Hi Friends,

Even as I launch this today ( my 80th Birthday ), I realize that there is yet so much to say and do. There is just no time to look back, no time to wonder,"Will anyone read these pages?"

With regards,
Hemen Parekh
27 June 2013

Now as I approach my 90th birthday ( 27 June 2023 ) , I invite you to visit my Digital Avatar ( www.hemenparekh.ai ) – and continue chatting with me , even when I am no more here physically

Translate

Saturday, 14 November 2026

f ISRO Can Track a Train, It Can Score a Car

 

If ISRO Can Track a Train, It Can Score a Car


Abstract: 


On 13 Nov 2026, PM Modi told Mysuru that ISRO and Indian Railways jointly track trains by satellite. Five days earlier I had proposed a terrestrial telemetry stack for ROTE, because NavIC alone can no longer give a reliable fix. The railway system does not use NavIC at all. It uses GAGAN for position and an ISRO satellite link for reporting. That makes it a working, nationwide proof that satellites can feed my Harm Quotient. It should be a layer in the ROTE stack, not just a rural fallback.

======================================================================

Dear Shri Gadkariji,

Today, while inaugurating the COSMOS planetarium at the University of Mysore, our Prime Minister listed the quiet ways ISRO serves ordinary Indians.

Fishermen get satellite warnings before a cyclone. Satellite images help settle crop insurance claims under PM Fasal Bima Yojana. And ISRO and Indian Railways jointly run a satellite-based system that tracks train locations in real time, so a passenger can check where the train is and when it will arrive.

My first thought on reading this was simple: if a satellite can track a train, it can track a car.

And if it can track a car, it can feed ROTE.

WHAT IS ROTE? (FOR NEW READERS)

ROTE stands for Reduction of Transport Emissions. The idea, which I have been developing since 2023, works in three steps:

  1. Harm Quotient: every vehicle gets a pollution footprint, weighted by energy source, age, capacity, location, time of use and driving conditions.
  2. Trans-Score: the monthly Harm Quotient readings roll up into a 0 to 100 eco-behaviour index.
  3. Trans-Tax: a score of 75 or above earns a monthly credit, and a score of 25 or below pays a monthly penalty. Settlement happens through a FASTag-linked e-wallet.

The weak link was always this question: how do we know where, when and how each vehicle was driven?

WHERE I LEFT IT, FIVE DAYS AGO

In my last post (8 Nov 2026) I argued that NavIC no longer has enough working satellites for a standalone fix. So I proposed a terrestrial hybrid: in-vehicle sensors, V2V pings, cellular positioning and a map-matching engine. In that design, satellite positioning was kept only as a fallback check on sparse rural highways.

The Prime Minister's speech made me revisit that last point.

HOW TRAINS ARE ACTUALLY TRACKED

The satellite does not "see" the train from space. The system is called RTIS (Real-Time Train Information System). It went live on its first routes in January 2019.

  • A box on each locomotive works out the train's position and speed using GAGAN, the satellite-augmentation system built by ISRO and the Airports Authority. GAGAN sharpens the GPS signal over India.
  • Every 30 seconds or so, the box reports through an ISRO communication satellite, with 3G/4G as a backup, to a central server at CRIS.
  • That server plots the control charts and feeds the National Train Enquiry System. No station master has to phone anything in.

Note what is not in this recipe: NavIC.

So the NavIC shortfall that pushed me towards a terrestrial design does not touch the GAGAN-plus-satellite-link route. Railways have been running it across the country for years.

WHAT THIS CHANGES FOR ROTE

I would now upgrade the satellite layer from "rural fallback" to a full partner in the ROTE stack:

  1. Position: GAGAN-augmented GNSS sits alongside the in-vehicle sensors and cellular positioning. It is not just a back-up for them. Three independent sources make a Harm Quotient much harder to fake than two.

  2. Reporting: Where there is no mobile signal (ghats, deserts, border roads), a satellite link can carry the signed Harm Quotient reading home, exactly as RTIS does for locomotives.

  3. Start with fleets: Trucks, buses and taxis already carry trackers under AIS-140, and those trackers already talk to a central server. That fleet can be the first ROTE pilot, before a single private car is touched.

  4. Borrow the RTIS architecture: CRIS has already built a device, a satellite hub and a central server for one national fleet. A ROTE pilot could copy that blueprint rather than start from zero.

THE HONEST CAVEATS

  • Trains are easy, cars are not. About 13,000 trains run daily on fixed tracks under one owner. India has over 30 crore registered road vehicles, owned by individuals and free to go anywhere.

  • Privacy. A locomotive has no right to privacy; a citizen does. This is why ROTE must keep sending only the signed Harm Quotient reading to the centre, never the raw location trail.

  • Signal loss. Flyovers, tunnels and dense city blocks confuse satellite receivers. This is exactly why the terrestrial layers from my last post stay in the stack.

  • Two-wheelers. They are the majority of our vehicles. The device and its cost must suit them too.

MY REQUEST

  1. Ask CRIS and ISRO to share the RTIS architecture with MoRTH as a template for a ROTE pilot.

  2. Fold a GAGAN plus satellite-link layer into the Chennai 1,000-vehicle pilot I proposed on 8 Nov.

  3. Begin with AIS-140 commercial fleets, which already report their position.

The Prime Minister spoke of satellites "reaching great heights in space".

I only ask that they also help us breathe cleaner air down here on our roads.

My earlier posts on ROTE:

  1. Terrestrial Vehicle Telemetry for ROTE / Harm Quotient / Trans-Tax (08 Nov 2026) Keep the Harm Quotient framework, and replace its NavIC input with a sensor, V2V and cellular hybrid. https://myblogepage.blogspot.com/2026/11/errestrial-vehicle-telemetry-for-rote.html

  2. Dear Shri Gadkariji: Let GPS-Based Toll Tax Lead to Harm-Quotient Based Trans-Tax (Nov 2023) [link to add]

  3. Pigovian Trans Tax? We Are One Step Closer. ([date]) [link to add]

Sources for this post:

With regards,

Hemen Parekh

www.HemenParekh.ai / www.Iam-Immortal.ai / www.YourContentCreator.in

Mumbai

13 Nov 2026

Posted by No comments:

Friday, 13 November 2026

Let every Indian have a Praja Doot




==============================================

 To: 

Shri Ashwini Vaishnaw, 

Hon'ble Union Minister of Electronics & Information Technology, Railways, and Information & Broadcasting 


Subject: 

Spam is carpet bombing. Let every Indian have a Praja Doot to guard the gate


==================================================

Respected Vaishnawji,


Yesterday's TRAI figures for July to September 2026 read like a war report.


In just one quarter, TRAI's AI systems flagged 19.9 billion suspected spam

calls and SMS (18.28 billion calls and 1.62 billion messages). That is more than

21 crore attempts every single day. Citizens filed 15.29 lakh complaints, and

Vodafone Idea topped the list with 5.39 lakh. TRAI barred 1,10,325 numbers for

15 days, including 47,249 on Vodafone Idea, disconnected 13,777 repeat

offenders for a year, and blacklisted 374 senders.


These are serious efforts, and TRAI deserves credit for them. But the picture that

comes to my mind is the carpet bombing of the Second World War. Thousands of

bombers fly over a city, and the defenders shoot down a few hundred. The city

still burns.


Barring 47,000 numbers removes a few bombers after they have already struck.

Spam works because it costs the sender almost nothing, and the entire cost of

the annoyance, the lost time and the fraud falls on the person who receives it. As

long as that equation holds, a new number will replace every barred one by the

next morning.


A different defence : a Praja Doot for every Indian


Today the telco and the regulator stand between the advertiser and the citizen,

and they are always one step behind. I suggest we move the gate to where it

belongs: with the citizen.


IImagine every Indian having a personal AI agent, a Praja Doot (people's

messenger), that answers on their behalf. 


Every brand, bank, builder or insurer would have its own agent, a Raja Doot.

Nothing reaches the citizen unless the two Doots have first negotiated it:



  • The Raja Doot states what it is offering, why it is relevant, and what it

  • will give in return (a discount, cashback, or even a small payment for the

  • citizen's attention).

  • The Praja Doot checks this against the citizen's own rules. For example, no

  • loan offers ever, property offers only after 6 PM on weekends, health

  • insurance only from my existing insurer, and none of my personal data shared

  • without asking me.

  • If the terms fit, the message is delivered. If they do not, it is politely refused,

  • and the citizen never even hears the phone ring.

Carpet bombing then turns into something closer to a market. Advertisers stop

paying for volume and start paying for consent. The citizen is no longer a target.

The citizen becomes a party to a deal.


No app to download, no smartphone needed


This must not become one more app that only the young and the urban can use. 

A Praja Doot would not live on the citizen's phone. It would live in the network, 

run by the citizen's telecom operator or on a public DPI platform, and it would

sit on every incoming call and SMS before the phone even rings.


So it would protect a ₹1,000 feature phone exactly as well as a costly smartphone.


Setting it up would follow, more or less, the same steps by which our social benefit

schemes already reach crores of families through Direct Benefit Transfer:


  1. Enrol once. 

  3. The mobile number is already linked to the citizen's identity through the

  5. telco's KYC, so the Praja Doot is switched on for that number, just as a bank

  7. account is seeded with Aadhaar to receive DBT.


  10. Tell it your wishes, in your own language. 

  12. The citizen dials a toll-free number (or the existing 1909), sends an SMS,

  14. uses USSD, or chats on WhatsApp, and simply speaks in Hindi, Tamil,

  16. Odia or any other language :


  19. " No loan calls. Property offers only on Sundays. Nothing from gaming apps "


  22. Bhashini turns this into rules for the agent. 

  24. A Common Service Centre, a post office or the telco's own store can help

  26. those who want a human hand.



  30. Consent is recorded, not assumed. 

  32. Like the consent screen in DBT and Account Aggregator flows, every rule the

  34. citizen sets is recorded, can be seen at any time, and can be changed with

  36. one call or SMS.


  39. Delivery only to the verified. 

  41. Just as benefits reach only a verified, seeded account and not a middleman,

  43. offers reach the citizen only from a verified Raja Doot that has agreed to the

  45. citizen's terms. Everything else is stopped at the network.


  48. A grievance line. 

  50. If anything slips through, one call or SMS to report it penalises the sender's

  52. Raja Doot automatically, much as the DND complaint system works today, but

  54. with real teeth.

For the 112 crore subscribers who have never filled in a DND form, the default

could be a sensible, protective rule set that is switched on automatically. The

citizen does nothing, and is protected anyway.


See it working


To make the idea concrete, I have prepared a clickable mockup of this whole

agentic workflow on an ordinary feature phone. It shows switching on with

*1909#, speaking the rules in Hindi, five Raja Doots negotiating with one Praja

Doot, the offer that finally reaches the phone, the consent ledger, and a one-key

spam complaint:


Praja Doot: a clickable walkthrough



What it would take


India is better placed than any country in the world to build this, because we


already have the Digital Public Infrastructure on which it can stand.




  3. Verified identity for every Raja Doot. 



  6. Only agents of businesses verified through the existing DLT registration, GST,


  8. or a DigiLocker or UPI-style credential would be allowed to negotiate. 


  10. An anonymous spammer would not even get a seat at the table. 


  12. TRAI has already put 34,025 senders through e-KYC this quarter, so


  14. the beginning is there.




  1. A machine-readable consent protocol. 


  3. Only 22.4 crore of our 134.9 crore subscribers, about 17 percent, have


  5. registered any preference on the DND registry. The other 112 crore have not,


  7. not because they welcome spam, but because a form with a few checkboxes


  9. is too blunt a tool. The DND registry and the DLT consent framework could


  11. grow into an open protocol, much like UPI did for payments, through which


  13. Praja Doots and Raja Doots talk to each other in a standard language. 


  15. MeitY, TRAI and DoT could develop it jointly under the IndiaAI Mission.




  19. Loyalty of the Praja Doot to the citizen. 


  21. This is the most important condition of all. A Praja Doot built by a platform


  23. could quietly serve the platform and sell the citizen's attention to the highest


  25. bidder. The law must say clearly that a Praja Doot owes its duty to the  


  27. citizen alone, and the Digital Personal Data Protection Act gives us a natural


  29. place to say it, perhaps by recognizing such agents as a kind of :



  1. "Consent Manager", acting for the Data Principal .



  4. Indian languages from day one, 


  6. through Bhashini, so that a farmer in Odisha can tell her Praja Doot in Odia


  8. what she does and does not want to hear about.



Why this is also a privacy issue


I have been writing about privacy on my blog for many years. Back in 2018, in


"Parekh's Law of Privacy,"  I wrote that "concern for privacy increases in direct


proportion to the amount of personal information which a person wants to hide


from the public."



Spam is the other side of that coin. Every unwanted call is proof that someone has


our number, our name, and often our loan history or our property search, without


our consent. A Praja Doot would reverse that flow. Instead of brands holding


data about us, our own agent would hold our preferences for us and


reveal only what we choose, only to whom we choose and only in exchange for


something we value




Dear Shri Vaishnawji, 



With UPI, Aadhaar and DigiLocker, India has already shown the world that public


digital rails can serve 140 crore people. 



Let us now build the rails on which every Indian can say "yes" to the  


offers they want and a quiet, firm "no" to everything else.



Let us end the carpet bombing not with more anti-aircraft guns, but by giving



every citizen the power to decide who may fly over their home.




With regards,


Hemen Parekh 


www.Iam-Immortal.ai  /  www.IndiaAGI.ai  /  www.YourContentCreator.in


================================================


Sources




References from my blog


Posted by No comments:
Subscribe to: Posts (Atom)