| ??? 06/28/05 10:10 Read: times |
#96081 - another solution Responding to: ???'s previous message |
Having read this thread I don't think that the possibility of a repeater network has been taken seriously. If every light has another light in a N meter radius, (where N is an acceptable distance for a cheap data transmission system) you could just implement a very simple routing algorithm where signal has to hop from one light to the nearest one until destination is reached.
As you probably only need one master, you could take a tree layout network with a practical adress space. For example master has adress (0,0). the first num is branch index, the second is branch host in branch index. The data frame could contain : - original sender - last repeater - destination - data - crc With a routing algorithm like : signed char compare_addr(aX,aY,bX,bY) { if(aX < bX) return -1; if(ax > bX ) return 1; if(aY < bY ) return -1; if(aY > bY) return 1 return 0; } my_addr =( X,Y) // defined for each controller unsigned char route(packet P) { signed char comp_dest = compare_addr(my_addr,packet->dest); signed char comp_hop = compare_addr (my_addr,packet->last_repeater); signed char comp = comp_dest+comp_host; if ( comp == -2 || comp == 2) { reemit(P); //change the last_repeater to self return 0; } if ( ! comp_dest ) return 1; } typedef struct addr__ { ADDR_TYPE X; ADDR TYPE Y; } addr; typedef struct packet__ { addr dest, last_repater, from; unsigned char length; unsigned char crc; } frame; typedef frame * packet; //// frame current; packet P; main() { P = & current; while(1) { wait_packet(P); //Packet is received if (!check_integrity(P)) continue; if (route(P)) treat(P); } } This is just a draft : feel free to comment. Best regards Benjamin Ps : adress attribution should ensure that adresses are unique, and that a lower(mother) tree branch always have a smaller branch ids on it. This system is far from beeing 'armor piecing bullet proof'... |
| Topic | Author | Date |
| Power Line Channel | 01/01/70 00:00 | |
| broadcast in noise | 01/01/70 00:00 | |
| Broadcast Vs Addressing | 01/01/70 00:00 | |
| erk | 01/01/70 00:00 | |
| Just some things | 01/01/70 00:00 | |
| Got it | 01/01/70 00:00 | |
| Horses for Courses | 01/01/70 00:00 | |
| just curious, why not a photocell at eac | 01/01/70 00:00 | |
| Radio | 01/01/70 00:00 | |
| Zigbee? | 01/01/70 00:00 | |
| great minds think alike | 01/01/70 00:00 | |
| you have my deepest sympathy | 01/01/70 00:00 | |
| Baby steps | 01/01/70 00:00 | |
| and.. | 01/01/70 00:00 | |
| Connection | 01/01/70 00:00 | |
| Hmmm | 01/01/70 00:00 | |
| MSN | 01/01/70 00:00 | |
| Nop | 01/01/70 00:00 | |
| K k | 01/01/70 00:00 | |
| kilometer | 01/01/70 00:00 | |
| what would he mean? | 01/01/70 00:00 | |
| Binary | 01/01/70 00:00 | |
| Is no guess, is only know. | 01/01/70 00:00 | |
| Mexico | 01/01/70 00:00 | |
| X10 -like | 01/01/70 00:00 | |
| Actualization | 01/01/70 00:00 | |
| K k | 01/01/70 00:00 | |
| kHz | 01/01/70 00:00 | |
| Andy Neil alias Mr. Perfectionist !! | 01/01/70 00:00 | |
| X10 like | 01/01/70 00:00 | |
| Pulse position? | 01/01/70 00:00 | |
| Existing systems | 01/01/70 00:00 | |
| Pulse position | 01/01/70 00:00 | |
| Yes | 01/01/70 00:00 | |
| AT89S8252 : lcd | 01/01/70 00:00 | |
| Huh? | 01/01/70 00:00 | |
| Off-Topic | 01/01/70 00:00 | |
| impedance matching | 01/01/70 00:00 | |
| I see dead computers | 01/01/70 00:00 | |
file transfer using power lines | 01/01/70 00:00 | |
| X2 Rated | 01/01/70 00:00 | |
| RF switching | 01/01/70 00:00 | |
| Try it ! | 01/01/70 00:00 | |
| Ya, sure | 01/01/70 00:00 | |
| $200? | 01/01/70 00:00 | |
| Not necessarily Xbee | 01/01/70 00:00 | |
| another solution | 01/01/70 00:00 | |
| well... | 01/01/70 00:00 | |
| maybe... | 01/01/70 00:00 | |
| mains born network | 01/01/70 00:00 | |
| 8000 pieces | 01/01/70 00:00 | |
| re: | 01/01/70 00:00 | |
| not really | 01/01/70 00:00 | |
| re:cost | 01/01/70 00:00 | |
| Cost? | 01/01/70 00:00 | |
| Cost | 01/01/70 00:00 | |
| SNR | 01/01/70 00:00 | |
| BER | 01/01/70 00:00 | |
| checksum to the hilt | 01/01/70 00:00 | |
| CRC, not checksum | 01/01/70 00:00 | |
| Duplex | 01/01/70 00:00 | |
| if he can afford an occasional miss | 01/01/70 00:00 | |
| checksum | 01/01/70 00:00 | |
| checksum and short | 01/01/70 00:00 | |
| Test set | 01/01/70 00:00 | |
| 1143.12345 Hz | 01/01/70 00:00 | |
| maybe once in a year | 01/01/70 00:00 | |
| Idea CRC | 01/01/70 00:00 | |
| SNR ? | 01/01/70 00:00 | |
| Frequency | 01/01/70 00:00 | |
| Lower frequencies | 01/01/70 00:00 | |
| Noise Floor | 01/01/70 00:00 | |
| 10% | 01/01/70 00:00 | |
| Data Rate | 01/01/70 00:00 | |
| re:10% error rate | 01/01/70 00:00 |



