Communicating with the Beamcars

Who is General Failure and why is he reading my hard disk?

Anfang he beam cars need to be in continuous contact with the computers that are controlling traffic on the network.
The task of communicating with these moving cars from stationary computers consists of 3 parts:

The physical equipment needed to facilitate communication

The form that this communication will take, i.e. the communication protocol.

The type of informaton that needs to be exchanged.
On this page, we will deal with:
The Principle for Communication in the Beam Network
Waveguides inside the beams
Using Microwave Antennas
Using Cellular communication with Hand-over.

How Communication is Controlled
What is Communicated?
Bluetooth is discussed on a separate page. Overview of communication needs with a beamcar

he beam cars need to be in continuous contact with the computers that are controlling traffic on the network. The task of communicating with these moving cars from stationary computers consists of 3 parts: The physical equipment needed to facilitate communication The form that this communication will take, i.e. the communication protocol. The type of informaton that needs to be exchanged. On this page, we will deal with: The Principle for Communication in the Beam Network Waveguides inside the beams Using Microwave Antennas Using Cellular communication with Hand-over. How Communication is Controlled What is Communicated? Bluetooth is discussed on a separate page.

Anfang hose vehicles carrying passengers need to provide for the needs of those passengers as well. Thus, the communication between the beamcars and the outside world would be of 3 different kinds:

Control- and Trafficinformation between vehicles, nodes and the central computer.

Traffic information for the travelers, coming from the central computer to be shown on displays in the beamcars. It could be information about delays, altered travel routes, expected time of arrival at destination, etc.

Booking information; prompted by travelers waiting for booked cars. Also, persons waiting at stops would press buttons on a terminal to indicate where they want to go, and the next scheduled car with empty seats which is going that way, would announce to passengers aboard (as well as to the persons waiting at the stop) its intention to stop and pick them up. This decision is made by the beamcar itself, without interference from any node computer.

It would also be feasible to provide the travelers with:

Connection to the telephone network and with the Internet
A choice of radio and TV channels, by way of earphones and TV monitors

This page is a rather perspicuous description of the communication needs. It describes a few more or less practical ways of realizing this. For more details, see other pages.

1. The Principle for Communication in the Beam Network

he "best" way to control an automatic traffic network like the one we are talking about here, is to use the point-synchronous traffic system. In this traffic system, the various beam segments are functionally divided into 2 traffic categories:

Asynchronous traffic; Trunkbeams and beams outside the nodes´ booking points
Synchronous traffic; beams inside booking points, stops, depots, etc.

It is proposed that when beamcars run in asynchronous traffic mode, they have a choice of communicating with either the Central Computer or with the next upcoming node computer in their path. When they run on beam segments for synchronous traffic, they have to communicate with the local node in charge of the area where the beamcar in question happen to be.

Every part of the synchronous network is under supervision by some node or another, i.e. the nodal computers have to know, at all times, which cars are in their allotted segments, the status of those cars and where they're headed (if they are running). The beamcars have to handle all their communications by way of this assigned node computer.

Since the cars are mobile, their communication has to be done by way of modulated carrier waves. The nodes could use individually fixed transmission frequencies. But the cars would be too many (in a large network) to be alloted individual transmission frequencies, and this would also be unnecessarily complicated for the nodes to handle. There are 2 ways to deal with this:

All cars communicating with a particular node would use a common transmission frequency that gets assigned by that node.
Each node has a pool of frequencies that it would successively hand out to arriving cars, for them to use as transmission frequencies. As the cars leave the area, those frequencies would be returned to this pool and be available for use by other arriving cars.

The first method is the simplest, but it has potential shortcomings. If the cars in an area are using one, common, carrier frequency for their transmission, there would be situations when more than one car wanted to transmit at the same time. In short beam segments the communication needs might be too urgent for the cars to resort to collision detection and similar data communication schemes, which requires that they wait out each other when they want to transmit simultaneously. This could result in cars having to slow down or stop until they get their chance to communicate. Thus, this could lead to traffic congestion; not very satisfactory! It should be emphasized, though, that transmission times are in the range of milliseconds, so this situation would only be annoying if a node has to handle more than about a hundred active cars, that want to transmit at about the same time.

The second method means that all cars announcing their approach to a node would be assigned a temporary carrier frequency on which to transmit. This assignment would be valid only while the car is in that particular beam segment (i.e. while the car is controlled by that particular node). The pool of available frequencies would have to be big enough to meet all occasions. That should not be a problem.

This method means that all cars in the area could transmit simultaneously, which would save lots of time, compared to the first method. The node computer would then buffer these messages and deal with them chronologically, and it is possible that some parallel buffering and processing in the node computer might be necessary in traffic-intensive areas.

These are 2 ways to handle the distribution of frequencies. We will next take a quick look at 3 ways to handle the physical transmission and reception.

1.1 Waveguides inside the beams

Figure 2	ne proposed solution to the communication issue, is to let every vehicle's propulsion car (= that vehicle that is inside the beam) have a small microwave antenna on its top, as is illustrated in figure 3. This antenna would at all times be positioned underneath a waveguide, which would be mounted in the ceiling of the beam, and having a slit underneath. In that respect, the waveguide would look like a miniature of the beam itself, with the antennas travelling along underneath these slits as the beamcar moves. Each beamcar´s reception would take place by way of the leakage field through the slit, and the beamcar´s transmission would likewise go through the slit and into the waveguide, possibly using a different carrier frequency. As can be seen in figure 4, these waveguides would have to be broken off into segments at all shunts in the beam network. There are several reasons for this: to isolate the beam segments from each other, communicationwise, so that the beamcars only communicate with the nodes they are assigned to at the moment. to keep wave reflections and attenuation under better control to keep down the number of vehicles that at any moment are in the same segment.	Figure 3

Figure 2

ne proposed solution to the communication issue, is to let every vehicle's propulsion car (= that vehicle that is inside the beam) have a small microwave antenna on its top, as is illustrated in figure 3. This antenna would at all times be positioned underneath a waveguide, which would be mounted in the ceiling of the beam, and having a slit underneath. In that respect, the waveguide would look like a miniature of the beam itself, with the antennas travelling along underneath these slits as the beamcar moves. Each beamcar´s reception would take place by way of the leakage field through the slit, and the beamcar´s transmission would likewise go through the slit and into the waveguide, possibly using a different carrier frequency.

As can be seen in figure 4, these waveguides would have to be broken off into segments at all shunts in the beam network. There are several reasons for this:

to isolate the beam segments from each other, communicationwise, so that the beamcars only communicate with the nodes they are assigned to at the moment.
to keep wave reflections and attenuation under better control
to keep down the number of vehicles that at any moment are in the same segment.

Figure 3

Figure 4	Figure 5

Every such segment leading up to a node would be under the control of that local node's computer. Thus, if the traffic moves in the downward direction as in figure 2 (which illustrates just the waveguides, not the beams), the stationary antenna controlled by the node would be placed at the lower end of each waveguide segment (the antennas of vehicles and nodes are indicated by red dots). The cars indicated by 1, 2 and 3 would be communicating with node A, and the cars 4 through 7 with node B.

This scheme could be fairly complicated, as figure 5 gives a hint about. The red dots are stationary antennas in the waveguides, belonging to the node that is responsible for these shunts. Since it would also be responsible for the depots and stations in the area, it could be quite a number of antennas. The green dots indicate beam segments that would be under control of other nodes. The booking points for the next node could for instance be placed at some of those green dots. As can be seen, these segments could get interweaved quite a bit, it depends on how the network is designed.

1.2 Using Microwave Antennas

nother solution might be to dispense with the waveguides and use antennas with parabolic reflectors for the stationary transmitters inside the beams. All neighboring stationary transmitters would use different carrier frequencies to avoid interfering with each other.		Microwave antenna with parabolic reflector. Reflector Waveguide Open end that faces the reflector

1.3 Using Cellular communication with Hand-over

third solution would be similar to the cellular phone system. The beam network would be divided into cells, where each individual cell would cover a sufficiently large geographical area to include all beam segments that the node in the cell have to reach. Because of the nature of the beam network, these cells would be overlapping each other to a considerable degree. This is roughly (and brightly!) illustrated to the right, where yellow denotes overlapping areas, and orange denotes trunk beams where the cars have to communicate directly with the Central Computer by way of local antennas. The other colors, red green and blue, symbolizes 3 individual set of frequencies, so arranged that neigboring cells do not have the same set (i.e. the same color in the illustration).

Each node would have its own transmission frequency in the microwave-range, separate from those of its immediate neighbors. Each node would also have a pool of available frequencies in the microwave-range, to distribute to arriving cars so that every car gets its own transmission frequency for the duration of its stay in this particular node´s area. Neighboring areas would have different sets of such frequencies, so that they would not interfere with each other.

Since microwaves do not reach very far, areas that are not neighbors could well have the same set of frequencies. Actually, there are several alternative methods to handle this carrier-frequency business:

Showing overlapping wireless communication cells

All arriving cars receive their individual transmission frequencies, as outlined above.
The pool of available frequencies is limited to, say, 10 frequencies, which the beamcars would have to share in a brotherly fashion. Collision detection would be used to handle simultaneous transmissions.
The beamcars would have to share one common frequency and use Collision detection.

The Hand-over Procedure
Anfang n the cellular phone system, the quality of the carrier signal is the criteria for when it´s time for the traveling vehicle or person to switch to another base station. One way of doing this is for the base station to send a supervisory signal which gets returned by the mobile station. The signal-to-noise ratio of the returned signal is measured, and when this value dips below a treshold, an alert signal is sent to a base station higher up in the hierarchy. This higher base station orders the base station as well as the base stations in the surrounding cells to measure the signal strength of the mobile station. These mesurements are then forwarded to the higher base station for comparative evaluation. The base station that measures the strongest signal gets ordered to start handshaking procedures with the mobile station, and the mobile station gets ordered to switch channel to the new base station. When the handshaking procedure has led to an established connection with the new base station, the old base station is disconnected from this mobile station, and the connection as a whole is rerouted in the network from the old to the new base station. This procedure is termed "hand-over". The details may vary between different systems. In the beamtraffic system, the criteria for hand-over would be different. The signal strength and signal-to-noise ratio would have to be satisfactory everywhere, because of considerable overlapping of reception areas of neighboring nodes. The primary reason for this overlapping is that the nodes´ areas of jurisdiction are rarely circular. They are a function of the network structure, where various booking points (in the point-synchronous traffic system), station areas, etc. that belong to a particular node are situated.
Instead of carrier wave quality, the sensors in the beams would tell the traveling beamcar when it´s time to start talking to a new node. The car would then make an announcement on a frequency that the car knows is available for transmission in this particular area. The car also knows on which frequency to listen. This kind of information, for the whole network, would be stored in every beamcar´s computer memory. During the ensuing handshaking procedure, the node would, if required, tell the car to switch to another transmission frequency.
There is a considerable advantage to this system as compared to the ordinary cellular phone system. Beamcars would never have to switch base station in the middle of a conversation. They would have time to finish all information exchange with the old node before taking contact with the new node.

2. How Communication is Controlled

ommunication is conducted in the form of packets. The node attaches a receiving address on its transmitted packets (as well as its own sending address). All cars listening to this frequency can receive all such packets, but only the addressee would open and read its content. All cars would likewise add sender and receiver addresses to its transmitted packets, and, although other nodes might receive such packets, only the destination node would read them. All such transmissions would be encrypted, this being the best way to foil ill-intentioned interferens. There are other methods available to avoid such interferens, but they fall outside the scope of this webpage.

As stated elsewhere, messages from beamcars could be destined for the Central Computer or for other beamcars, in which case the node would act as an intermediary (if the sending car is in a synchronous area).

All node computers have of course a cable link to the Central Computer. As for finding the recipient if it is another beamcar, the node could send a broadcast, asking for this car. Since all beamcars are accounted for att every single moment, the node that recognize the recipient as being a car within its own jurisdiction would send an acknowledgement to the sending node, and then forward the message to the recipient.

As for communication between beamcars and the Central Computer; for geographical reasons (the cells being rather small), the best solution to this would be:

In areas with synchronous traffic; ask the node to relay the message.
In areas with asynchronous traffic; use antennas that cover all such beams. These antennas would be directly controlled by the Central Computer.

This solution would also enable communication both ways with cars on beams with asynchronous traffic.

3. What is Communicated?

his chapter is meant to list all types of messages that need to be communicated. In analogy with established rules for synchronous communication between computers, there would be 2 general types of messages:

Control messages, that would have direct bearing on the control of traffic.
Data messages, that would convey information to and from:
1. passengers in beamcars,
2. information monitors and
3. various administrative information to and from the Central Computer.

The first group (the Control messages) would always have first priority. This list of messages is not finalized yet, but a lot of information regarding this can be found on other (technical) webpages on this site.

Copyright © 2004, SwedeTrack System.	Last Updated: 2007-01-17	Webmaster	This site is maintained by Johnson Consulting