

 

For added assurance, consider recruiting a cybersecurity professional to review your logs for any red-flag trends, such as an unusually large amount of time spent on a social media site or a high frequency of viruses consistently found on a single computer. This activity may present an important information security problem that requires stronger protection.

 

TRONSERVE

 

Data is routed through the Internet’s several metaphorical tubes using what's called the Border Gateway Protocol (BGP). Any data moving over the Internet needs a physical path of networks and routers to make it from A to B. BGP is the protocol that moves information through those paths - though the downside, just as a person in a game of telephone, is that each junction in the path only knows what they have been told by their immediate neighbor.

 

Since a particular junction in a route recognizes only where the data it’s transmitting just came from and where it’s headed next, it’s relatively simple for someone to step in and divert the data. At these unique junctions, autonomous systems establish BGP connections. Like a party pooper intentionally ruining a game of telephone by whispering a totally different phrase than the one that was told to them, a hacker may perhaps insert their own autonomous system to reroute information. The worst offenders are serial hijackers, who continuously deviate data to skim information or enable distributed denial-of-service (DDOS) attacks. In 1998, various hackers testified to the U.S. Congress that the Internet could be taken down by a dedicated hacker in 30 minutes by deploying BGP hacking.

 

Over the years, serial hijackers have been difficult to stop. One recent example was Bitcanal, a Portuguese web hosting firm that devoted years helping serial hijackers in their attacks. It took years of coordinated effort from legitimate service providers to shut down Bitcanal, and meanwhile, many other serial hijackers still roam the Web. What’s worse, serial hijackers have to, as the name suggests, launch several attacks before it becomes clear that they're a bad-faith actor.

 

“BGP [hacking] is one way to sniff at traffic, or steal traffic,” says Cecilia Testart, a graduate student at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL). ”Given that the Internet is becoming more and more critical, we should try and prevent these attacks.”

 

Testart is the lead author on a paper posted today [PDF] by a few researchers at CSAIL and the Center for Applied Internet Data Analysis (CAIDA). They have suggested that machine learning can be used to pro-actively stop serial hijackers from their hijinks. Serial hijackers, the researchers suggest, show some characteristic traits that make them be noticeable in comparison with ordinary network providers. They reveal that machine learning could dig out serial hijackers a lot faster than the standard method of identifying them only after multiple attacks.

 

The joint team used a machine learning technique which is called an extremely-randomized trees (extra-trees) classifier. In a test with their classifier, the classifier flagged 934 out of 19,103 autonomous systems it tested as potential serial hijackers. You can find extra-trees classifiers as though you were growing a forest of trees, where each tree represents a vote of confidence - as for instance, whether anybody is a serial hijacker - in line with a randomized subset of available information.

 

The resulting forest represents a consensus. If most trees have arrived at the choice that someone is a serial hijacker using the limited information available to them, then you possibly have one on your hands. Testart says extra-trees classifiers and other forest classifiers do not have the same bias toward a set of training data that a machine learning technique such as deep learning may have. Because the available data on known serial hijackers is so small, deep learning techniques may have skewed toward uncovering only ones most similar to known attackers and missed ones that might differ more.

 

Of course, for individual trees to cast a vote, they have to know what they are looking for. The research group identified a few ways in which serial hijackers differ from authentic network providers that usually route Internet traffic. For example, authentic providers tend to be online more regularly, as they are providing Internet service to genuine customers. Serial hijackers, on the contrary, would only be online while they are skimming data.

 

Serial hijackers furthermore usually have more diversified Internet Protocol (IP) blocks - primarily the street addresses of the Internet. Testart explains that an institution like MIT normally has a block of consecutive IP addresses that it uses. Hijackers, however, choose to pick up small strings of IP addresses as they become defunct from other users. One user with a bizarre selection of IP blocks, therefore, is more likely to be a serial hijacker.

 

These rules won't be set in stone. Testart notes there are times when a recognized network provider could go offline - for example, during an earthquake or blackout. Fat finger errors can also lead to typos and misconfigurations that could make a legitimate provider look suspicious at first glance. Testart says there is still plenty to be done with the work the research team has published so far. She suggests that an extra-trees classifier like the one the group developed could give network operators a sort of reputation score, to ensure serial hijackers would see their reputations drop quickly as they went about their nefarious business.

 

The other alternative is to update BGP and turn the game of the telephone into something more secure. But Testart doesn’t think that’s likely. “The Internet is a huge network,” she says. “It’s running on infrastructure set up many years ago. If you update a major protocol, you need to update all that infrastructure.” Think about the headaches of trying to get every network provider in the world to agree to change a protocol - it is far easier just to build a tool that can sniff out serial hijackers.

 

TRONSERVE

 

Huawei Technologies Co Ltd, the world’s biggest telecoms gear maker, has been put on a U.S. trade blacklist since May, when trade talks between Washington and Beijing broke down. The United States says the company can spy on users, which Huawei denies. The blacklisting prevented Huawei from buying parts and components from U.S. companies without U.S. government approval, constraining its access to essential technologies particularly Google Mobile Services.

 

U.S. companies can seek a license for particular products to be exempted from the ban. The U.S. Commerce Department has accepted more than 130 applications from companies for licenses to sell U.S. goods to Huawei, Reuters reported in August. Government officials advised U.S. companies to apply for licenses following U.S. President Donald Trump’s pledge of relief, saying exports to Huawei of non-sensitive items that are readily replaced by foreign competitors would be permitted.

 

Trump's administration gave the green light last week to start approving licenses for a few American companies to bypass the curbs, the New York Times said nyti.ms/35pED2e, stating people familiar with the matter. A U.S. Commerce Department spokesman told Reuters that no official direction has been granted to the department on the matter as of Wednesday afternoon.

 

TRONSERVE

 

The thickness of the package (720?) remains the same as current 8-layer High Bandwidth Memory-2 (HBM2) products, which is a tremendous advancement in component design. This will help customers release next-generation, high-capacity products with better performance capacity without having to change their system configuration designs. Additionally, the 3D packaging technology also features a shorter data transmission time between chips than the currently existing wire bonding technology, resulting in considerably faster speed and lower power consumption.

 

“Packaging technology that secures all of the intricacies of ultra-performance memory is becoming tremendously important, with the wide variety of new-age applications, such as artificial intelligence (AI) and High Power Computing (HPC),” said Hong-Joo Baek, executive vice president of TSP (Test & System Package) at Samsung Electronics.

 

“As Moore’s law scaling reaches its limit, the role of 3D-TSV technology is expected to become even more critical. We want to be at the forefront of this state-of-the-art chip packaging technology.”

 

Relying on its 12-layer 3D-TSV technology, Samsung will offer the ultimate DRAM performance for applications that are data-intensive and amazingly high-speed. Also, by enhancing the number of stacked layers from eight to 12, Samsung will soon be able to mass produce 24-gigabyte (GB)* High Bandwidth Memory, which provides three times the capacity of 8GB high bandwidth memory on the market today.

 

Samsung will be able to meet the aggressively expanding market demand for high-capacity HBM solutions with its cutting-edge 12-layer 3D TSV technology and it hopes to solidify its leadership in the premium semiconductor market.

 

TRONSERVE

 

Limit Access to Your User Data & Information

 

Controlling having access to your valuable user data lessens the chance for human error, which is the number-one information security threat. If a worker leaves your company, or transfers to another company location, take protective action right away, including deleting passwords and accounts from all systems and collecting company ID badges and entry keys. An ounce of access prevention can equal a pound of protection when it comes to limiting the impact of a disgruntled ex-employee.

 

Install Surge Protectors & Uninterruptible Power Supplies

 

Uninterruptible power supplies (UPS) can provide just enough battery life and time to save your data in the event of a power disruption. Check to make sure the UPS type and size meets your standards and requirements. Every computer and networked device should be plugged into a UPS. For less-sensitive electronics and non-networked equipment, standard surge protectors should suffice. Make sure to test and replace each UPS and surge protector as recommended by the manufacturer.

 

Patch Your Operating Systems & Software Regularly

 

Whatever new app can open the door to a cyber attack if you don’t on a regular basis patch and update all software on every device used by your employees. Regularly check for updates when purchasing a new computer or installing a new software system. Understand that software suppliers aren't required to provide security updates for unsupported products. For example, Microsoft® will stop supporting Windows 7 in January of 2020, so if you've not up-graded yet, now is the time. Do not holdup downloading operating system updates. Updates often include new or enhanced security features.

 

Install & Activate Software and Hardware Firewalls

 

Firewalls can combat malevolent hackers and hinder employees from browsing inappropriate websites. Install and update firewall systems on every employee computer, smartphone, and networked device. Include off-site employees, even in the event you use a cloud service provider (CSP) or a virtual private network (VPN). You may also want to install an intrusion detection/prevention system (IDPS) to provide a greater level of protection.

 

Secure All Wireless Access Points & Networks

 

For secure wireless networking, use these router best practices:

- Change the new device administrative password

- Set the wireless access point so that it does not broadcast its service set identifier (SSID)

- Set your router to use WiFi Protected Access 2 (WPA-2), with the Advanced Encryption Standard (AES) for encryption

- Avoid using WEP (Wired-Equivalent Privacy).

 

For guest WiFi access, use a separate network from your business account.

 

Set up Web & Email Filters

 

Use email and web browser filters to dissuade hackers and restrict spam from clogging employee inboxes. You can also download “blacklist” services to block users from browsing speculative websites that pose malware risks. Caution your employees against visiting sites that are frequently linked with cybersecurity threats, such as pornographic websites or social media. This might appear to be a no-brainer; but it only takes one employee to visit the wrong website to inadvertently download malware.

 

Use Encryption for Sensitive Business Information

 

Use full-disk encryption to shield all your computers, tablets, and smartphones. Save a copy of your encryption password or key in a protected place distinct from your stored backups. Email recipients typically need the same encryption capability in order to decrypt. Never send the password or key in the same email as the encrypted document. Give it to them via phone or some other method.

 

Dispose of Old Computers & Media Safely

 

Before donating or trashing old computers, you must remove all important hard drive information. Remove any sensitive business or personal data on old CDs, flash drives, or other old media. Then destruct these items or take them to a company that will shred them for you. Destroy sensitive paper information with a crosscut shredder or an incinerator.

 

Train Your Employees

 

Cyber-vigilant employees are your best protection against information security threats.

 

Every employee should know:

- What business and personal use is permitted for emails

- How to treat business information at the office or at home

- What to do if a cybersecurity incident occurs

 

Train any new employee to protect valuable data and get them to sign your information policy. Use newsletters and/or ongoing training to boost your culture of cybersecurity. Now that we've covered the key steps to protect your valuable data and information, we will show you how to install mechanisms for detecting and recognizing a cyber attack in part three of our series on “Cybersecurity for Manufacturers” from the MEP National Network.

 

TRONSERVE

 

Such hi-tech gadgets are central to Google's strategy to draw consumers with its own-branded hardware and thereby integrate them ever-closer with the company's better-known search engine and artificial intelligence-driven software. ''The two [Pixel 4] smartphones are already proceeding into mass production and will be ready to ship after Google's [new products] unveiling next week,'' one of a few sources close to the situation said. ''Google is also working on a version with 5G technology, which is in test production.''

 

A marketing splashing generated by Google's step into 5G smartphones, which would be the first by a U.S. company, presents a direct challenge to Apple and a few other leading hardware and handset makers just like Samsung Electronics and Huawei Technologies. It may also throw down the gauntlet at Microsoft, Google's biggest U.S. rival in search and software, which proclaimed a return last week to the tremendously competitive smartphone market with its own folding model, after it exited in 2016.

 

Google has made no trick of its aim to move into hardware although the apparent acceleration of its investment plans, joined with the company's deep pockets, will raise the pressure on competitors. Alphabet, Google's parent, has around $117 billion in cash, while Apple has $102 billion.

 

''Bringing together software and hardware ... does have a lot of synergistic value ... and the main way to do that today for our core [search and software] products is by using hardware,'' Google CEO Sundar Pichai, told analysts in a February conference call. ''As we scale up our hardware efforts ... you can definitely glimpse the future.'' As a relative newcomer to the global smartphone market, Google's Pixel phone series nevertheless has less than a 0.5% market share. Nonetheless that is flourishing exponentially. This year the company strives ship as many as 10 million phones, sources said, more than twice last year's level.

 

''Although Google's smartphone shipments are still small, it's one of the clients that still has healthy volume growth,'' another source said. ''It offers better prices for suppliers too.'' Google's 5G model is currently being test produced in China. But to escape U.S. tariffs from the China trade dispute, and to source cheaper labor, all Google production meant for the U.S. will eventually be made outside China. A leading option is a Vietnamese facility due to be ready by the end to the year, as the Nikkei Asian Review first reported.

 

Google may eventually make a decision to release its new 5G model in spring next year, the sources said, together with the release of a budget Pixel phone. But that release date would still put Google prior to Apple in the race to produce the next generation of 5G phones. The Google 5G handset will include a Snapdragon 855 mobile platform, made by U.S. chipmaker Qualcomm, the sources added. For now the new flagship Pixel 4 series will sport advanced organic light-emitting diode screens, like Apple's premium iPhone 11 Pro and 11 Pro Max models launched in September.

 

TRONSERVE

 

“Europe needs a common approach on crypto-assets such as Libra,” Valdis Dombrovskis told EU lawmakers in a confirmation hearing. “I intend to propose new legislation on this.”

 

The EU doesn't have particular regulations on cryptocurrencies, which, until Libra was released in June, had been considered a marginal issue by most decision-makers because only a fraction of bitcoins or other digital coins are translated into euros. Dombrovskis has resisted regulating digital currencies in the five years he has served so far. He have cleared his change of heart stemmed from Facebook’s plans for Libra, a digital currency that “could have systemic effects on financial stability,” he told lawmakers.

 

The EU is now also stressing the G20 for global action on “stablecoins,” an EU document circulated last week said. Facebook’s planned Libra is the best-known of the stablecoins - cryptocurrencies backed by assets like conventional money deposits, short-term government securities or gold.

 

Libra’s scale would likely pose issues, Dombrovskis said, seeing that Facebook’s millions of users in Europe would be able to pay with the new digital currency. An EU Commission official said there was no timetable yet for advising the new rules. Dombrovskis said the crypto regulation should focus on protecting financial stability, protecting consumers and tackling the risks of money-laundering using crypto-assets, which can easily cross borders.

 

TRONSERVE

 

To serve this increasing need, a lot of robotics systems integrators have come online and promise complete packages to guide manufacturers from initial assessment to fully realized industrial automation. But deferring to these experts can feel a bit imposing to manufacturers who rely on established processes they've developed internally.

 

So how does a manufacturer contemplating a first robot integration project participate fully so that the project is a success on their terms? Here are four suggestions to guide you during the process.

 

1. Be honest about the level of support you need.

 

You might have in-house robotics expertise or want to use your first robot integration project as the chance to learn. This is a good idea and yet do not forget to consider how much time it will take. While every small manufacturer is different, one thing is common across the sector: Everyone is always strapped for bandwidth. Even if you have all the capabilities required to implement your own robotics cell, if you don't have the time to dedicate to the project, it will definitely not be successful and could delay your ability to recoup your ROI. In which case, using a reputable systems integrator may be the best way to go.

 

2. Empower your existing experts.

 

Your existing processes work — and they work well because you have good people who own and administer them. Ensure these individuals are closely engaged with the implementation project so the new, automated process can build on the success of the existing process, while improving on deficient areas. Small details like occasional process inconsistencies can throw a big wrench in an automation project. The team members in charge of the manual process will be able to help head those issues off at the pass.

 

3. Identify a robotics champion at your company.

 

In a way, the work really starts after a robot implementation is complete — i.e., when your team starts to work with the new equipment. To make certain that rapid ROI and ongoing success, identify an in-house champion who will work alongside the implementation team and learn the system. Make sure, too, that this person has real cross-departmental authority and can broker engineering and production cooperation, which will be critical to success.

 

4. Keep it simple.

 

Introducing a robot or robots into your facility is an essential change. There are quite a few variables to any project of any scale — e.g., appropriately converting a manual cell, training key personnel, and minimizing the impact on production. Start your robot implementation simply and take that principle to heart as you begin to evolve how you use robots in your facility.

 

TRONSERVE

 

Find a New Market

 

Henry Ford was not a newcomer to the car business when he began producing the Model T in 1908.  He had already been part of several automotive companies before the Ford Motor Company was established, and built different other car models including his Quadricycle and the 999.  But he imagined of a vehicle for ‘the great multitude,’  and so the Model T was born.  Unfortunately,  the original Model T was still too expensive for most Americans.  When Ford began churning the cars out via assembly line, however, their price dropped significantly.  In 1909, when workers were still using traditional methods to piece the cars together, a Model T was priced at $825, and below 11,000 were produced.  But in 1916, three years once Ford started using assembly line production,  the Ford Motor Company manufactured over 500,000 Model Ts and sold each one for $345.

 

Automation can lead you to consider opportunities in ways you have not before.  A shift in production capabilities and costs allow you to reconsider your market from a new perspective. Larger productivity and efficiency equals a lower per unit cost that will change how competitive you could end up within your market.

 

Help Your Workers

 

Henry Ford really improved the lives of his workers. In the beginning of assembly-line production within the Ford Motor Co, most workers managed 9-hour days for about $12 a week. But the shift was awkward, the work was hard, and turnover was high. So Ford changed the work periods to three 8-hour shifts and doubled the worker pay through a bonus structure.  This, in turn, diminished his labor turnover and allowed for smooth, uninterrupted production of his cars.

 

Inspite of fears, current technology does not take away jobs. In fact, automation may help increase the number of skilled, high-paying jobs within the manufacturing sector, since an automated shop has need of better trained, higher skilled workers. This could be a win-win for an existing company as well as their current workforce if management offers educational reimbursement to employees who want to retrain to gain more skills: your employees gain 21st century skills that can allow them to remain with the company in a better-paid position, and you retain good employees who have a proven track record of reliability.

 

Decrease Waste, Increase Efficiency

 

One of the top benefits to Henry Ford’s assembly line was its increased efficiency. The Model T went through 84 individual assembly processes using interchangeable parts that were all mass-produced anywhere else and assembled into a single car by workers trained to do one particular job. While this appears like common sense today, it was a revolutionary idea in 1913, and one that increased worker productivity to such a level that the time to put a single Model T together dropped from over 12 hours in 1908 to 93 minutes in 1914.

 

Automation can make the same kinds of advances in efficiency for your company.  When Factory Five Racing, Inc was looking to decline the time it took to produce their hot rod trim kits, they turned to robotic automation. The change allowed them to produce a higher quality trim kit consisting of four sets of panels (four trunks, four hoods, and eight doors) in 24 minutes, down from 7.5 hours. Increased efficiency tends to improve throughput, reduces environmental impact due to lower energy use, and cuts costs.

 

Be Flexible

 

Henry Ford famously said, “Any customer can have a car painted any color that he wants so long as it is black.”  While this doesn’t imply a flexible business outlook, Ford’s assembly line was great in its flexibility. At the height of Model T production, there were eleven different bodies built upon the basic Model T chassis, including a racer, a snowmobile, a police wagon, and a woody wagon, and each one could be custom-fitted with any of thousands of accessories.

 

Flexible automation makes your plant to switch over to take on new processes without a full retrofit. This is made less difficult by thinking ahead about what challenges might occur down the road and preparing for them during the original design phase. For example, choosing robots with an open interface will allow for the later connection of third-party equipment for customized processes. This is where working with an experienced robotics/AI design firm can pay substantial dividends later.

 

Another famous saying is “The more things change, the more they stay the same.”  While Henry Ford’s assembly line and basic business practices have long been replaced by modern lean manufacturing and Industry 4.0, we can still apply Ford’s reasons for improving processes to today’s changing industrial landscape.

 

TRONSERVE

 

Apple's iPad series, which accounts for the biggest portion of the brand-name tablet shipments, may not receive upgrades as keenly as before in the next few years, while iPads' price cuts are only having limited effect on stimulating demand, Digitimes Research noted.

 

Global tablet shipments will see hardships remaining at above 130 million units in 2020. After 2020, the global tablet market will become an area of rigid demand with shipments to get smaller 2-3% every year, and by 2024, it will be challenging to keep volumes above 120 million units.

 

White-box tablet shipments will remain weak in 2020 as most makers' key products are small-size tablets that have been cannibalized by large-screen smartphones and Amazon's inexpensive tablets. Since many white-box tablet makers have begun turning to manufacture non-tablet products and customized tablets, more makers are likely to quit the regular tablet business in the next few years.

 

Shipments of tablets with above 10-inch display or using an in-cell touch solution will jump considerably in 2020 as Apple has replaced its inexpensive 9.7-inch iPad with a new 10.2-inch one, while panels makers have been keenly advertising their in-cell touch solution with advantages in production and pricing to replace GFF one.

 

Microsoft's Windows-based tablet shipments are anticipated to boost tremendously in 2019 and will see its shipment share rise to 5.2% by 2020, trailing closely behind Lenovo.

 

TRONSERVE

 

Distance learning programs have actually existed since the 1980s, but technological innovation, content scalability, and widespread mobile adoption have made possible the online degree program to become a competitive option for aspiring students. Long gone are the days of aggressive marketing tactics and empty promises made by degree mills and unaccredited for-profit universities. Today, a learner can enter in competitive bachelor's and master's programs at U Penn, Columbia, Johns Hopkins, NYU, and more. In many cases, the virtual programs can pass tuition savings of 25-50% to enrolled candidates. 

 

Take, as an example, the Georgia Institute of Technology. An elite public institution, Georgia Tech boasts proud figures of #35 for national university rankings and #8 for best computer science programs. Backed by a $2 million investment from AT&T and a tendency to democratize education by extending beyond physical class sizes, Georgia Tech spun up a Master of Computer Science degree provided asynchronously in a virtual environment. The total price tag? ~$7,000 USD.

 

Georgia Tech is absolutely not all alone. The University of Illinois Urbana-Champaign suspended their on-campus MBA program to transition into an iMBA online, enrollment in individual online courses is tracking upwards year-over-year, and 25% of universities are envisioned to fail in the next 20 years due to heavy losses incurred by pricy facilities and declining student registration.

 

The acceleration of online-based learning has its positive factors, but it is not a silver bullet. But, a set of next-generation improvements facilitated by artificial intelligence (AI) stands to totally change the virtual experience.

 

The AI Advantage

 

To improve the current teaching model in which the teacher is the source of knowledge and the student is the recipient, we need to fundamentally reimagine the role of an educator in the university system. Advances in automated assignment grading and remote monitoring services (e.g. Proctorio) allow instructors to forego repetitive, time-intensive tasks and instead dedicate their saved time to higher-value work. For students that do not thrive in the regular classroom setting, AI-enabled learning management systems (LMS) can deploy surveys to categorize individuals into distinct learning buckets (e.g. visual, auditory, text), which can provide effective and targeted content that fits with each preferred learning style. Apart from just identification of preferences, the platforms can also break down long-form lectures and reading assignments into smaller, atomic components that are conveniently digestible.  

 

For international students, the language barrier may complicate progress, but cutting-edge research in text translation and machine learning seeks to create deep-learning systems that can translate English lectures into the student’s native tongue. Similar technologies in voice recognition and text summarization can transcribe a complete lecture with stunning accuracy and minimize paragraphs of text into just the relevant bullet points for review. Machine learning algorithms can equally be deployed over a course curriculum to flag areas of bias, complexity, and ambiguity for much closer review by the instructor. 

 

The climb is slow, but students, teachers, and administrations will without a doubt enjoy the benefits of AI as it evolves and matures over the next decade. The question of whether this field will be disrupted by a tech company, university, or research organization is still up for debate.

 

TRONSERVE

 

Neural networking is a technology that copies the human brain in that it is able to distinguish complex patterns. Keeping that in mind, Siemens says that by adding AI via neural networks to traditional control programs - which were designed to execute a set task - the capabilities of the system can be extended to change based on the parameters of the product or process. Bottom line: machines turned out to be naturally flexible.

 

A year ago at the SPS/IPC/Drives showcase in Nuremberg, Germany, Siemens announced a module that will integrate AI capabilities into the company’s Simatic S7-1500 controller and the company’s ET 200MP I/O system. This year, at PACK EXPO Las Vegas, Siemens unveiled the offering in the U.S., setting the foundation for the future portfolio that will enable AI throughout all levels of SiemensTotally Integrated Automation (TIA) architecture, which is a combination of hardware and software that links everything together seamlessly. The goal with TIA is to apply AI within applications that span from Siemens’ MindSphere, a cloud-based Internet of Things (IoT) operating platform, out to the industrial edge and even to the controller and field devices.

 

With the launch of the S7-1500 TM NPU (neural processing unit) module for the Simatic S7-1500 controller and the ET 200MP I/O system, Siemens has brought AI right to the controller. The S7-1500 TM NPU module is more appropriate for use on the field level at the machine – and wherever reliable, fast, deterministic decisions are required — as it enables the transfer of human expertise to the machine through training, the company said.

 

“With artificial intelligence we are able to train, recognize, and adjust to allow more flexible machinery,” said Colm Gavin, Factory Automation digitalization specialist at Siemens during a press conference at PACK EXPO. “Because, do we want 10 machines to package 10 different types of products, or a tool that accommodates different packages and different sizes and automatically adjusts to the new format?”

 

The S7-1500 TM NPU module works using a trained neural system on an SD card. Users can connect Gigabit Ethernet- and USB 3.1-compatible sensors like cameras and microphones to the module’s integrated interfaces. CPU data transmitted by the backplane bus can also be used as input data. The processing results then are assessed in the CPU program.

 

In packaging, for instance, bottles are coming down a conveyor belt quickly and if the system is trained for pass/fail, the moment something goes out of tolerance, it will fail. But if using AI, where the camera is trained with neural networks to recognize a billion pictures of every possible combination, the system will be able to figure out the rules on its own.

 

According to Gavin, the benefits of the S7-1500 TM NPU module are:

 

Flexibility - it makes handling unknown objects easy without resource-intensive programming.

Quality - Expert knowledge for fast and reliable quality checks is transferred directly to the module thanks to the higher-level training of a neural network.

Greater efficiency - Machines can respond flexibly and automatically to situations that used to require manual intervention, which reduces downtime and increases availability.

Cost-effectiveness - The module makes it possible to detect problems in production early on and avoid the cost of having to rework or even discard the product.

 

Applications in the areas of robotics, quality assurance, and condition monitoring are certainly suitable for the TM NPU module, but applications are limited only by the user’s imagination. At PACK EXPO, Siemens was demonstrating a robot with “flexible grasping” using AI, which looks at a shape and calculates the optimal point the gripper can pick it up (see video here). Once it understands the best grasping point the AI tells it where to go.“You don’t need to program the robot as AI makes it possible to grasp arbitrarily shaped and positioned objects,” Gavin said, adding that the ability to mimic the human hand in manufacturing has the potential to be a very big business.

 

TRONSERVE