SRAMI: Secure and Reliable Advanced Metering Infrastructure Protocol for Smart Grid

: The emergence of Advanced Metering Infrastructure (AMI) into the Smart Grid applications receiving significant attention by researchers of the Internet of Things (IoT) assisted smart city projects. Communication networks (WiFi/WLAN) is one of the key building blocks of AMI for information exchange. The wireless communication networks are vulnerable to various security threats that lead to problems in designing the AMI system in smart city projects. The main requirements of cyber-physical systems like AMI include confidentiality, integrity, availability, and privacy. To satisfy the requirements of cyber-physical systems, we focused on the wireless communication protocol that addresses data transmission reliability and data security with minimum power consumption and overhead. The Secure and Reliable AMI (SRAMI) protocol using the unique trust-based mechanism for reliability and lightweight cryptography mechanism for the security proposed for Wireless Sensor Network (WSN)-assisted AMI. The trust-based algorithm introduces to establish a reliable data transmission technique among two communicating entities of AMI. In this regard, next-hop selected according to trust-evaluation using parameters energy consumption, geographical distance, and bandwidth availability parameters. Furthermore, the lightweight Elliptic Curve Cryptography (ECC)-based hybrid technique introduces to satisfy the data integrity and privacy requirements in the AMI network. The performance of SRAMI evaluated using throughput, Packet Delivery Ratio (PDR), average energy consumption, delay, and communication overhead compared to state-of-art methods. The throughput and PDR improved by 5 % and 3.14 % compared to existing methods. The energy consumption, delay, and overhead reduced by 7.36 %, 12.16 %, and 17.66 % compared to existing techniques.


Introduction
AMI (Advanced Metering Infrastructure) is the common terminology to explain the entire infrastructure of Smart Meter to two way-communication interfaces to manage center devices and all the pertinence that facilitate the collection and transfer of power utilization data in imminent real-time [Fadlullah et al. 2018].AMI delivers two-way communications with consumers desirable and is the resolution of the smart grid.The purposes of AMI can be network difficulty credentials, smart meter reading for error-free data, partial load reduction, energy examination, and load profiling in place of load molting [Huh et al. 2016].AMI consists of different hardware and software segments, all of which represent a function in regulating power consumption and communicating data about power, gas, and water usage to service organizations and consumers.The technological elements of AMI incorporate:  Smart Meters: These AMI elements having the capacity to gather data about power, water, and gas usage at periodic intervals and communicating the data over established communication networks to the utility, and getting data like pricing signals from the utility and sending it to the customer.The smart meters can represent as sensor devices. Communication Network: The data transmission among smart meters and the utilities performed using two-way communication technique.The commonly used communication methods are Power Line Communications (PLC), Broadband over Power Line (BPL), Fixed Radio Frequency (FRD), Fiber Optic Communication (FOC), or public networks (e.g., landline, cellular, paging). Meter Data Acquisition System: As the name indicates, this component deals with data acquisition from the meters through the communication network and transmits it to the Meter Data Management System (MDMS) using communication networks.


Meter Data Management System (MDMS): This is the host component that collects, stores, and examines the metering data.The AMI benefits several ways in the Internet of Things (IoT) enabled smart city applications such as operational benefits, financial benefits, and customer benefits [Muhanji et al. 2019].For operational benefits, AMI helps the whole grid by advancing the efficiency of meter reading, power theft detection, and reply to power interruptions while reducing the demand for an on-site meter reading.For financial benefits, AMI produces economic profits for utility, water, and gas organizations by decreasing material and support prices, facilitating the faster rehabilitation of electrical assistance through blackouts, and streamlining the billing method.For customer benefits, AMI avails electric consumers by identifying meter malfunctions early, providing quick assistance recovery, and enhancing the efficiency and versatility of billing.It supports time-based rate choices that can assist consumers in managing their power consumption and save money [Al-Turjman et al. 2019].However, AMI deployment suffers from various challenges such as security against cyber threats, reliability, energy theft, energy costs, integration, etc. AMI is a kind of cyber-physical system in IoT applications and suffers from different security threats.Hence, protecting the AMI data communications between smart meters and AMI hosts from various cyber threats is a research problem.From the perspective of the smart city project, AMI becomes a crucial part of the IoT-assisted systems.As discussed earlier, the smart meters periodically sense the meter data and transmit it to the AMI host via insure wireless communications.Thus, to protect such insecure communications in AMI from cyber threats, two mechanisms need to be done at the network layer like reliable route discovery and secure data transmission.These two mechanisms bring security benefits using AMI to the smart grid systems and consumers.Smart Grid saves millions of dollars in the electricity sector by providing smart and secure wireless communication infrastructure to AMI.The communication from smart meters (acts sensor nodes) to the AMI hosts (acts base station node) represents the Wireless Sensor Network (WSN)-assisted IoT system for AMI [Barsana et al. 2021].The wireless networks like Mobile Ad hoc Network (MANET) and WSN are self-configured enabling technologies and enhancing the performance of Quality of Services (QoS) for these enabling technologies is a challenging task [Mahajan et al 2018].WSN and MANET suffer from performance degradation due to diverse threats, corresponding forging, Replay, Colluding, Denial of Service (DoS), and Malicious attacks [Singh et al. 2018].This research focuses on designing routing protocol to address the challenges of reliability and data security in presence of cyber threats via the calculation of parameters PDR, output quantity (throughput), communication delay, communication overhead, and energy consumption.Several routing protocols introduced categories-wise on-demand, table-driven, hybrid, etc., but such protocols failed to protect network communications from security threats [Lekshmi et al. 2020].Proactive Ad-hoc On-demand Distance Vector (AODV) and reactive Destination-Sequenced Distance-Vector Routing (DSDV) are non-secure routing protocols with no provisions to tackle malicious attacks.The AODV and DSDV investigators were annoyed to provide routing security for MANET and WSN considering routing calculation parameters, and their effect on security was despicable for wireless communication [NC et al. 2018].Subsequently the failures of the non-secure protocols, researchers are tried cryptography-based protocols like Authenticated Anonymous Secure Routing (AASR) [Liu et al. 2014] and trustbased Trusted and Energy-efficient Routing Protocol (TERP) [Shen et al. 2017], but such approaches failed to satisfy all the security requirements.To gain the security benefits in AMI, we proposed the Secure and Reliable AMI (SRAMI) routing protocol using the lightweight ECC-based cryptography and trustbased approach for reliable route discovery.The goal of the SRAMI protocol is to satisfy all the security requirements of AMI communications that include confidentiality, integrity, availability, and privacy.Figure 1 demonstrates the mechanism of SRAMI protocol for smart grid AMI application.The trust-model introduces reliable next-hop selection and ECC-based lightweight cryptography for secure data transmission.The performance of SRAMI is satisfied by the five QoS parameters PDR, throughput, communication delay, communication overhead, and average energy consumption of smart meters.Section 2 presents a brief study on the state-ofart methods.Section 3 presents the algorithms of the SRAMI protocol.Section 4 presents the simulation results and discussions.Section 5 presents the conclusion and future recommendations.

Related work
Since the past decade, several attempts introduced for the security of wireless networks such as WSNs, MANET, and IoT-enabled networks to protect against the various threats.The methods are broadly categorized into trust-based and cryptography-based for attack detection and mitigation during the wireless data transmissions.This section reviewed some recent trust-based and cryptography-based mechanisms for wireless communications.After that, research motivations and contributions have been disclosed.

A. Wireless Security Methods
Conviction management is a big challenge in a different wireless communication network, even though researchers are attempting to give security issues are not finished [Kraounakis et al. 2015].All the aspects of the SRAMI algorithm are demonstrated.Researchers, who had worked on security issues for different types of networks, still face difficulties.The performance of the routing method is based on a reliable path discovery.And finding the trusted path is based on the trust evaluation algorithm.The Internet of Things (IoT) supports smart systems and its wireless communication based on WSN.Trust monitoring scheme reduces the communication cost, minimizes overhead, and increases the network life.Correspondingly, the logic of the OLSR protocol was used for trust management schemes in routing.None of the existing works re-designed or considered for the infrastructure of AMI communications.For AMI deployment, we are focusing on two concerns in this paper such as security and energy-efficiency.The reliability of communicating the periodic electric meter readings with the intended recipient and the security of protecting sensitive data from the various vulnerable threats are important goals for a smart AMI system.

C. Research motivations and Contributions
The State-of-the-art shows that wireless communication security is a big issue in IoT enabled smart systems.The problem becomes challenging for smart AMI systems.DoS attack, malicious attack, black hole attack, and man in the middle attack collapse the system and eventually, Smart Grid (SG) degrades the performance.AMI is a part of SG, and we can save electricity by providing security for the com-munication infrastructure of AMI.Ultimately, SG enhances performance.In short, the key requirement of AMI systems is the securities from the various attacks in wireless communications such as WI-FI/WLAN networks.In general, the cybersecurity requirements of AMI include confidentiality, integrity, availability, etc. that can be vulnerable to wireless security threats during wireless communications.This work motivates by providing a reliable and secure path in the routing of WSN called SRAMI.Communication infrastructure is the main element of AMI.Consequently, through reliable and secure communication infrastructure to AMI, the present work is to support and develop the SG of the electricity sector.
The contributions are:  Optimizing the AMI system by providing the smart communication protocol at the network layer supports reliable route discovery and a lightweight security algorithm for the transmission of electrical data. For reliable route discovery, we used the trust-based approach to select the next relay node for data transmission.In trust-computation, each AMI node is analyzed by computing its trust score using the trust parameters mentioned in figure 1.  For secure data transmission, the lightweight cryptography algorithm is designed that depends on the efficient key management technique, data encryption, and its verification at each intermediate AMI node.The proposed protocol SRAMI is more effective than the above-stated protocols in the state-of-the-art.SRAMI mainly focused on finding a reliable path of communication and the security algorithm works to provide security on a reliable path. The SRAMI protocol simulated and evaluated with state-of-art protocols by considering the different network conditions in terms of parameters mentioned in figure 1.

A. System Design and Assumptions
This section presents the complete design of the proposed SRAMI protocol to address the significant requirements for the calculation process of the reliable path using trust parameters such as energy, geographical distance, and bandwidth and cryptography-based secure data transmission.Figure 2   As per the above system design and assumptions, we proposed SRAMI protocol to address the challenges of reliable and secure data transmissions under cyber threats.SRAMI algorithm goes to provides reliable and secure communication.As per the problem statement, we prepared the below network design parameters for the evaluations of different routing methods for AMI network security.SRAMI algorithm proposed to enhance the performance of SG than the way of suggested literature survey schemes for reliable and secure communication.Batch rekeying operations tried to provide secure communication for AMI by using key management schemes [Benmalek et al. 2018], but it failed to provide reliable communications.

I. Reliable Route Discovery
As observe in figure 3 (2)   −   = ε, ℎ   = 0 (3)   −   < , ℎ   = −1 (4) Where   enduring energy of succeeding hop,   is essential energy to spread the current data and ε is threshold to satisfy.If the equation 2 satisfied then   is trust value set to true from current node  to next node .Else if equation 4 is satisfied then   trust value is set to false from current node  to next node .Otherwise in rare case, trust value is set to 0. This helps to improve reliability through selecting the more stable path for the reliable data transmission.Similarly, the bandwidth is evaluated as:   +   < , ℎ   = 1 (5)   +   = σ, ℎ   = 0 (6)   +   > , ℎ   = −1 (7) Where   bandwidth occupancy at the current node,   is vital bandwidth to convey the recent data and σ is the lower bandwidth limit to satisfy.Geographical the distance from node  to next node  measured as: Where,    &    is the location value   of node  and .The  node is by default the destination node.We considered the maximum distance between two nodes in 750 meters.We received the distance trust value in the range of 0 to 1.The maximum the distance trust value, the better probability of the node to select.The proposed approach is its simple and having minimum overhead of discovering the reliable route.These all equations utilized in algorithm 1 for reliable route discovery in SRAMI protocol.

II. Secure Data Transmission
After discovering the reliable route to start data transmission from source node to destination node, we designed lightweight cryptography approach with effective key management technique.Algorithm 2 shows the processing of sending the data from source to destination node using Elliptic Curve Cryptography (ECC).The ECC is a recent technique introduced as an alternative to Rivest-Shamir-Adleman (RSA) cryptography.ECC provides security among the key pairs using the elliptic curves mathematics.In RSA, a similar kind of approach adopted using prime numbers rather than elliptic curves.ECC gained significant attention recently due to strong security with small key sizes.ECC depends on the structure of elliptic curves for public-key cryptography operations and hence its keys remain very difficult to crack.Due to the security and computational efficiency using ECC, we designed cryptography functions to transmit the electric data from AMI node to intended utility node in algorithm 2. This algorithm not only achieves the data security but also achieves the user privacy preservation.Signing received packet and forward to next  in selected path using set of keys 3.8 END IF 3.9 ELSE 3.10 Packet is received from malicious node 3.11  () 3.12 END IF 4. STOP As demonstrated in algorithm 2, the group of keys generated using ECC key generation technique.The private key  is randomly generated.The  is generated by the elliptic curve parameter.The session key  is generated by using the current session ID for appropriate key management in network as well as protecting the communications.To improve the security further, the group private key  is generated using the public and current session key.The group private key is used for generating the digital signature of encrypted data and its verification.The key size selected for all is 256 bits (which is very small compared to conventional mechanisms) with high-security provisions.As the ECC technique does not provide in-built encryption and decryption capabilities, we combined this approach with the Advanced Encryption Standard (AES-128).AES-128 encrypts and decrypts the electric data from smart meters using the set of keys generated using ECC.

Simulation Results and Discussions
The selection of a reliable and secure path of routing in WSN ended with the SRAMI algorithm.The simulation is possible with the help of NS2 as it evaluates the proposed protocol with exiting protocols following QoS parameters compared.
The proposed SRAMI algorithm is evaluated by choosing a random and grid type of topology [Halle et al. 2020].As the AMI network deployed randomly or grid type, we designed the networks of both scenarios with the number of AMI nodes.Tables 1 and 2 show the set of simulation parameters for random and grid network topologies respectively.In a random network scenario, a varying number of AMI nodes were deployed to verify the scalability and reliability of the SRAMI protocol with a fixed data rate of 20 packets/second.In grid topology, we designed a grid network of 25 AMI nodes with varying data rates in the range of 10 packets/second to 50 packets/second.In both scenarios, we introduced the presence of 10 % malicious attackers.The performance of SRAMI protocol compared with two conventional non-secure protocols AODV and DSDV, and two secure protocols such as AASR (cryptography-based) [Liu et al. 2014] and TERP (trust-based) [Shen et al. 2017].We selected AASR and TERP protocols for comparative study, as our goal to claim the efficiency of using the hybrid approach in SRAMI protocol with the minimum computational burden.For comparative analysis, we computed the five well-known parameters such as average throughput, Packet Delivery Ratio (PDR), delay, overhead, and energy consumption.Figure 4 shows the examples of both scenarios that demonstrate the deployed topologies and communications among the AMI nodes and utility nodes in presence of malicious nodes (redcolored).

A. Random AMIs Evaluation
In random topology, we varied the smart meters density from 10 to 60. Figures 5-9 demonstrate the comparative results for parameters average throughput, PDR, delay, overhead, and energy consumption respectively.Figure 4 demonstrates the outcome of average throughput for a varying number of smart meters.The conventional routing protocols AODV and DSDV have poor throughput results compared to other protocols as they do not have provisions to defend against the malicious nodes in the network.Among other secure protocols, SRAMI improved the throughput performance compared to AASR and TERP protocols due to the provision of establishing the trust-aware route and lightweight cryptography-based data security.
Figure 6 showing the PDR outcomes that overlapping the trend of throughput results.As the number of smart meters increases, the performance of throughput and PDR decreases significantly.It is mainly because of an increasing number of malicious nodes in the network and long-distance communications.Among all the protocols SRAMI leads to significant improvement in PDR performances as minimizes the number of link break probabilities due to malicious users and protects the data from being compromised.It also impacts communication delay performances (figure 7) for SRAMI protocol compared to other protocols.As the frequent route discovery and re-transmissions are reduced in the SRAMI protocol, it significantly reduces the communication delay as well.In some networks, the nonsecure protocol (AODV & DSDV) shows the minimum communication delay compared to secured protocols (AASR & TERP).It is because of extra provisioning provided in AASR and TERP protocols to defend against the malicious nodes.However, the QoS (throughput & PDR) of AODV and DSDV protocols degraded badly due to malicious nodes.Figure 9 shows the outcome of average energy consumption results for all the protocols.The SRAMI protocol minimizes the average energy consumption for each AMI network compared to other protocols; hence, it improves the network lifetime performance.
Figure 9. Performance analysis of average energy consumed for random topology

B. Grid AMI Evaluations
This section presents the simulation results for grid topology with the varying data rate.The purpose of grid AMI designing is to consider the real-time deployment of smart meters for each home.In rural or semi-urban areas of India, housing societies in a grid manner where there is the possibility of electricity theft.By deploying the smart meters in such regions, the electricity theft probability can be neutralized.However, wireless threats may introduce challenges for managing the SG effectively.The result in figures 10-14 demonstrates the outcome of throughput, PDR, delay, overhead, and energy consumption respectively.The data rate variations investigated with grid topology of AMI networks.The higher data rate introduces a higher communication burden, communication delay, and communication overhead for AMI networks. of SRAMI shows the promising among all the secured and non-secured protocols.Among the cryptography-based (AASR) and trust-based (TERP) protocols, the latter one has a high throughput performance and PDR (figure 11) performances (by considering random and grid topologies).TERP focused on establishing more reliable routes without data security and AASR focused on data security without reliable routes.In both cases, complete protection against malicious attackers cannot be achieved.The PDR results show that increasing data leads to increasing packets dropped in the network.
The communication delay shows in figure 12 concerning an increasing number of data rates.The communication delay has increased significantly since after 30 packets/second data rate as it increasing the significant congestions on transmission links.The SRAMI protocol able to keep minimum delay in all cases compared to all the protocols.A similar reflection was noticed for communication overhead (figure 13) as well.The AASR protocol shows high computation overhead compared to the TERP protocol due to cryptography operations.The SRAMI protocol reduces communication overhead because of a reduction in retransmissions and routes discovery operations compared to AASR and TERP protocol.The average energy consumption performance in figure 14 demonstrates no fluctuations with a varying data rate as the fixed number of smart meters in the network.The proposed one reduced the energy consumption performance significantly compared to all the protocols.

Compliance with Ethical Standards:
Funding: No Funding.
Conflict of Interest: All authors declares that they has no conflict of interest.
Ethical approval: This article does not contain any studies with human participants performed by any of the authors.

Figure 1 .
Figure 1.AMI Optimization via security benefits using trust and cryptographybased approaches

Figure 2 .
Figure 2. Structure of AMI systemThe design of AMI systems based on assumptions such as:-All the AMI nodes are equipped with the functionality of periodical meter reading of electricity consumptions of all connected devices.In short, such smart meter nodes act as the sensing node that periodically senses the electricity data reading and transmitting towards the utility node.-The AMI nodes are constrained by processing capabilities and processing power.-Theutility node is outside of the network without any resource constraint.-Themalicious nodes are part of the AMI network that performs the malicious activities by spreading false information among the neighboring AMI nodes.-The data from source AMI node to destination utility node transmitting in a multi-hop manner.

Figure 5 .Figure 7 .
Figure 5. Performance analysis of average throughput for random topology

Figure 10 .
Figure 10.Performance analysis of average throughput for grid AMI network

FigureFigure 13 .
Figure 12.Performance analysis of delay for grid AMI network

further
They designed modified Diffie-Hellman for secure communications in WSN.AlMajed et al. (2020) proposed authenticated encryption based on plain text improved mapping phase into the elliptic curve to protect against various security threats.Chaitra et al. (2021) proposed SEEDT (Secure and Energy-Efficient Data Transmission) proto-col.The clustering performed by multi-objective function and ECC used for secure data transmission in WSN.
Ali et al. (2020)rust-based Secure Directed Diffusion Routing Protocol) using ETM for WSN.Kore et al. (2020) proposed a cross-layer trust model called IC-MADS (IoT enabled Cross-layer Man-in-Middle Attack Detection System).They designed IC-MADS in two phases clustering and attack detection.Poomagal et al. (2020) proposed secure data transmission among the vehicular nodes using ECC.They designed ECC for satellite communication and key agreement for secure message transmission.Ali et al. (2020)proposed data security mechanisms in WSN with minimum response time and computational efforts.

Table 1 .
State-of-the-art of the wireless network with security solution

Table 1
demonstrates the comparative study of all the recent security solutions to protect the wireless networks (WSN, IoT, MANET, etc.) in terms of methodology, performance parameters, attacks, etc.The security methods include the trust-based, cryptography-based, and combination of both trust-based and cryptography-based methods.The SRAMI work proposed in this paper by considering the AMI system requirements of security.Designing wireless communication security for AMI is the basic need in the electricity sector to make the smart grid.AMI worsens the performance because of no security provisions for AMI.Hence, SRAMI proposed to address the concerns of reliability and security for data transmission operations in a WSN-assisted AMI network.B.

Trust Management for Communication Infrastructure Trust
[Mahajan et al. 2020]cation infrastructure becomes essentials for reliable data transmissions.Some recent works introduced by considering the real-time communication infrastructure.AMI communication infrastructure, designing factors should be logical, which gives faithful end to end delivery.Some of the parameters are Network topology design, secure routing protocol, secure forwarding, end to end communication, secure broadcasting, and DoS defense.For any wireless network, the selection of a trusted node is one of the vital tasks[Mahajan et al. 2020].In this paper, we proposed WSN for wireless communication to AMI and its result increases sensor nodes to transfer the information from one place to another place.Secure selection of sensor node performs according to the trust-evaluation algorithm.Table2presents the literature review of some trust-based algorithm for security improvement of communication infrastructure.The trust management algo- demonstrates the AMI design considered in this paper.As showing in figure 2, the AMI system consists of  number of AMI nodes { 1 ,  2 , …   } called smart meters deployed at edge layer randomly in area of size  × .The data collected by AMI nodes transmitted to corresponding local gateway nodes and then to destination node  called utility node via intermediate relay AMI nodes.Figure2also demonstrates that how each smart meter connected to various electric equipments such as fridge, television (TV), bulbs, etc.
, after the AMI network deployment with a group of sensor nodes and utility node, the reactive route construction process starts by any source node  in the network by generating and spreading Route Request packets near the actual receiver node .RREQs are broadcasted to all the sensors within the near to  as per demand to search the trustworthy and a reliable route.All the neighbors of node source or current intermediate node are recorded into the set    (set of all neighbors of node  towards endpoint or receiver D (Utility node)).Thus, the path as of existing node  to succeeding node  is constructed by computing the trust score of each neighboring node of node .The three most important parameters are calculated for finding a trust-based routing path.The parameters are such as energy capabilities, bandwidth capabilities, and geographic distance between  ℎ sensor nodes to Utility node () in the AMI network.The computed values of each neighboring node are considered as the trust value and at each session, it can be updated in the routing-table.Once the RREQs received by neighboring nodes, then the probability is computed as: is the trust value of node  for becoming the relay of node .The   and   is the energy capabilities and bandwidth capabilities of node .As the sensor nodes are fixed position, our aim is to select the path with minimum geographic distance from  to .   is the geographic distance from node  ℎ to  ℎ node.The conditions for energy and bandwidth at each node are evaluated as: The calculations of energy consumption founded to next hop selection are below three equations:   −   > , ℎ   = 1