System and method for rejoining sleeping subnets in a wireless mesh network

1. A method of joining nodes from a first and a second subnet into a network, wherein each subnet includes two or more nodes and wherein each subnet is in a different, non-overlapping wake-sleep timeslot, the method comprising:

establishing a probability that a node within the first subnet will enter scout mode during which the node will wake between regular synchronized wake periods of the subnet, wherein the probability is less than 100%;

if the node enters scout mode, broadcasting a message from the node in scout mode to other nodes inviting nodes in other subnets to join with the first subnet, wherein the message includes the node's network identifier; and

detecting, within a detecting node in the second subnet, the message broadcast by the node in scout mode; and

determining whether nodes in the second subnet have the same network identifier as the network identifier in the message broadcast by the node in scout mode; and

if the nodes in the second subnet have the same network identifier as the network identifier in the message broadcast by the node in scout mode, combining the nodes in the first and second subnets into a combined network;

wherein each node generates a first and second random number during its sleep interval, wherein, if the first random number is less than a threshold value, the node enters scout mode during that sleep interval, wherein the second random number defines when the node enters scout mode during that sleep interval; and

wherein combining includes:

determining which subnet has fewer nodes;

if the first subnet has fewer nodes than the second subnet, instructing all nodes in the first subnet to synchronize to the second subnet; and

if the first subnet has more nodes than the second subnet, instructing all nodes in the second subnet to synchronize to the first subnet.

2. The method of claim 1 , wherein combining includes synchronizing the subnet with fewer nodes to time parameters of the subnet with more nodes.

3. The method of claim 1 , wherein each node is designed to enter scout mode at a random time during the sleep interval.

4. The method of claim 1 , wherein each node generates a random number during its sleep interval and, if the random number is less than a threshold value, the node enters scout mode during that sleep interval.

5. The method of claim 1 , wherein each node generates a first and second random number during its sleep interval, wherein, if the first random number is less than a threshold value, the node enters scout mode during that sleep interval, wherein the second random number defines when the node enters scout mode during that sleep interval.

6. The method of claim 1 , wherein combining includes:

if the first subnet has fewer nodes than the second subnet, sending a message from the node in scout mode to the other nodes in the first subnet, wherein the message instructs all nodes in the first subnet to synchronize to the second subnet; and

if the first subnet does not have fewer nodes than the second subnet, sending a message from the detecting node in the second subnet to the other nodes in the second subnet, wherein the message instructs all nodes in the second subnet to synchronize to the first subnet.

7. The method of claim 1 , wherein combining includes:

if the first subnet does not have more nodes than the second subnet, sending a message from the node in scout mode to the other nodes in the first subnet, wherein the message instructs all nodes in the first subnet to synchronize to the second subnet; and

if the first subnet has more nodes than the second subnet, sending a message from the detecting node in the second subnet to the other nodes in the second subnet, wherein the message instructs all nodes in the second subnet to synchronize to the first subnet.

8. The method of claim 1 , wherein combining includes:

if the first subnet has fewer nodes than the second subnet, sending a message from the node in scout mode to the other nodes in the first subnet, wherein the message instructs all nodes in the first subnet to shift their sleep-wake time interval, as well as other sleeping mode parameters, to match that of the detecting node in the second subnet; and

if the first subnet has more nodes than the second subnet, sending a message from the detecting node in the second subnet to the other nodes in the second subnet, wherein the message instructs all nodes in the second subnet to shift their sleep-wake time interval, as well as other sleeping mode parameters, to match that of the node in scout mode in the first subnet.

9. The method of claim 8 , wherein each node generates a first and second random number during its sleep interval, wherein, if the first random number is less than a threshold value, the node enters scout mode during that sleep interval, wherein the second random number defines when the node enters scout mode during that sleep interval.

10. The method of claim 1 , wherein combining includes:

if the first subnet has fewer nodes than the second subnet, sending a message from the node in scout mode to the other nodes in the first subnet, wherein the message instructs all nodes in the first subnet to shift their sleep-wake time interval, as well as other sleeping mode parameters, to match that of the detecting node in the second subnet; and

if the first subnet does not have fewer nodes than the second subnet, sending a message from the detecting node in the second subnet to the other nodes in the second subnet, wherein the message instructs all nodes in the second subnet to shift their sleep-wake time interval, as well as other sleeping mode parameters, to match that of the node in scout mode in the first subnet.

11. The method of claim 1 , wherein combining includes:

if the first subnet does not have more nodes than the second subnet, sending a message from the node in scout mode to the other nodes in the first subnet, wherein the message instructs all nodes in the first subnet to shift their sleep-wake time interval, as well as other sleeping mode parameters, to match that of the detecting node in the second subnet; and

if the first subnet has more nodes than the second subnet, sending a message from the detecting node in the second subnet to the other nodes in the second subnet, wherein the message instructs all nodes in the second subnet to shift their sleep-wake time interval, as well as other sleeping mode parameters, to match that of the node in scout mode in the first subnet.

12. An article comprising a computer readable medium having instructions thereon, wherein the instructions, when executed by a machine, create a system for executing the method of claim 1 .

13. A wireless mesh network, comprising:

a first and a second subnet, wherein each subnet includes two or more nodes and wherein each subnet is in a different, non-overlapping wake-sleep timeslot;

wherein each node communicates wirelessly to one or more other nodes in the network;

wherein each node includes program code for executing a wake/sleep cycle having a wake interval followed by a sleep interval;

wherein each node includes program code for creating a probability that the node will wake during the sleep interval to execute a scout function in which the node will wake between regular synchronized wake periods of a subnet, wherein the probability is less than 100%;

wherein each node includes program code for detecting a node in scout mode and for responding to a message from the node in scout mode if it is in a different subnet and has a common network identifier; and

wherein each node includes program code for combining its subnet with a different subnet when they share a common network identifier;

wherein each node includes program code for determining whether nodes in a different subnet have the same network identifier as the network identifier in said message from the node in scout mode; and

if a node in the first subnet has the same network identifier as the network identifier in the message broadcast by the node in scout mode, combining the nodes in the respective subnets into a combined network;

wherein each node generates a first and second random number during its sleep interval, wherein, if the first random number is less than a threshold value, the node enters scout mode during that sleep interval, wherein the second random number defines when the node enters scout mode during that sleep interval; and

wherein combining includes:

determining which subnet has fewer nodes;

if the first subnet has fewer nodes than the second subnet, instructing all nodes in the first subnet to synchronize to the second subnet; and

if the first subnet has more nodes than the second subnet, instructing all nodes in the second subnet to synchronize to the first subnet.

14. The network of claim 13 , wherein the program code for creating a probability that the node will wake during the sleep interval is designed to increase the probability that most nodes will sleep the entire sleep interval.

15. The network of claim 13 , wherein the program code for creating a probability that the node will wake during the sleep interval includes program code for selecting a random time to awake during the sleep interval.