Machine learning based clustering/routing algorithms to reduce energy consumption

Dear Jannat Hegazy,

to set up a WSN that contains many Low-power nodes (sensors) and is expected to work well for a long time, the following Design Principles (DPs) can be used:

· DP1: Energy Efficient Design with an energy-aware static routing

This DP consists of designing WSN so that Low-Power Nodes (LPNs) do not serve as forwarders (routers) of transit / foreign data. It should be noted that the LPNs are not installed on the "important and heavily loaded" static routes in the WSN.

· DP2: Reduction of power consumption in LPNs thanks to load-controlled power supply

To maximize battery life, LPNs spend most of their time in a "sleep" mode and then consume little power. Often they are "awakened" periodically according to a fixed communication plan to perform their functions. To increase battery life, the internal timer of the microcontroller (MCU) in LPNs may be load-controlled so that LPNs consume less power. This means a load-dependent power supply of LPNs. For more information about a load-dependent power supply, see my answer to Q1.

· DP3: Use of the IoT Routing Protocol RPL with Energy-aware Metrics

The RPL is intended for use in WSNs with LPNs. For this reason, RPL can also use so-called energy-aware metrics, which take into account the energy state of bateries in LPNs. Such metrics cause the optimal routes in WSNs to not pass through the low power nodes. The RPL thus contributes to extending the lifetime of a WSN with LPNs. This goal can be achieved with the help of RPL and with the objective function MRHOF. The RPL option "DAG Metric Container" is very important for this; see RFC 6550 and RFC 6551 (sections 2.1 (Fig. 1) and 3.2)). In this container you can specify the details of routing metric / constraint types. For you, the Node-based metrics - RE (Remaining Energy) and MLT (Maximum Life Time) - are important.

· DP4: Self-Healing WSN with redundancy nodes

A WSN as at IoT part should be able to repair or even avoid some node failures due to energy shortage, as well as various faulty situations. The RPL can help with energy-aware metrics. There are two principles of self-healing: reactive self-healing and proactive self-healing.

In reactive self-healing, an attempt is made to identify and locate the faulty system components. Then an attempt is made to compensate for the failed functions as far as possible. To enable reactive self-healing in a WSN, some redundancy (redundant nodes, redundant wireless connections) must be "built-in" in the WSN. Reactive self-healing in a WSN means that fault tolerance or even fault tolerance can be activated.

Proactive self-healing attempts to monitor and manage a WSN to prevent possible overloads or failures of system components.

Self-healing in a WSN can be supported using the RPL routing protocols. See my presentation of RPL - and there are the Figures 8361, 8363 and 8366.

· DP 5: WSN with Fault Tolerance

In a WSN with RPL, several DODAGs can overlap. Some important nodes may be part of several DODAGs, so these nodes can be accessed via the Internet via several Border Routers. This would lead to Fault Tolerance. The Figures 008361 and 008363 in my work RPL can help you. See my answer to the question Q6.

My answers to questions Q1, Q2, ..., Q6 can help you.

Much success and best regards

Anatol Badach

RPL – IPv6 Routing Protocol for LLNs

https://www.researchgate.net/publication/320244798_RPL_-_IPv6_Routing_Protocol_for_LLNs

Q1: Static and dynamic power estimations?

https://www.researchgate.net/post/static_and_dynamic_power_estimations

Q2: How can I calculate an using energy of WSN node?

https://www.researchgate.net/post/How_can_I_calculate_an_using_energy_of_WSN_node

Q3: Sleep phase between every two frames?

https://www.researchgate.net/post/sleep_phase_between_every_two_frames

Q4: What features or Softwares can be build to make IT greener in our Environment

https://www.researchgate.net/post/What_features_or_Softwares_can_be_build_to_make_IT_greener_in_our_Environment

Q5: What methods can I use to solve the self-healing problem on the Internet of Things (IoT)?

https://www.researchgate.net/post/What_methods_can_I_use_to_solve_the_self-healing_problem_on_the_Internet_of_Things_IoT

Q6: Fault Tolerance in Smart City?

https://www.researchgate.net/post/Fault_Tolerance_in_Smart_City

Q7: How can you maximize lifetime in a heterogeneous omnidirectional wireless sensor network while preserving target coverage?

https://www.researchgate.net/post/How_can_you_maximize_lifetime_in_a_heterogeneous_omnidirectional_wireless_sensor_network_while_preserving_target_coverage



Dear Jannat Hegazy,

as addition to my answer, an interesting literature source:

Qianqian Li, Sarada Prasad Gochhayat, Mauro Conti and FangAi Liu: "EnergIoT: A solution to improve network lifetime of IoT devices"

https://www.researchgate.net/publication/320400757_EnergIoT_A_solution_to_improve_network_lifetime_of_IoT_devices

Best regards

Anatol Badach

thanks dr Shubham Gupta, if you have any paper about your answer please send it to me.

thanks a lot prof. Anatol Badach, i will follow the steps you identify.