Augmented mesh delivery system

1. A method comprising:

receiving, by a controller at a destination node, signal samples and associated sampling time indications, the signal samples and the associated sampling time indications being received from a source node via a mesh network, the signal samples being delivered with sampling time indications generated at the source node to form a series of signals corresponding to one or more characteristics related to electricity sensed by a sensor;

applying, by the controller, a compacting algorithm to the signal samples and the associated sampling time indications to generate a paired vector comprising compacted signal samples and an associated compacted sampling time indication;

transforming the paired vector representing the compacted signal samples in a time domain into a discrete Fourier spectrum of the signal samples in a frequency domain using a modified Approximate Fourier Transform such that a discrete Fourier spectrum of the compacted signal samples is uniformly sampled in the frequency domain;

applying a filter in the frequency domain to the discrete Fourier spectrum of the signal samples to generate a filtered discrete spectral representation; and

applying an inverse discrete Fourier transform to the filtered discrete spectral representation to generate a recovered signal in the time domain that is uniformly sampled,

wherein a weighting of sample values in the modified Approximate Fourier Transform is determined at least partially based on information indicative of a statistical behavior of a corresponding realized sample process.

2. The method of claim 1 , further comprising providing an adjustment signal to adjust the one or more characteristics of the electricity based on the recovered signal.

3. The method of claim 1 , wherein applying the filter in the frequency domain to the discrete Fourier spectrum of the signal samples to generate the filtered discrete spectral representation comprises:

applying a bandpass filter in the frequency domain to the discrete Fourier spectrum of the signal samples.

4. The method of claim 1 , wherein applying the filter in the frequency domain to the discrete Fourier spectrum of the signal samples to generate the filtered discrete spectral representation comprises:

applying a low pass filter in the frequency domain to the discrete Fourier spectrum of the signal samples.

5. The method of claim 4 , wherein attenuation characteristics of the low pass filter are set related to the rate at which the one or more characteristics of the electricity can be adjusted.

6. The method of claim 1 , wherein applying the compacting algorithm to the signal samples and the associated sampling time indications to generate the paired vector comprising the compacted signal samples and the associated compacted time indication comprises:

determining void indications in the signal samples; and

indicating a sampling time for each of the void indications.

7. The method of claim 1 , wherein the one or more characteristics related to the electricity sensed by the sensor comprise at least one from the list consisting of: voltage, current, real power, power flow direction, reactive power, temperature or humidity.

8. The method of claim 1 , wherein the mesh network includes a plurality of emitters, wherein a first emitter of the plurality of emitters is configured to connect to a second one of the plurality of emitters to form a path to deliver signals between the source node and the destination node, and wherein at least the first emitter is configured to automatically connect to a third emitter of the plurality of emitters to form a new path between the source node and the destination node when interference is present between the first emitter and the second emitter.

9. The method of claim 1 , wherein applying the inverse discrete Fourier transform to the filtered discrete spectral representation to generate the recovered signal in the time domain that is uniformly sampled comprises:

extrapolating future values of the recovered signal in the time domain by setting a time basis in the inverse discrete Fourier transform to be greater than a total time span of the compacted signal samples; and

providing an adjustment signal to adjust the one or more characteristics of the electricity based on the recovered signal,

wherein the adjustment signal is provided based on the future values to compensate for a group delay of the filter.

10. The method of claim 1 , wherein the signal samples are non-uniformly delivered to the destination node from the source node via the mesh network.

11. The method of claim 1 , further comprising: delivering the signal samples with an associated sampling time indication or with delivery time indications.

12. The method of claim 1 , wherein the associated compacted sampling time indication is derived from an actual sample interval for each sample, wherein the actual sample interval is a time difference between that of a present sample and a previous sample.

13. A method comprising:

receiving, by a controller at a destination node, signal samples and associated sampling time indications, the signal samples and the associated sampling time indications being received from a source node via a mesh network, the signal samples being delivered with sampling time indications generated at the source node to form a series of signals corresponding to one or more characteristics related to electricity sensed by a sensor;

applying, by the controller, a compacting algorithm to the signal samples and the associated sampling time indications to generate a paired vector comprising

compacted signal samples and an associated compacted sampling time indication;

transforming the paired vector representing the compacted signal samples in a time domain into a discrete Fourier spectrum of the signal samples in a frequency domain using an Approximate Fourier Transform such that the discrete Fourier spectrum of the compacted signal samples is uniformly sampled in the frequency domain; and

based on the discrete Fourier spectrum, applying an iterative Adaptive weight conjugate gradient Toeplitz (ACGT) procedure to generate a recovered signal in the time domain that is uniformly sampled.

14. The method of claim 13 , further comprising, prior to applying the ACGT procedure, specifying an initial convergence criteria for the ACGT procedure.

15. The method of claim 14 , further comprising estimating a uniformly sampled recovered signal using the ACGT procedure such that a recovered sampling interval is a desired nominal interval for the signal samples.

16. The method of claim 15 , wherein the recovered signal includes a specified number of forecasted samples.

17. The method of claim 13 , further comprising determining convergence information indicative of a number of iterations of the ACGT procedure required to achieve an initial convergence criteria.

18. The method of claim 17 , further comprising determining a revised convergence criteria based on the convergence information and information indicative of sample availability of each input signal realization.

19. The method of claim 17 , wherein the convergence information is determined based on the initial convergence criteria other than an absolute error criterion.

20. The method of claim 19 , wherein the initial convergence criteria is based at least in part on a rate of decrease of absolute error.

21. The method of claim 13 , further comprising constraining an allowable number of iterations of the ACGT procedure.

22. The method of claim 13 , further comprising providing an adjustment signal to adjust the one or more characteristics of the electricity based on the recovered signal.

23. The method of claim 13 , further comprising applying a bandpass filter in the frequency domain to the discrete Fourier spectrum of the signal samples to generate a filtered discrete spectral representation.

24. The method of claim 13 , further comprising applying a low pass filter in the frequency domain to the discrete Fourier spectrum of the signal samples to generate a filtered discrete spectral representation of the compacted signal samples.

25. The method of claim 24 , wherein attenuation characteristics of the low pass filter are set related to a rate at which the one or more characteristics of the electricity are adjusted.

26. The method of claim 13 , further comprising:

applying a filter in the frequency domain to the discrete Fourier spectrum of the signal samples to generate a filtered discrete spectral representation of the compacted signal samples; and

applying an inverse discrete Fourier transform to the filtered discrete spectral representation to generate the recovered signal in the time domain that is uniformly sampled.

27. The method of claim 13 , wherein applying with the controller the compacting algorithm to the signal samples and the associated sampling time indications to generate the paired vector comprising the compacted signal samples and the associated compacted time indication comprises:

determining void indications in the received signal samples; and

indicating a sampling time for each of the void indications.

28. The method of claim 13 , wherein the one or more characteristics related to the electricity comprise at least one from the list consisting of: voltage, current, real power, power flow direction, reactive power, temperature or humidity.

29. The method of claim 13 , wherein the mesh network includes a plurality of emitters, wherein a first emitter of the plurality of emitters is configured to connect to a second one of the plurality of emitters to form a path to deliver signals between the source node and the destination node, and wherein at least the first emitter is configured to automatically connect to a third emitter of the plurality of emitters to form a new path between the source node and the destination node when interference is present between the first emitter and the second emitter.

30. The method of claim 13 , further comprising:

applying a filter in the frequency domain to the discrete Fourier spectrum of the signal samples to generate a filtered discrete spectral representation of the compacted signal samples;

applying an inverse discrete Fourier transform to the filtered discrete spectral representation to generate the recovered signal in the time domain that is uniformly sampled;

extrapolating future values of the recovered signal in the time domain by setting a time basis in the inverse discrete Fourier transform to be greater than a total time span of the compacted signal samples, and

wherein an adjustment signal is provided based on the future values to compensate for a group delay of the filter.

31. The method of claim 13 , wherein the signal samples are non-uniformly delivered to the destination node from the source node via the mesh network.

32. The method of claim 13 , further comprising:

delivering the signal samples with an associated sampling time indication or with delivery time indications.

33. The method of claim 13 , wherein the associated compacted sampling time indication is derived from an actual sample interval for each sample, wherein the actual sample interval is a time difference between that of a present sample and a previous sample.

34. An apparatus comprising:

a controller having one or more processors that;

receives at a destination node signal samples and associated sampling time indications, the signal samples and the associated sampling time indications being received from a source node via a mesh network, the signal samples being delivered with sampling time indications generated at the source node to form a series of signals corresponding to one or more characteristics related to electricity sensed by a sensor;

applies a compacting algorithm to the signal samples and the associated sampling time indications to generate a paired vector comprising compacted signal samples and an associated compacted sampling time indication;

transforms the paired vector representing the compacted sampled signal in a time domain into a discrete Fourier spectrum of the signal samples in a frequency domain using a modified Approximate Fourier Transform such that a discrete Fourier spectrum of the compacted signal samples is uniformly sampled in the frequency domain;

applies a filter in the frequency domain to the discrete Fourier spectrum of the signal samples to generate a filtered discrete spectral representation; and

applies an inverse discrete Fourier transform to the filtered discrete spectral representation to generate a recovered signal in the time domain that is uniformly sampled,

wherein weighting of sample values in the modified Approximate Fourier Transform is determined at least partially based on information indicative of a statistical behavior of a corresponding realized sample process.

35. A non-transitory computer readable storage medium storing computer executable instructions that when executed by one or more processors, cause the one or more processors to perform operations comprising:

receiving at a destination node signal samples and associated sampling time indications, the signal samples and the associated sampling time indications being received from a source node via a mesh network, the signal samples being delivered with sampling time indications generated at the source node to form a series of signals corresponding to one or more characteristics related to electricity sensed by a sensor;

applying a compacting algorithm to the signal samples and the associated sampling time indications to generate a paired vector comprising compacted signal samples and an associated compacted sampling time indication;

transforming the paired vector representing the compacted signal samples in a time domain into a discrete Fourier spectrum of the signal samples in a frequency domain using a modified Approximate Fourier Transform such that a discrete Fourier spectrum of the compacted signal samples is uniformly sampled in the frequency domain;

applying a filter in the frequency domain to the discrete Fourier spectrum of the signal samples to generate a filtered discrete spectral representation; and

applying an inverse discrete Fourier transform of the filtered discrete spectral representation to generate a recovered signal in the time domain that is uniformly sampled,

wherein the weighting of sample values in the modified Approximate Fourier Transform is determined at least partially based on information indicative of a statistical behavior of a corresponding realized sample process.