Wifi seeing through walls, Commercializing Breakthrough Technology and our Cars are Spying on us

September 13

This week we find out how to use WiFi to see through walls. We investigate how moving from discovery of a new technology to commercialization requires a range of unexpected problems to be overcome. We examine a new theory of human history that proposes that our ancestral line fell to only 1,300 individuals around 900,000 years ago. We almost became extinct. Finally we discover how our new cars are spying on us. The information that they collect and resell is astounding!

WiFi seeing through Walls

A team at UC Santa Barbara have developed a new way for WiFi signals to image still objects through walls. The technique also enable the reading (or imaging) of the English alphabet. Previously this was too difficult due to the complexity of the letters, principally due to the curves on letters.

This latest breakthrough builds upon the team’s previous work in using everyday radio signals for a range of applications such as crowd analytics, person identification, smart health and smart spaces. A range of startups have been launched using these technologies.

The team faced a few challenges. Imaging still images is challenging due to the lack of motion. A change of approach to focus on tracing the edge of objects allowed them to overcome the difficulty. When sending a wave at an edge point, a cone of outgoing waves emerges according to Keller’s Geometrical Theory of Diffraction. This is known as a Keller cone. The team discovered that this works on sharp edges and a broader set of surfaces with a small enough curvature.

The edge orientation causes the cone to leave different footprints on a receiver grid. A mathematical framework that uses the conic footprints as signatures to infer the orientation of object is used to create a map of the scene. The English alphabet application allows them to read words through walls. This application is not exactly spy grade as yet however it is just the start.

Commercializing Breakthrough Technology

We talk about a lot of new breakthroughs here and sometimes we wonder why it takes so long for us to see those breakthroughs in products that we use every day. The road from the lab to the shop can be a long and winding one.

Lithium-sulfur batteries are one example. First invented in the 1950’s, development has been slow. In 2017 batteries with up to 1,500 charge and discharge cycles were demonstrated however none are yet commercially available.

Lithium-sulfur batteries have 3 main advantages. They can store 2 to 3 times more energy in a given volume, they are lower cost and they don’t use critical elements such as cobalt and nickel. The problem however is that when scaled up to commercial size the performance of the batteries died off quickly. There were rapid declines in performance with repeated charge and discharge cycles.

The reason for this decline is the dissolution of sulfur from the cathode during discharge. This loss of sulfur hinders the battery’s performance during cycling.

A team at the US Department of Energy may have finally found a solution to this problem. A few years ago the same institute developed a catalytic material that essentially eliminated the sulfur loss problem. The catalyst has shown promise in testing however the mechanism was not understood. This has now been solved.

Now that they understand the mechanism the team can apply cutting edge characterization techniques. They found that the catalyst structure affects the shape and composition of the final battery upon discharge.Additionally a new technique developed at Xiamen University allowed the team to visualize the electrode-electrolyte interface at the nanoscale. Combined this will allow design of more effective cathodes.

The techniques may also able to be applied to next generation batteries such as sodium-sulfur (cheaper still and hold even more power per kilo). What is clear is that the batteries of the next decade will be very different to the batteries that we use today.

Human Ancestors nearly died out 900,000 years ago

We currently have a lot of evidence that modern humans evolved within the past 200,000 years. We believe that they spread out of Africa commenced about 60,000 years ago. We don’t know much about what happened before that time. Somewhere about 600,000 years ago our lineage split off from the Neanderthals and Denisovans. We know that after this split, both those lineages interbred with modern humans before dying off.

A new method for analyzing our modern genome has shed some further light on our origins. Populations that are of significant size and that are not inbred have genetic diversity. With the advances in decoding the human genome we are now able to investigate our lineage more closely.

The new work is based upon three genetic differences. If we have enough genomes (we do) it is possible to work out the ancestral states of different areas of chromosomes. If there is a variation that is only present in a set of closely related individuals it is likely a recent change generated from a common ancestor. We know the rate at which new mutations arise in modern humans. This allows us to create a molecular clock. The number of variations in a population is related to the population’s size. Smaller populations tend to be inbred as it is difficult to avoid mating with relatives. This leads to a loss of genetic diversity. Larger populations support higher levels of genetic diversity.

Putting all of this together the team was able to look at the variations in today’s genomes and used them to determine the existence of various ancestral states. By estimating how many different ancestral states were present at any one time they could estimate population.

The algorithm that was developed showed a huge drop in the population size in Africa about 930,000 years ago to roughly 1,300 individuals that contributed to maintaining the population. Please note that there may have been more individuals alive as some may never have reproduced, others may have moved away and formed ancestral lines that died out and do not form part of the modern human lineage.

1,300 individuals is a very small population and put our ancestors at risk of extinction. An equivalent sized animal population today would have them on the endangered species list.

The authors traced diversity back even further and they estimate that some event wiped out 98.7% of the population in a sudden collapse. It is likely that the population remained small for the following 100,000 years. There is very little fossil record of this time (maybe because of the tiny number of individuals alive). The release of whatever environmental pressure restricting the population size may have allowed our ancestors to expand to new habitats and form distinct new populations. There is also evidence that the Neanderthals and Denisovans split off from modern humans around the time.

The collapse of population in this way would have had a significant impact on our development and may even have driven the evolution of important features of modern humans such as brain size. All of this is well informed speculation at this stage. Hopefully we will learn more as fossils are discovered and new techniques improve the algorithms.

Your Car and Data Privacy

Mozilla (a software company founded in 1998 by former members of Netscape) has just completed an analysis of the data that you nice new car is keeping (and reporting) on you.

The company tested 25 different brands and found that every major car brand fails to adhere to the most basic privacy and security standards. For example Volkswagen reports if you have fastened your seat belt and how hard you brake. BMW, Ford, Toyota, Tesla and Subaru collect data about race, facial expression, weight, health information (so does Google by the way) and where you drive. Some cars collected data about sexual activity and immigration status.

Your car is not a private space. Modern cars use microphones, cameras and the phones that we connect to our cars to harvest data. Manufacturers also collect data through their apps and websites. All of this data is then sold to third parties.

The worst offender was Nissan. Their privacy policy suggests that they collected information about sexual activity, health diagnosis and genetic data. They then reserve the right to sell “preferences, characteristics, psychological trends, predispositions, behavior, attitudes, intelligence, abilities and aptitudes” to data brokers, law enforcement and other third parties. Kia’s privacy policy reserves the right to monitor your sex life. Mercedes Benz ships cars with TikTok preinstalled (hello Chinese Government).

BMW allows customers to make granular choices on what data is collected (how many of you have even tried to read that in your service manual or online and have tried to turn data collection off) and they say they don’t resell in vehicle data to third parties. Subaru says that by being a passenger in a car means that you have consented to their data collection. Toyota has 12 different policies regarding privacy. It is unlikely that anyone has ever read all of them.

Just remember, always behave as if someone is watching, because they are!

Paying it Forward

If you have a start-up or know of a start-up that has a product ready for market please let me know. I would be happy to have a look and feature the startup in this newsletter. Also if any startups need introductions please get in touch and I will help where I can.

If you have any questions or comments please comment below.

I would also appreciate it if you could forward this newsletter to anyone that you think might be interested.

Till next week.