IT201900018935A1 - COMPUTER-IMPLEMENTED METHOD FOR REACHING A DISTRIBUTED CONSENSUS IN A BLOCKCHAIN NETWORK AND NODE IMPLEMENTING THE METHOD. - Google Patents

Description

Description of the industrial invention entitled:

"METHOD IMPLEMENTED BY MEANS OF A COMPUTER TO REACH A DISTRIBUTED CONSENT IN A BLOCKCHAIN AND NODE NETWORK IMPLEMENTING THE METHOD"

DESCRIPTION

FIELD OF TECHNIQUE

The present invention relates to a computer-implemented method of achieving bet-based distributed consensus for a distributed ledger, such as blockchain technology.

STATE OF THE TECHNIQUE

The invention is particularly suitable, but not limited, to reach a higher level of distribution of a network based on bets, in order not to have too many bets concentrated in too few nodes, thus increasing the solidity of the network and reducing its vulnerability. to any node that is compromised.

In this paper, the term blockchain is used to encompass all forms of electronic, computer-based and distributed ledgers. These include, but are not limited to, blockchain and transaction chain technologies, authorized and unauthorized ledgers, consensus based ledgers, shared ledgers and their variations. The best-known application of blockchain technology is the Bitcoin ledger, although other blockchain implementations have been proposed and developed. A blockchain is a consensus-based electronic ledger, which is implemented as a computer-based decentralized and distributed system, consisting of blocks which in turn are made up of transactions and other information. Bitcoin is an example of a proof-of-work blockchain where traders perform expensive calculations to facilitate transactions on the blockchain. Proof-of-work blockchains have been criticized due to the large computing resources required, which requires large energy consumption to operate, and also due to the fact that block generation can be erratic and slow. Also, several blocks must be built on top of a given block before the given block is considered confirmed (that is, it is sufficiently unlikely that it will be restored).

Proof of stake blockchains have been proposed as an alternative to proof of work blockchains. In a proof of stake blockchain network, the blockchain is protected by proof of stake rather than proof of work. In a proof of stake, traders hold a stake (by depositing some tokens) on a special account. This stake can be referred to as a security deposit and the likelihood of being selected as the node to mine a block is proportional to the amount of digital assets provided as a security deposit. Proof of stake blockchain networks can be used to avoid the computational fees and energy required for proof-of-work blockchain mining. Additionally, proof-of-stake blockchains can allow for higher frequency and smoother block creation than proof-of-work blockchains. At least some proof of stake blockchains also have a low probability of bifurcation and a block can be effectively confirmed as soon as it is added to the blockchain.

As mentioned above, although the economy of scale leads to the concentration of mining resources to increase the efficiency of the network, in the development of a distributed network, the concentration is a problem related to the control of the network, the robustness of the network since it is unable to perform scheduled maintenance operations on nodes. The main problem in small and concentrated networks is the low redundancy in case of failures, be it network problems, hardware or software. The opposite situation, in which there is excessive fragmentation, can also be counterproductive. This is because the transmission and replication of information always present an overload within a real system. Furthermore, the assignment of bets directly to mining nodes can lead to overload problems if there are distribution constraints.

Therefore, it is necessary to address the concentration problems mentioned above and improve the consensus protocol with proof of stake to obtain a protocol that encourages distribution and at the same time the efficiency of the network.

AIMS OF THE INVENTION

An object of the present invention, according to a first of its aspects, is to obtain a computer-implemented method capable of encouraging the distribution of the network and the efficiency of miners, and at the same time optimizing the scalability of the blockchain.

A second object of the present invention is to obtain a computer-implemented method capable of improving the robustness of the network.

A further object is to optimize the method implemented by the computer, in order to increase the overall security of the blockchain network.

A further object of the present invention is to provide a computer-implemented method which encourages the use of honestly performing nodes to extract the next block and to discourage the use of nodes which do not perform well or have intentionally bad or malicious behavior. harmful.

An object of the present invention is also that of keeping the energy consumption of the network low and possibly increasing the computing power of the node only when necessary.

A further object of the present invention is to provide a system which implements the computer-implemented method in an easily designable way, with an optimized and extremely efficient energy and calculation consumption.

BRIEF DESCRIPTION OF THE INVENTION

Some technical aspects of the present invention are summarized below, which allow to achieve some of the most important purposes.

According to a first aspect, this invention refers to a computer-implemented method to reach a distributed consensus based on proof of stake for a blockchain network, thus allowing to reach a higher level of network distribution and not to have too many bets concentrated in too few nodes, said blockchain network comprising one or more MAIN addresses, for each MAIN address being associated in turn with one or more OVERFLOW addresses / nodes, the latter being able to act as potential mining nodes, only if a predetermined minimum quantity of bets to said one or more OVERFLOW addresses / nodes, including the following processes:

- a redistribution process configured to calculate the minimum stake needed for an OVERFLOW node to be extracted, being minimum_stake_value = total_amount_at stake / predetermined_target_number_of_nodes, to calculate the maximum stake that allows efficient mining, being maximum_stake_value = minimum_stake_value * 2, for each MAIN address to distribute the stake among the OVERFLOW nodes, to iterate the OVERFLOW nodes of a given MAIN address and to assign the minimum_stake_value to each OVERFLOW node, until the remaining stake is sufficient to activate a miner on each overflow node that has been assigned to the his stake;

- a penalty process configured to freeze an amount of reward corresponding to the computational work for having mined a number of slots beyond a predetermined limit Slot_MAX in a period, and to recover it in the following periods starting from period i + 1, a little at a time based on a number of parts proportional to the number of slots mined on said predetermined Slot_MAX limit, being a part assigned to each subsequent slot that the miner can be delegated to extract, this in order to naturally incentivize the distribution of the blockchain network.

Such a computer-implemented method naturally allows for a high distribution of the blockchain network.

On the basis of a second aspect, the present invention refers to a method implemented by computer, in which the redistribution process can be further configured in such a way that, if any bet remains, it is distributed among the OVERFLOW nodes up to when the greatest possible number of nodes has a stake equal to the maximum stake value and each further stake is assigned to the first OVERFLOW node in a predetermined order.

It allows to optimize the redistribution of the total quota assigned to a pool of OVERFLOW nodes connected to a MAIN address and ensures that only the first OVERFLOW nodes on the basis of a predetermined univocal progressive numbering can be under a penalty regime According to a third aspect, the present invention relates to a computer-implemented method further comprising an initialization process configured to list all the MAIN addresses of the blockchain network, for each list of MAIN addresses being assigned OVERFLOW addresses / nodes, and therefore for each OVERFLOW address / node it transfers the quota , if it has one, to its assigned MAIN address.

It allows you to optimize the redistribution process in the event that there may be any quota on an OVERFLOW node before the implementation of the distribution process.

According to a fourth aspect, the present invention relates to a computer-implemented method which further comprises a maintenance process, comprising the following operations, to which a MAIN address and at least two or more OVERFLOW addresses are assigned and a node to be maintained:

110: sending a transaction comprising a first input that provides a replacement OVERFLOW address using its private code, which means that said replacement OVERFLOW address is a node that replaces the node you want to maintain

120: assignment of said substitute OVERFLOW address to the corresponding MAIN address using the private code of the MAIN address

130: cancellation of the desired registration to be maintained using your private code. Alternatively, the MAIN address can UNASSIGN an overflow node previously assigned to it.

It allows you to maintain the highly distributed blockchain network even in the event of the failure of one or more mining nodes.

According to further aspects, the following invention relates to:

- a node configured to implement a method to reach a distributed consensus based on proof of stake for a blockchain network, thus allowing to reach a higher level of network distribution and not to have too many bets concentrated in too few nodes, said node including an interface device, a processor coupled to said interface device, a memory coupled to the processor, the memory having stored therein executable instructions which, when executed, configure the processor to perform the corresponding operations of the method implemented by the computer described.

- a computer readable storage medium comprising computer executable instructions which, when executed, configure a processor to execute the method implemented by the described computer.

- a computer network to perform a blockchain-based transaction from a first user to a second user in an electronic transfer network of connected nodes, where the nodes comprise at least one first node, a second node described, in which each node shares a same blockchain, and in which each node stores a described computer-readable storage medium.

BRIEF DESCRIPTION OF THE FIGURES

The structural and functional characteristics of the present invention and its advantages with respect to the prior known art, will become even clearer on the basis of the following claims, and in particular from an examination of the following description, carried out with reference to the attached figures which show a shape of preferred but not limited schematic embodiment of the method, system, device implemented by computer according to the invention, in which:

Figure 1 illustrates a flow chart of the computer implemented method according to the present invention;

Figure 2 illustrates an operational example of the registration process of the computer-implemented method according to the present invention;

Figure 3 illustrates a diagram of an example of the network comprising a MAIN address and three OVERFLOW nodes according to the present invention;

Figure 4 illustrates a flow diagram of the initialization process of the computer implemented method according to the present invention;

Figure 5 illustrates an operational flow diagram of the initialization process of the computer implemented method according to the present invention;

Figure 6 illustrates an operational example of the initialization process of the computer-implemented method according to the present invention;

Figure 7 illustrates a flow diagram of the redistribution process of the computer implemented method according to the present invention;

Figure 8 illustrates an operational flow diagram of the redistribution process of the computer implemented method according to the present invention;

Figure 8 illustrates an operational flow diagram of the redistribution process of the computer implemented method according to the present invention;

Figure 9 illustrates an operational example of the redistribution process of the computer implemented method according to the present invention;

Figure 10 illustrates a flow diagram of the penalty process of the computer implemented method according to the present invention.

DETAILED DESCRIPTION OF THE INVENTION

In general, this disclosure describes a computer-implemented method, a system for achieving distributed consensus based on proof of stake for a distributed ledger, such as blockchain technology. In particular, the present invention relates to a proof of stake-based blockchain technology, in which stake holders cannot and must not communicate with mining nodes.

The blockchain technology according to the present invention provides for two main actors: stake holders, i.e. those who possess tokens that allow control of the chain, and miners, i.e. those who control mining nodes actively operating in the network and charged with a sufficient amount of you bet to get the job of a certain number of job slots, where they extract a corresponding determined number of blocks thus lengthening the chain.

The blockchain is considered a meta-actor and is the union of the nodes operating on the network. Sending a transaction to the blockchain involves this transaction reaching each operating node which constantly updates its internal status. These update procedures are constructed in such a way as to ensure that the same initial conditions produce identical status updates in each node in a strictly deterministic way. This meta-actor represents the massive upgrade procedure on each node in the network.

From now on it is understood that:

- one slot is a time interval of 30 seconds. These slots are uniquely assigned to the mining nodes and within a given slot one and only one node has the right to produce a block.

This assignment is unique and strictly deterministic and is performed before the start of each period;

- for a period is meant a grouping of contiguous slots. Each period consists of an equal number of slots. The allotment and redistribution of slots and shares are calculated once for each period and remain unchanged until the end of the period.

- a bet is a numerical representation of the voting rights of each stakeholder in the network. The bet can be delegated with specific rules and operations to the nodes involved in the creation of a block.

- MAIN address: this address is not associated with a physical object (a MAIN address is not associated with an active mining node) and acts as a placeholder which indicates a group of physical resources to which you want the blockchain to be able to access, thus exposing its private code only during an OVERFLOW address assignment transaction to the MAIN address and not during another operational transaction, increasing the overall security of the chain; this MAIN address (for example alice_main_address) is declared by its owner and is managed through its private code (for example alice_key). Only the MAIN address can declare which OVERFLOW nodes are assigned to it;

- OVERFLOW address: this address is associated with a physical object and acts as an active mining node; this OVERFLOW address (for example node_1_overflow_address) is managed using its private code (for example node_1_key); it is possible that the MAIN address manager and the OVERFLOW node manager are the same entity, but they use different codes when impersonating Alice or Node_1; the owner of each MAIN address (for example Alice) using a specific transaction associates each registered overflow node to its MAIN address. Only the MAIN address in the network can link an OVERFLOW address to itself, which means that only the MAIN address can send an ASSIGN type transaction which links an OVERFLOW address to a MAIN address. This operation is not commutative, ie an OVERFLOW address may not assign itself to a MAIN address. The OVERFLOW address / node may not be shared between MAIN addresses, an OVERFLOW address can only be assigned to one MAIN address. Only an address previously declared as OVERFLOW can be associated with a MAIN address. An OVERFLOW node can also unregister and thereby removes itself from both the pool of available nodes and the MAIN address to which it was previously assigned. This is to give both the MAIN and OVERFLOW nodes the ability to protect them from unwanted situations in which they are linked together, whether this is the result of an error in good faith or of malicious behavior.

If an OVERFLOW address is not bound to any MAIN address, it will only operate as a replication node.

The computer implemented method according to the present invention (FIG. 1) comprises a redistribution process, a penalty process and may comprise a registration process, an initialization process and a maintenance process.

The computer-implemented method according to the present invention is set according to the following parameters and values, where said values are to be understood only as examples and can be modified according to specific technical requirements:

Hourly slot duration = 30 seconds

Slots in period (E_S) = 24000 => number of slots for each period

Penalty Factor = 2 => determines the amount of time penalty for violating the current protocol

Recovery factor = 1 => determines the age of return from the penalty regime

Target number of nodes (MAX_N) = 400 => optimal number of nodes forming the network

Minimum number of nodes (min_N) = 200

Minimum slot per node Slot_min (E_S / MAX_N) = 60

Maximum slot per node Slot_MAX (E_S / min_N) = 120

Evaluation point of a period (EEP) (E_S / 3) = 8000

The registration process is configured to register one or more MAIN addresses, to register one or more OVERFLOW addresses / nodes and to associate one or more OVERFLOW addresses / nodes to a MAIN address and to register such association.

The registration process (fig. 2, 3) implemented on each overflow node of the blockchain network includes the following operations:

5: via the blockchain network, receive a transaction including a MAIN address as input and register it in an address register using a private code of the address manager MAIN 10: via the blockchain network, receive a transaction including one or more addresses as input OVERFLOW and register one or more OVERFLOW addresses in the address register using a private code of the manager of each OVERFLOW address

15: through the blockchain network, receive a transaction including as input the assignment of one or more OVERFLOW addresses to the MAIN address registered in operation 5.

The method implemented by computer according to the present invention provides for setting a minimum amount of bets (S_min_VAL) assigned to an OVERFLOW node to be activated and considered as a mining node and in order to avoid that all bets concentrated only on a few nodes are set , by setting an upper limit corresponding to the amount of bets that can be assigned to a node, beyond which the mining node begins to incur penalties. This should make a concentration of wagers beyond efficiency ranges less appealing to nodes looking to reap the highest possible rewards. The maximum amount of desired bets (S_MAX_VAL) can be double the value of S_min_VAL.

The algorithm is designed to maximize the number of active nodes by spreading the excess bet among as many overflow nodes as possible in order to expand the network thereby increasing its resilience.

The computer-implemented method comprises an initialization process in order to retrieve the incorrectly addressed bet on an OVERFLOW node and to collect them in the corresponding MAIN address before carrying out a redistribution process according to the computer-implemented method of the invention. The initialization process is useful when bets are incorrectly addressed on one or more OVERFLOW nodes, thus making distribution between OVERFLOW nodes reliable. Also, the initialization process is useful for ensuring that OVERFLOW addresses are not assigned bets before running the redistribution process.

The initialization process (fig. 4) is configured to list all the MAIN addresses of the blockchain network, for each MAIN address list its assigned OVERFLOW addresses / nodes, and then for each OVERFLOW address / node to transfer the action, if has, to its assigned MAIN address. The initialization process (figs. 5, 6) implemented on each overflow node of the blockchain network in particular includes the following operations:

20: have a blockchain network comprising one or more MAIN addresses, for each MAIN address being associated in turn with one or more OVERFLOW addresses / nodes which act as potential mining nodes only if a predetermined minimum amount of bets is assigned to said one or more OVERFLOW address / node;

25: through the blockchain network, receiving a transaction comprising as input addresses from the address register which includes MAIN addresses and their corresponding OVERFLOW addresses;

30: associate a unique progressive number to each MAIN address;

35: for each MAIN address check if it contains a quantity of bets and if this value is> 0, assign this value to parameter M_A_S, and if this value is <= 0, perform operation 35 until the next MAIN address element of the blockchain network and generate a transaction including as output the execution of the value M_A_S associated with the corresponding MAIN address,

40: associate a unique progressive number to each OVERFLOW node

45: for each OVERFLOW node assigned to each corresponding MAIN address check if it contains a defined amount of OVERFLOW_ADDRESS.assigned_stake and if this value is> 0, assign this value to the O_A_S parameter, calculating the MAIN_ADDRESS.assigned_stake value as M_A_S O_A_S and assign O_A_S = 0 and generate a transaction including as output the updated value of the M_A_S value associated with the corresponding MAIN address, then carry out operation 45 up to the next OVERFLOW address assigned to the corresponding MAIN address considered in operation 35,

and if OVERFLOW_ADDRESS.assigned_stake is <= 0, perform operation 45 up to the next OVERFLOW address assigned to the corresponding MAIN address considered in operation 35, 50: generate a transaction including as output the updated value of the TOTAL_AT_STAKE value for each MAIN address of the blockchain network corresponding to the total amount of bets on each MAIN address after performing the above operations.

The redistribution process (fig. 7) is configured to calculate the minimum stake needed to extract an OVERFLOW node, being minimum_stake_value = total_amount_at stake / predetermined_target_number_of_nodes, to calculate the maximum stake that allows effective mining, being the maximum stake = value minimum stake * 2, for each MAIN address by distributing the stake among the OVERFLOW nodes, scrolling through the OVERFLOW nodes of a given MAIN address and assigning each OVERFLOW node the minimum_stake_value, until the remaining stake is sufficient to trigger a miner or each overflow node has not been allocated its stake. The redistribution process can also be configured in such a way that if a stake remains, it is distributed among the OVERFLOW nodes until as many nodes as possible have a stake equal to the maximum stake and each other stake is awarded to the stake. first OVERFLOW node in a predetermined order.

The redistribution process (figs. 8, 9) in turn includes a first iteration redistribution process in which the algorithm assigns to each available OVERFLOW node assigned to each MAIN address the minimum amount of bets necessary to become a active mining, and in a second iteration redistribution process the algorithm distributes the stake trying to maximize the number of overflow nodes operating in a full stake in view of the S_MAX_VAL set for any OVERFLOW node.

The redistribution process includes in particular the following operations:

52: have a blockchain network comprising one or more MAIN addresses, for each MAIN address being associated in turn with one or more OVERFLOW addresses / nodes which act as potential mining nodes only if a predetermined minimum amount of bets is assigned to said one or more OVERFLOW addresses / nodes;

55: get as input a TOTAL_AT_STAKE value for each MAIN address of the blockchain network corresponding to the total amount of bets on each MAIN address

60: calculate S_min_VAL = TOTAL_AT_STAKE / TARGET NUMBER OF_NODES being TARGET NUMBER OF_NODES = total number of OVERFLOW mining nodes in the chain network,

Calculate S_MAX_VAL = S_min_VAL * 2

Assuming a counter value ITERATION = 0

65: If the ITERATION value is <2, then update the ITERATION = ITERATION 1 counter value

70: for each MAIN address that gets as input the value MAIN_ADDRESS.assigned_stake and if this value is> 0, which means that the MAIN address in question contains a quantity of bets, assign this value to the parameter M_A_S,

and if the M_A_S value is> = S_MAX_VAL repeat operation 70 up to the next MAIN address, 75: if the M_A_S value is <S_MAX_VAL, obtain as input the OVERFLOW_ADDRESS.assigned_stake value for each OVERFLOW address associated with the MAIN in question, and

if the OVERFLOW_ADDRESS.assigned_stake value is = 0 repeat operation 75 until the next OVERFLOW address associated with the MAIN in question and

if the OVERFLOW_ADDRESS.assigned_stake value is ≠ 0, assign the OVERFLOW_ADDRESS.assigned_stake value to the O_A_S parameter, calculating MAIN_ADDRESS.assigned_stake = M_A_S-S_min_VAL, and calculating OVERFLOW_ADDRESS.assigner_stake_address = O_A_Stake, and repeat the OVERFLOW_ADDRESS.assigner_stake_ address = O_A_Stake 75 to the MAIN under consideration.

The penalty process makes mining operations that do not respect the established rules disadvantageous, thus incentivizing the nodes to naturally expand the network thus increasing its resilience, as already discussed.

From now on we mean:

periodi, the period in which miners have registered their MAIN and OVERFLOW addresses, periodi + 1, the period in which miners mine their assigned slots,

periodi + 2, the period in which prizes and penalties are evaluated.

The subscript "j" refers to the blockj written in slotj

The subscript "i" is about periodi and what is relevant to periodi.

The computer-implemented method according to the present invention includes a bonus for miners for remuneration for their computing effort during their mining operation. If a miner is unable to write a block in the corresponding slot or mines in a number of slots beyond a predetermined limit, he will not receive the corresponding reward for the unwritten block or for the mined block beyond the predetermined limit at the next time ( only one block can be written in each slot), but this prize (for blocks mined over the limit) is frozen and will be recovered very slowly a little at a time (for example according to a number of parts equal to twice the number of slots mined under penalty regime) for each subsequent slot that will have been mined by the miner, in particular on the basis of a number of parts proportional to the number of mined slots beyond said predetermined limit, Slot_MAX, being a part assigned to each of the following slots that miner it can be delegated to mine, therefore in order to naturally incentivize the distribution of the blockchain network.

As mentioned above, a prize i + 2 corresponds to the prize for mining a blockj 1 in periodi 1. This prizei 2 includes a monetary base (i.e. a new coin created) for the block, i 1 mined, and and a collection feej, i i + 1. This collection commissionj, i 1 in turn includes the collection commission paid by users who wish to execute a transaction and written in the relevant blockj, i 1 of the blockchain network.

From now on we mean:

Penalty_slotsi + 1 = number of slots mined over the limit in the periodi + 1 or number of slots on Slot_MAX * 2

Frozen commissioni 1 = is the amount of reward not paid to miners in periodi 1 for having mined a number of slots above the Slot_MAX limit, thus having been subject to a penalty regime.

The penalty process (fig. 10) implemented on each overflow node of the blockchain network includes in particular the following operations:

80: calculate the total frozen commissioni + 1 as the prize amounti + 2 not paid out to miners at a specific timei 1 for mining a number of slots over the Slot_MAX limit

85: calculate the penalty_slotsi + 1 preferably twice (or according to another proportional law) the number of slots mined over the limit, i.e. the number of slots on Slot_MAX

90: calculate recovered_feej, i + 1 value as frozen commissioni + 1 / penalty_slotsi + 1, i.e. the commission to be recovered for each slot

95: add the recovered_feej value, i + 1 to the MAIN_address prize for blockmining i in the period i 1 as a prizej, i 1 = a collect_feej coinbase, i 1 recovered_feej, i 1

100: calculate the new value of the parameter frozen_feei + 1 as frozen_feei + 1 -recovered_feej, i + 1 and the new value of penalty_slots i 1 as penalty_slotsii 1 -1 and repeat operation 95 for the following blockj + 1 in the periodi + 1 while frozen_fee i 1> 0 or while penalty_slotsi + 2> 0.

At the end of each period, corresponding to the last mined block, penalty_slots and frozen_fee are transferred from the first OVERFLOW node to its MAIN address. Similarly, at the beginning of each period, corresponding to the first mined block, the penalty_slots and frozen_fee are transferred from the MAIN address to its first OVERFLOW node.

This procedure is necessary to avoid malicious behavior such as undue transfers of tokens from one pool to another.

Furthermore, it is always possible to register a new overflow node and calculate its address so that it precedes the one with the assigned penalties in the overflow order. In doing so, a pool would recoup the frozen fees without getting back into optimal parameters or extending the network. Considering that under normal circumstances the fees exceed the generated monetary base this would significantly reduce the effectiveness of the penalties.

As regards scheduled maintenance events in order to maintain the highly distributed blockchain network even in the event of a failure of one or more mining nodes, the invention implemented by computers according to the present invention can comprise a maintenance process comprising in turn the following operations, receiving a MAIN address and at least two or more OVERFLOW addresses and a node to be maintained:

110: send of a transaction comprising a first input that provides a replacement OVERFLOW address using its private code, which means that said replacement OVERFLOW address is a node that replaces the node you want to maintain

120: assign said substitute OVERFLOW address to the corresponding MAIN address using the private code of the MAIN address

130 unregister the node you want to maintain using its private code. Alternatively, the MAIN address can UNASSIGN an overflow node previously assigned to it.

As regards the security aspects of the blockchain according to the present invention, according to the way in which the assignment system is structured, private codes of the MAIN address come into play only for the short moment of time necessary to generate the registration and the transaction of assignment, however, they are no longer involved in any subsequent processing phase. This exempts it from being available online on an object that could be subject to malicious attacks. On the other hand, a constant online presence of a physical server and the corresponding codes necessary to operate on the network allow a malicious third party to steal their private codes and to take all the actions delegated to the set of MAIN addresses.

The present invention also relates to a node (preferably an OVERFLOW node) configured to implement the described computer-implemented method, said node comprising an interface device, a processor coupled to said interface device, a memory coupled to the processor, the memory having stored on it instructions executable by a computer which, once executed, configure the processor to perform the corresponding operations of the method implemented by the described computer.

The present invention also relates to a computer readable storage medium comprising computer executable instructions which, when executed, configure a processor to execute the described computer implemented method.

The present invention also relates to a computer network for performing a blockchain-based transaction from a first user to a second user in an electronic transfer network of connected nodes, in which the nodes comprise at least a first node, a second node, as described previously, in which each node shares a same blockchain, and in which each node stores a computer-readable storage medium described above.

Some operational examples of the computer-implemented method according to the present invention are described below.

First example: attacker

Suppose two MAIN addresses:

� Alice, alice_main_addr, node_a1, node_a2, ..., node_a10 episode 30.000

� Mallory, mallory_main_addr, node_m1, node_m2,…, stake node_m530.000

Total bets: 1,200,000

Minimum bet: 3,000 Maximum bet: 6,000

Min slot: 60 Max slot: 120

Each block generated is awarded a reward of 1 token

In period 14 Alice and Mallory completed the registration of their main addresses and assigned their nodes to them.

These will be active starting from period 15

In period 16 the premium is calculated.

At the start of period 15 Alice will have 10 active nodes (node_a1, ..., node_a10) each with 60 assigned slots. At the start of period 15, Mallory will have 5 active nodes (node_m1, ..., node_m5) each with 120 assigned slots.

Alice and Mallory's slots are the same as they have the same active bet. However, the distribution changes. Mallory has only 5 knots while Alice has 10. None of these are penalized. Alice's nodes are active with the minimum, Mallory's nodes with the maximum.

Suppose a perfect execution of the protocol in period 15, with no transactions to simplify the example.

The ten Alice nodes have assigned 60 slots, thus producing 60 blocks; in each block there will be a reward of 1 token. Since there are ten nodes (60 * 10 * 1) the prize will be 600 tokens. At the start of period 16, Alice's balance will increase from 30,000 to 30,600 tokens.

Let's now turn to Mallory: Mallory's five nodes have assigned 120 slots, thus producing 120 blocks; in each block there will be a reward of 1 token. As there are five nodes (120 * 5 * 1), the expected prize will be 600 tokens. At the start of period 16, Mallory's balance is expected to increase from 30,000 to 30,600 tokens. Only three of the Mallory nodes, node_m3, node_m4, node_m5, were modified to function incorrectly. The causes of this behavior can be various and now we will list some of them.

● An intentional change was made in order to perform invalid operations (duplicate transactions, accepting false transactions with wrong code, ...) = Intentional malicious interaction.

● The server on which the node is installed is malfunctioning and fails the mathematical operations by marking as valid transactions invalid for example Carol, with a budget of 100 tokens, and sends 25 to Bob. The operation is legal but is marked as illegal in the product block. = Unintentional malicious interaction.

● The server's network connection is unreliable and the block is not transmitted in time. = Unintentional malicious interaction.

● The block is not linked to its immediate valid predecessor but to another predecessor that violates the rule of extending the heaviest chain. In this way, the node aims to make unwanted transactions, such as a payment, disappear from the history of the chain = Unintentional Malicious Interaction.

These aspects ensure that the blocks generated by node_m5 are rejected by the other nodes and do not become part of the blockchain.

In this way the prize collected by Mallory will be:

We now pass to an evaluation of the remaining part of the network.

Alice and Mallory have a total (30,000 Alice and 30,000 Mallory) 60,000 stake tokens.

On the rest of the network (1,200,000 - 60,000) 1,140,000 tokens remain in play.

Assuming that all the rest of the network works in ideal conditions and without transactions with the other nodes that make up the network, we will assign:

22,800 blocks will be generated and therefore a prize of 22,800 tokens.

In period 15 the total stake of the network is 1,200,000, the stake assigned to non-Alice or Mallory nodes is 1,140,000. In period 16 the prize for the aforementioned nodes will be 22,800 tokens, bringing their stake to 1,162,800 tokens.

Considering that the main Mallory has collected a lower bet, we have recalculated the values for the following period.

The overall stake is 1,200,000 600 240 22,800 = 1,223,640

The table below compares the control rate of the network at time 15 with the revaluation in period 16 (expected) and 16 (real)

As you can see, Mallory's computing power has not been reduced (the algorithm is not punitive in the classic sense of removing something) but has grown more slowly than that of the other components of the network. This caused, at the recalculation point of moment 16, a change in the allocation of the slots with Mallory for which 0.03% of the computing power was removed from the total network slots. Instead, comparing it to its previous slot number, the reduction corresponds to the loss of 1.17% of the slots.

Second example: unbalanced bet with commission

Suppose two MAIN addresses:

� Alice, alice_main_addr, node_a1, node_a2,…, stake node_a10

Stake 30,000

� Mallory, mallory_main_addr, node_m1

Stake 30,000

Total stake: 1,200,000

Minimum bet: 3,000 Maximum bet: 6,000

Min slot: 60 Max slot: 120

Each block generated is awarded a reward of 1 token

In time 14 Alice and Mallory have completed the registration of their main addresses and have assigned their nodes.

These will be active starting from period 15

In period 16 the premium will be calculated.

At the start of period 15 Alice will have ten active nodes (node_a1, ..., node_a10) each with 60 assigned slots.

At the start of period 15, Mallory will have an active node_m1 node with 600 assigned slots.

Alice and Mallory's slots are the same as they have the same active bet. However, the distribution changes. Mallory has only one knot while Alice has ten. Mallory is penalized. Alice's nodes are active with the minimum, Mallory's node is 480 slots over the limit and for this she will receive a penalty.

Assuming a perfect execution of the protocol in period 15, let's assume that an average of 5 tokens / block is the amount of the commission collected, for simplicity.

Alice's ten nodes have assigned 60 slots, thus producing 60 blocks, and in each block there will be a prize of 1 token. Since there are ten nodes (60 * 10 * 1) the prize will be 600 tokens. At the start of period 16, Alice's balance will increase from 30,000 to 30,600 tokens. To these we will add the rewards of the operations 600 * 5 = 3,000 tokens. Alice will total 33,600 tokens as a result of operations in period 15.

Let's now move on to Mallory: Mallory's node has assigned 600 slots, so it will produce 600 blocks, and for the first 120 blocks a prize of 1 token will be awarded. For the following 480 the prize will be 0. As for the prizes, using the even distribution indicated in the preliminary conditions, Mallory will be awarded the prize only in the first 120 slots. For these he will receive 5 * 120 = 600 tokens. The remaining 2,400 are frozen to be returned over a period of 480 * 2 = 960 slots within limits. Assuming that the node configuration does not change from period 15 to period 16, we evaluate what happens to the EEP of period 16 for period 17.

At the start of period 16, Mallory's balance is expected to increase from 30,000 to 30,600 tokens as a result of the monetary base and an additional 3,000 tokens should be awarded due to fees. Only that Mallory has activated too few nodes to meet the network requirements.

Let us now move on to an assessment of the remainder of the network.

Alice and Mallory have a total (30,000 Alice and 30,000 Mallory) 60,000 stake tokens.

On the rest of the network (1,200,000 - 60,000) = 1,140,000 tokens remain in play.

Assuming that all the rest of the network works in ideal conditions and without transactions with the other nodes that make up the network, we will assign:

22,800 blocks will be generated and therefore a prize of 22,800 tokens for the monetary base part. For the commission portion 114,000 tokens

Previous bets of other nodes 1,140,000

Other monetary base nodes 22,800

Other concession nodes 114,000 =

Total other nodes 1,276,800

Pre-Alice bet 30,000

Monetary base of Alice 600

Alice's commission 3.000 =

Total Alice 33,600

Stake prior to the period of 30,000

Mallory

Mallory monetary base 120

Mallory commission 600 =

Mallory's total 30,720 =

Total stake for new period 1,341,120

The table below compares the control rate of the network at time 15 with the revaluation in period 16 (expected) and 16 (real)

As can be seen, Mallory's computational power has not been reduced (the algorithm is not punitive in the classic sense of removing something) but has grown more slowly than that of the other components of the network. This caused, at the point of recalculation of period 16, a change in the allocation of the slots with Mallory which was seen to take away 0.21% of the computing power on the network. Compared to its previous slot number, the reduction corresponds to the loss of 8.33% of the slot. The algorithm, in the presence of commissions (therefore of a loaded network), becomes highly punitive. This is because the greater the network load, the more critical the breaches become. Third example: Retrieving a frozen bet

In this section we will present the refund mechanism for bets that have been frozen starting from the simulated condition in the unbalanced bet with commission. In this way, the return to the optimal operating parameters of the network is rewarded.

Mallory reconfigures the pool to fit the optimal mining parameters

Starting from the aforementioned condition (second example), before the EEP of period 16 it adds 4 nodes. With the addition of 4 nodes its 550 slots are divided into 550/5 = 110 <120. As always, the change will only take effect from the next period.

With this change, it leaves the penalty zone entirely.

As initial conditions, consider those highlighted in gray in the last column.

Now simulate the execution in case Mallory corrects its behavior by adding 4 nodes.

In this case you will get a reward for its 550 slots:

coinbase = 550 * 1 = 550

collected_fee = 550 * 5 = 2750

From the previous moment, Mallory inherited a frozen commission of 2400 tokens and a number of penalty slots of 960 (twice as many slots for which it remained out of parameters). The formula to calculate the commission recovered for a slot is as follows: recovered_feeE17S1 = frozen_feeE17S1 / penalty_slotsE17S1

At this point the frozen_fee and penalty_slots values are updated:

frozen_feeE17S2 = frozen_feeE17S1 - recovered_feeE17S1

penalty_slotsE17S2 = penalty_slotsE17S1 - 1

For convenience of note, the period of the absolute value is indicated in the subscript, in this case 17, but in slot 2 the slot assigned to Mallory after one is indicated. The two slots may not be consecutive. In this case, after the "s", the progressive number of the slot assigned to Mallory is shown.

The following table shows the evolution of the values at the passage of the slots.

By the end of the period, Mallory will have recovered 275 tokens for the 2400 frozen. This allows it to partially bridge the gap with the other nodes in the network and constitutes an incentive to favor the virtuous behavior of having added nodes.

In frozen commissions, there are no monetary bases that have not been generated. It is always and in any case more convenient for a node to operate in accordance with the rules rather than using this recovery procedure.

Claims (11)

CLAIMS 1. Method implemented by computer to reach a distributed consensus based on a proof of stake for a blockchain network, thus allowing to reach a higher level of network distribution and not to have too many stakes concentrated in too many nodes, called blockchain network including one or more MAIN addresses, for each MAIN address being associated in turn with one or more OVERFLOW addresses / nodes, the latter being able to act as potential mining nodes only if a predetermined minimum amount of bets is assigned to said one or more OVERFLOW addresses / nodes, including the following processes: - a redistribution process configured to calculate the minimum stake needed for an OVERFLOW node to be extracted, being minimum_stake_value = total_amount_at stake / predetermined_target_number_of_nodes, to calculate the maximum stake that allows efficient mining, being the maximum stake value = minimum stake value * 2, for each MAIN address to distribute the bet among the OVERFLOW nodes, scrolling through the OVERFLOW nodes of a given MAIN address and to assign each OVERFLOW node the minimum_stake_value, until the remaining stake is sufficient to activate a miner or to make sure that each overflow node has been assigned its stake; - a penalty process configured to freeze an amount of reward corresponding to the computational work for having mined a number of slots on a predetermined limit Slot_MAX in a periodi 1 and to recover it in the following periods starting from periodi 2, a little at a time, based on a number of parts proportional to the number of slots mined on said predetermined Slot_MAX limit, being a part assigned to each subsequent slot for which the miner can be delegated to mining, therefore in order to naturally incentivize the distribution of the blockchain network. 2. A computer implemented method according to claim 1 wherein the redistribution process can be further configured such that, if any bet remains, it is distributed among the OVERFLOW nodes until as many nodes as possible have a bet equal to the maximum value of the bet, and each further bet is assigned to the first OVERFLOW node according to a predetermined order. Computer-implemented method according to one of the preceding claims, wherein the redistribution process comprises in particular the following steps: 52: have a blockchain network comprising one or more MAIN addresses, for each MAIN address being associated in turn with one or more OVERFLOW addresses / nodes which act as potential mining nodes only if a predetermined minimum amount of bets is assigned to said one or more OVERFLOW addresses / nodes; 55: Obtain as input TOTAL_AT_STAKE value for each MAIN address of the blockchain network corresponding to the total amount of bets on each MAIN address 60: calculation of S_min_VAL = TOTAL AT STAKE / TARGET NUMBER OF_NODES being TARGET NUMBER OF_NODES = total number of OVERFLOW mining nodes in the chain network, Calculation of S_MAX_VAL = S_min_VAL * 2 Assuming a counter value ITERATION = 0 65: If the ITERATION value is <2, update the ITERATION = ITERATION 1 counter value 70: for each MAIN address that obtains as input the value MAIN_ADDRESS.assigned_stake and if this value is> 0, which means that the MAIN address in question contains an amount of bet, assigning this value to the parameter M_A_S, and if the value M_A_S is> = S_MAX_VAL repeat operation 70 until the next MAIN address, 75: if the M_A_S value is <S_MAX_VAL, obtain as input the OVERFLOW_ADDRESS.assigned_stake value for each OVERFLOW address associated with the MAIN in question, and if the OVERFLOW_ADDRESS.assigned_stake value is = 0 repeat operation 75 until the next OVERFLOW address associated with the MAIN in question and if the OVERFLOW_ADDRESS.assigned_stake value is ≠ 0, assign the OVERFLOW_ADDRESS.assigned_stake value to the O_A_S parameter, calculating MAIN_ADDRESS.assigned_stake = M_A_S - S_min_VAL, and calculating OVERFLOW_ADDRESS.assigner_stake_stake_address = O_A_Stake_Stake_ 75, and repeating OVERFLOW_ADDRESS.assigner_stake_stake_ address = O_A_Stake_Stake_ 75 exam. Computer implemented method according to one of the preceding claims 1, 2 wherein the penalty process comprises in particular the following steps: 80: calculate the total frozen_fee i 1 as the amount of the prize i 2 not paid to miners in a specific period i 1 for mining a number of slots over the Slot_MAX limit 85: calculate the penalty_slotsi + 1 preferably twice (or according to another proportional law) the number of slots mined over the limit, i.e. the number of slots on Slot_MAX 90: calculate the recovered_fee j, i 1 as frozen_feei + 1 / penalty_slotsi + 1, i.e. the fee to be recovered for each slot 95: add the value recovered_feej, i + 1 to the MAIN_address reward for mining block i in period i + 1 as rewardj, i + 1 = a coinbase collect_fee j, i 1 recovered_fee j, i 1 100: calculate the new value of the frozen_fee i 1 parameter as frozen_fee i 1 - recovered_fee j, i 1 and the new value of penalty_slot i 1 as penalty_slot i 1 - 1 and repeat operation 95 for the following block j 1 in period i 1 while frozen_fee i 1> 0 or while penalty_slot i 2> 0, so when the last blockj is mined at the end of each periodi, penalty_slotsi and frozen_feei are transferred from the first OVERFLOW node to its MAIN address and when the first block j is mined at the beginning of each periodi, penalty_slotsi and frozen_feei are transferred from the MAIN address to its first OVERFLOW node. 5. Method implemented by computer according to one of the preceding claims further comprising an initialization process configured to list all MAIN addresses of the blockchain network, for each MAIN address list its assigned OVERFLOW addresses / nodes and therefore for each OVERFLOW address / node transfers the bet, if present, to the assigned MAIN address. 6. A computer-implemented method according to claim 5 wherein the initialization process comprises in particular the following operations: 20: have a blockchain network comprising one or more MAIN addresses, for each MAIN address being associated in turn with one or more OVERFLOW addresses / nodes which act as potential mining nodes only if a predetermined minimum amount of bets is assigned to said one or multiple OVERFLOW addresses / node; 25: through the blockchain network, receive a transaction including as input addresses from the address register which includes the MAIN addresses and their corresponding OVERFLOW addresses; 30: associate a unique progressive number to each MAIN address; 35: for each MAIN address check if it contains a quantity of bets and if this value is> 0 assign this value to the parameter M_A_S, and if this value is <= 0 carry out the operation 35 up to the next MAIN address which is a member of the blockchain network and generate a transaction including the output providing the M_A_S value associated with the corresponding MAIN address, 40: associate a unique progressive number to each OVERFLOW node 45: for each OVERFLOW node assign to each corresponding MAIN address checking if it contains an amount of bets defined OVERFLOW_ADDRESS.assigned_stake and if this value is> 0 assign this value to the O_A_S parameter, calculating the MAIN_ADDRESS.assigned_stake value as M_A_S O_A_S and assigning O_A_S = 0 and generate a transaction that includes the output that provides the updated value of the M_A_S value associated with the corresponding MAIN address, then perform operation 45 until the next OVERFLOW address assigned to the corresponding MAIN address considered in operation 35, and if OVERFLOW_ADDRESS.assigned_stake is <= 0 execute operation 45 up to the next OVERFLOW address assigned to the corresponding MAIN address considered up to operation 35 50: Generate a transaction that includes the output of the updated value of the TOTAL_AT_STAKE value for each MAIN address of the blockchain network corresponding to the total amount of bets on each MAIN address after performing the above operations. 7. Method implemented by a computer according to one of the preceding claims, further comprising a registration process configured to register one or more MAIN addresses, to register one or more OVERFLOW addresses / nodes and to associate one or more OVERFLOW addresses / nodes to said one or more MAIN addresses and register said association. 8. A computer-implemented method according to one of the preceding claims which further comprises a maintenance process, in order to maintain the highly distributed blockchain network even in the event of the failure of one or more mining nodes, said maintenance process comprising in turn the following operations, given a MAIN address and at least two or more OVERFLOW addresses and a node to be maintained: 110: Send a transaction that includes a first input that provides a replacement OVERFLOW address using its private code, which means that this replacement OVERFLOW address is a node that replaces the node you want to maintain 120: assign said substitute OVERFLOW address to the corresponding MAIN address using the private code of the MAIN address 130 unregister the node you want to maintain using its private code 9. Node configured to implement a method to reach a distributed consensus based on a proof of stake for a blockchain network, thus allowing to reach a higher level of network distribution and not to have too many bets concentrated in too many nodes, said node including an interface device, a processor coupled to said interface device, a memory coupled to the processor, the memory having stored on it computer executable instructions which, when executed, configure the processor to perform the corresponding operations of the computer implemented method of any one of claims 1 to 8. A computer readable storage medium comprising computer executable instructions which, when executed, configure a processor to execute the computer implemented method of any one of claims 1 to 8. 11. Computer network for performing a blockchain-based transaction from a first user to a second user in an electronic transfer network of connected nodes, wherein the nodes include at least one first node, a second node, according to claim 9, in wherein each node shares a same blockchain, and in which each node stores a computer readable storage medium according to claim 10.