Sensing ignition by voltage monitoring

1. A method of sensing the ignition state of a vehicle engine, comprising the steps of:

a) sampling a voltage level of a vehicle electrical system as powered by a vehicle battery without relying on other physical phenomena related to the engine, to obtain a current voltage sample;

b) determining if said current voltage sample has a sufficient drop from a preceding voltage sample;

c) determining if said current voltage sample is reliable by determining if a trend of voltage samples preceding said current voltage sample is consistent with said current voltage sample; and

d) determining if said current voltage sample is below a specified threshold from which an accurate inference is that the vehicle engine ignition is off.

2. A method of sensing the ignition state of a vehicle engine, comprising the steps of:

a) sampling a voltage level of a vehicle electrical system as powered by a vehicle battery without relying on other physical phenomena related to the engine, to obtain a current voltage sample;

b) determining if said current voltage sample has a sufficient drop from a preceding voltage sample;

c) determining if said current voltage sample is reliable;

d) determining if said current voltage sample is below a specified threshold from which an accurate inference is that the vehicle engine ignition is off; and

e) calculating a Running Average Voltage;

wherein step b) determines if said current voltage sample has a sufficient drop from said preceding voltage sample by measuring whether the drop exceeds a prescribed Drop Delta below said Running Average Voltage.

3. The method of claim 2 , wherein step b) determines if said current voltage sample has a sufficient drop by further measuring whether there is a prescribed number of consecutive voltage samples each having a drop exceeding said Drop Delta.

4. The method of claim 3 , further comprising the step of determining whether the Running Average Voltage is below a prescribed voltage, and if it is, recognizing an Ignition Off state.

5. The method of claim 2 , wherein if it is determined that said current voltage sample is not reliable, further comprising the step of substituting said Running Average Voltage with a value of a preceding Running Average Voltage.

6. The method of claim 2 , wherein said Drop Delta is a function of local conditions being a weighted sum of vehicle battery age, engine-on time, ambient temperature, current voltage sample, engine temperature, and current Drop Delta, where respective synaptic weights are learned via a standard neural network learning algorithm.

7. The method of claim 2 , wherein said Drop Delta is a fuzzy logic function based on local conditions of vehicle battery age, engine-on time, ambient temperature, current voltage sample, engine temperature, and current Drop Delta.

8. A method of sensing the ignition state of a vehicle engine, comprising the steps of:

a) sampling a voltage level of a vehicle electrical system as powered by a vehicle battery without relying on other physical phenomena related to the engine, to obtain a current voltage sample;

b) determining if said current voltage sample has a sufficient drop from a preceding voltage sample;

c) determining if said current voltage sample is reliable;

d) determining if said current voltage sample is below a specified threshold from which an accurate inference is that the vehicle engine ignition is off; and

e) assigning default values to a plurality of parameters comprising Sample Interval, Sample Count, Transition Count On, Transition Count Off, On Threshold Voltage, Drop Delta and Off Threshold Voltage, and then tuning the parameters using guidelines and heuristics.

9. The method of claim 8 , wherein the parameter of Transition Count Off is assigned a value that differs from a value of Transition Count On.

10. The method of claim 8 , wherein the parameter of Transition Count Off is changed in response to a recent trend of voltage samples.