BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
D
Deleted member 118
Guest
Too small to read on my phone, but did see Brainchip mentioned in this book
Industrial Networks and Intelligent Systems
This book constitutes the refereed proceedings of the 7th EAI International Conference on Industrial Networks and Intelligent Systems, INISCOM 2021, held in Hanoi, Vietnam, in April 2021.The 39 full papers were selected from XX submissions and are organized thematically in tracks on...
books.google.com.au
Last edited by a moderator:
alwaysgreen
Top 20
Well I hope he is pushing Akida into as many products as possible at motherson!
Fact Finder
Top 20
Christmas 2024 that leaves plenty of time to leave the country so I will go with approximately $20.00 on the basis that all the automotive engagements have started to kick in revenue and full year 2022 closes with Brainchip achieving going concern status.With all this dot joining going on one requires a prediction on how many numbers there will be before the dot in our share price, let's say by end 2024
My opinion and speculation only so DYOR
FF
AKIDA BALLISTA
Fact Finder
Top 20
That'll do very nicely, thankyou FF. If this year is anything to go by then Christmas 2024 will be upon us before we know it
Fact Finder
Top 20
Hi @equanimous6G communication is evolving the current communication networking system smartly. Nayak & Patgiri [1], [2] envision the diverse future possibilities of 6G application in various domains, where 5G is the foundation for building the next generation of communication technology; for example, intelligent vehicles, Internet of Things, and savvy urban areas [10]. Although 5G has yet to be deployed worldwide and get experience at this time, researchers have turned their attention to the 6G communication framework [11]. In addition, the researchers have started working on integrating the 6G communication networks with AI’s capabilities to provide ubiquitous and reliable communication. Thus, it is becoming the backbone of the digital transformation of society [9], [12]. 6G communication technology provisions the connection with everything by integrating various technologies and applications. It also supports holographic, haptic, and underwater technologies. It will strengthen the Internet of Everything such as the Internet of Industrial Things, Internet of Medical Things, Internet of Nano-Things, etc. [7], [8], [13]. Therefore, the 6G communication technology can fulfill its promises with the help of advanced AI.
2. 6G core technologies
6G communication is all about sixth-sense communication. It will be a three-dimensional technology, particularly time, space, and frequency. 6G will truly be an artificial intelligence-driven communication.
AI-powered network management to control the fast monitoring system. Implementing several cloud services, network virtualization, and other individual components could simplify some activities. Still, it leads to more fragmentation in the long run and an increased need to regulate tools. 6G will help us to build and modify that advanced toolkit. Edge AI over management system would handle the complex integration in a hierarchical order [5].
A vision on the artificial intelligence for 6G communication
The 6G communication network will be a sixth-sense next-generation communication network, which will increase the worthiness of the intelligent Intern…www.sciencedirect.com
I have no doubt that AKIDA technology has a place with 6G but as far as 5G is concerned I know there was a presentation involving Brainchip and another where the need for intelligent nodes in 5G networks featured.
I assume it must have been a Rob Telson podcast and at the moment leaning towards the DELL Technologies Rob Lincourt interview.
Logically though we do not need an industry expert to tell us that bandwidth is just like a highway.
You build a new highway and this just encourages people to buy a car and drive on it and eventually it becomes congested.
You provide bandwidth to people and they buy extra devices to connect to it and it eventually becomes congested.
Just like widening roads increasing bandwidth is expensive and time consuming.
So those who manage roads rather than adding lanes add buses to reduce the numbers of cars to allow traffic to flow and thereby increase capacity through reducing vehicle movements and encourage working from home and flexible work hours to distribute demand.
In the same way bandwidth is increased by using accelerators to compress data and offering different rates for off peak use.
This is where AKIDA comes in it allows the equivalent solution to working from home by processing at the 5G node and localising decision making reducing the need for data to move back and forth taking up bandwidth.
Decisions regarding routing which are constantly required are enormous consumers of bandwidth but can be removed almost entirely as any information AKIDA does need to exchange will be in the form of meta data and will likely not require a return signal.
Any form of networked communication system can be enhanced by localised node intelligence be it 6G, 5G, 4G, 3G or wireless.
My opinion only DYOR
FF
AKIDA BALLISTA
Fact Finder
Top 20
Why would Mercedes vehicles feature more prominently in this Blu Wireless video???
Perhaps Mercedes loaned them their private test track for the trial cause Anil Mankar asked nicely after all he charmed that warehouse guy at Samsung into lending Brainchip DVS cameras and fridges.
My opinion only DYOR
FF
AKIDA BALLISTA
Perhaps Mercedes loaned them their private test track for the trial cause Anil Mankar asked nicely after all he charmed that warehouse guy at Samsung into lending Brainchip DVS cameras and fridges.
My opinion only DYOR
FF
AKIDA BALLISTA
Fact Finder
Top 20
Just had a thought would secure
data transfer AKIDA chip to AKIDA chip via accelerated meta data be of assistance to Blu Wireless when servicing the connected needs of high speed transportation.
I do hate how AKIDA’s ubiquitous nature makes excluding it from any and all applications very difficult for an old technophobe.
My opinion only DYOR
FF
AKIDA BALLISTA
Fact Finder
Top 20
In 2018 according to Google there was over 64 million kilometres of paved roads in the world.
At 4 AKIDA chips to a kilometre that is a lot of chips Blu Wireless would need to cover the needs of automotive. Add in an AKIDA chip in every connected vehicle and that becomes a very large number???
My opinion only DYOR
FF
AKIDA BALLISTA
Maybe we could give them a bulk order discount
thelittleshort
Top Bloke
@Diogenese I literally have no idea what this means but is is good for BrainChip?
www.linkedin.com
François Piednoël de Normandie on LinkedIn: #idc | 10 comments
Intel is now in position it was just before the launch of Conroe. Because competitive pressure, it was forced to push the power envelope and increase… | 10 comments on LinkedIn
Fact Finder
Top 20
I am pretty sure he is saying Intel, AMD & others are on the wrong track. That they have ignored power use in favour of processing power which anyone can do but on their present trajectory they will be selling ovens not semiconductors because of how hot they are running because of the power they consume.@Diogenese I literally have no idea what this means but is is good for BrainChip?
View attachment 19749
François Piednoël de Normandie on LinkedIn: #idc | 10 comments
Intel is now in position it was just before the launch of Conroe. Because competitive pressure, it was forced to push the power envelope and increase… | 10 comments on LinkedIn
He is then asking can they work out how to reduce power consumption and by implication if not will they survive. Reducing power consumption for them is going to be difficult.
He is setting the stage ready for AKIDA to emerge from the wings to rapturous applause.
In my opinion only DYOR
FF
AKIDA BALLISTA
Dozzaman1977
Regular
Not sure , but the following is a subsidiary of TDK. Probably not Akida IP involved but posting so the great mind of Diogenese and others can have a look .
October 18, 2022
ICsense, a TDK Group Company and independent subsidiary that focuses on the supply of application-specific integrated circuits (ASICs) and custom chip design services, has started mass production of a new chemical sensor interface ASIC.
The custom IC was developed for inclusion in a wearable design and is now being produced in large volumes to meet demand requirements. This project was undertaken for a large US-based client. It is expected that several hundred million units will be shipped over the coming 4 years.
ICsense was selected for the project because of the company’s long track record of developing/manufacturing high-performance sensor readout ICs, alongside its vast industrialization experience.
“We are proud to announce the successful ramp-up of this high-runner,” explains Bram De Muer, co-founder and CEO, ICsense. “Playing a pivotal role in the end-product, the custom chip was developed exclusively for this client. The design needed to be ultra-low power, due to it being incorporated into a battery-powered system. Our team succeeded in reaching a power consumption level of only a few µA in full measurement mode, which is currently state-of-the-art for this type of sensor.”
The ASIC contains an ISO15693-compliant NFC transceiver. Data from the ultra-high-accuracy sensor readout is transferred wirelessly to mobile devices. The customer was looking to integrate all functionality into an ASIC, in order to lower cost and save area, as the wearable design had very limited space available.
ICsense has taped (i.e. sent to the foundry) over 150 designs since its foundation in 2004. About 80% of these designs are for high-reliability markets such as medical and automotive. As the most prominent ASIC design house in Europe, ICsense employs close to 100 in-house design engineers at its headquarters in Leuven, Belgium and design center in Ghent.
“We see a steep increase in the number of ASIC requests, also as a result of the global chip shortage and accelerating digitalization and focus on customer experience. Customers are looking for supply certainty and develop their own exclusive ASIC,” De Muer continues. “In contrast to value chain aggregators or design houses, we have all the necessary expertise in-house to support mass production. This includes analog and digital design teams, test and product engineers and full in-house ATE test development capabilities.”
About ICsense
ICsense, an independent subsidiary of the TDK Group, is Europe’s premier IC design company. ICsense’s core business is ASIC development and supply and custom IC design services. ICsense has the largest fab-independent European design group with world-class expertise in analog, digital, mixed-signal and high-voltage IC design. The company develops and supplies customer exclusive ASIC solutions for the automotive, medical, industrial and consumer market compliant with ISO9001, ISO13485, IEC61508-ISO26262.
Last edited:
Fact Finder
Top 20
A quick look on their website I can only find analogue processing being discussed but as we know the NDA’s mean those who want them keep the whole thing under wraps.
For me it is a no evidence beyond the like by Chris Stevens but as RobbTelson said as he sat listening to all the presentations all he could think was AKIDA could improve this solution.
Perhaps Chris Stevens is similarly afflicted.
My opinion only DYOR
FF
AKIDA BALLISTA
That's why the ogre retired.
thelittleshort
Top Bloke
Fact Finder
Top 20
I think the following linked article is probably the starting point for François's post:@Diogenese I literally have no idea what this means but is is good for BrainChip?
View attachment 19749
François Piednoël de Normandie on LinkedIn: #idc | 10 comments
Intel is now in position it was just before the launch of Conroe. Because competitive pressure, it was forced to push the power envelope and increase… | 10 comments on LinkedIn
My opinion only DYOR
FF
AKIDA BALLISTA
PS: Definitely worth a read it reinforces why being like Fonzie is best:
Last edited:
Hi Tls,@Diogenese I literally have no idea what this means but is is good for BrainChip?
View attachment 19749
François Piednoël de Normandie on LinkedIn: #idc | 10 comments
Intel is now in position it was just before the launch of Conroe. Because competitive pressure, it was forced to push the power envelope and increase… | 10 comments on LinkedIn
Transistors have internal resistance, so, when a current flows, heat is generated. If we assume that each operation of a digital transistor switch (CMOS) in any specified geometric chip size uses the same amount of power no matter what chip it is in, then more TOPS means more power. Increasing the clock speed increases power consumption. In an IC, heat dissipates more slowly than it is generated, so there is an accumulation of heat. So to reduce power while obtaining an equivalent outcome means that the number of transistor switch operations must be reduced.
Pumping cooling fluid around and blowing air from a fan through a radiator much like a car's cooling system uses more power.
There are a couple of ways of reducing the number of switch operations, eg, not implementing multiply by zero operations in MAC, or lowering the voltage supply (which has the effect of reducing the signal-to-noise ratio (increasing error-proneness)).
However, these savings are small beer compared to the use of spikes instead of 8-bit or upward bytes to implement a function.
The article is talking about an Intel CPU, not Loihi, so it would be using 64-bit MACs or higher - very high precision, but very power-hungry.
Before Akida, most hardware SNNs used analog ReRAM/MemRistor because it is more closely analogous to real neurons. However, human technology lacks the millions of years of evolutionary refinement which lead to the human brain, and the manufacture of analog neurons lacks the precision necessary to derive consistently accurate outcomes. The performance of analog neurons is too variable, and considering that millions of neurons are included in an SNN, the accumulated errors can be significant. While some attempts to overcome this problem have been made, they generally end up using more power.
As we've discussed before, the discovery by Simon Thorpe's group that the receptor cells in the eye respond more quickly to stronger input signals, so most of the visual information is included in the earlier receptor cell signals. This can also be applied to the pixels of a camera's light sensor.
This lead to N-of-M coding in which only, say, 10% of incoming pixel data needs to be processed by choosing the first 10% of incoming spikes.
A lot of work of a CPU or GPU is image-related in that, apart from any image/video processing, the processor must control the screen display. I wonder if there is a role for Akida there?
Similar threads
