The year 2022 showed an urgent need to improve the existing systems for recognizing objects in the aerial space, which is caused by the significant increase in the number of technical means (especially unmanned aerial vehicles) on the battlefield. Such a sharp increase in the number of objects that simultaneously take part in combat operations in the air requires the improvement of military object recognition systems, both qualitatively and quantitatively. This requires the development and improvement of appropriate new generation Friend-or-Foe algorithms for the objects' recognition. The main requirements for recognition systems of aerial objects of civil application were determined. They include maximum compatibility; support for a large number of objects; outdated recognition complexes support; support for alternative ways of recognition; support for alternative data entry methods; determining the coordinates of aerial objects in an emergency situation. Friend-or-foe recognition systems for military applications were also considered and formulated. In contrast to civilian systems, the following basic requirements have been identified for them: 1) Maximum speed of the recognition process. 2) Protection against false positives. 3) Protection against legitimate aerial object imitation. 4) Support for a large number of objects. 5) Protection against cases of loss of a legitimate aerial object. 6) Rotation of the secret part. 7) Protection against false-negative results to prevent friendly fire. 8) Protection against man-in-the-middle attacks. 9) Flexible integration with the NATO block recognition system. 10) Availability of opportunities for purely domestic production and support of the object recognition system. 11) Protection against electronic warfare means. 12) Support for several recognition modes. 13) Automatic blocking of the launch of ground-to-air and air-to-air weapons against objects that confirm their legitimacy by a correct response to a request. 14) Determining the coordinates of aerial objects in an emergency. The statuses of objects in the “friend-or-foe” recognition system were formulated and described. Based on the formulated requirements and statuses, an improved friend-or-foe algorithm for the state identification system for military use is proposed, built based on the state standards, and considering the features of its software implementation to increase recognition speed and correctness rate. Its implementation will ensure sufficient scalability, stability, reliability, and multi-level recognition for military friend-or-foe recognition.
2022 Copyright for this paper by its authors.
The year 2022 showed an urgent need to improve the existing systems for recognizing objects in the aerial space, which is caused by the significant increase in the number of technical means (especially unmanned aerial vehicles) on the battlefield. Thus, the creation of an army of drones was announced in Ukraine [1]. This concept focused on procurement, repair and replacement of the massive number of UAVs. At the first stage of the program implementation 200 tactical level UAVs will be purchased for air reconnaissance. At the second stage each unit of the Armed Forces will have its own reconnaissance UAV.
It should be considered that this will lead to a huge increase of the UAVs number that could simultaneously be in the airspace control zone. Also, in addition to the tactical reconnaissance UAVs of the Armed Forces units, strategic reconnaissance UAVs, reactive and ballistic missiles can be present in the same airspace – and all this together with the usual airplanes and helicopters. And then this number should be at least doubled – to consider the corresponding number of enemy targets in the air.
Such a sharp increase in the number of objects that simultaneously take part in combat operations in the air requires the improvement of military object recognition systems, both qualitatively and quantitatively. This requires the development of appropriate new generation Friend-or-Foe algorithms for the objects’ recognition.
It is also necessary to consider the fundamental differences in the requirements for the aerial objects recognition systems of civil and military use. When describing them, let us use the definition "legitimate aerial object" – this is an aerial object that has the right to be in the given airspace, has a working respondent of the aerial objects identification system and provides correct answers to requests from the identification center.
Now let's consider the difference with the recognition systems made for military applications. The
basic operational principle of any modern government identification system used in military
applications is to process the incoming request according to a formula that is a cryptographic secret
and changes regularly (for example, every 24 hours). In contrast to civil systems, the following basic
requirements have been identified for them:
1. Maximum recognition process speed. Since the situation on the battlefield changes very
quickly, and for air combat this statement is even more relevant, any delay in the
recognition process can lead to losses, even human losses. So, for example, for
antiaircraft missile systems, the target's stay in the affected zone usually does not exceed a few
tens of seconds. This is especially relevant for UAVs, as UAVs usually have very strict
limitations on the amount of free space for installing the recognition system, and its weight
and power consumption – and, accordingly, on speed and the amount of available memory
for this system. So, the development process of the software component of this system
should take this into account.
2. Protection against false positives. For civil applications, cases of trying to pretend that
aerial object is not what it is truly is theoretically unlikely (and without involvement in
military applications have not yet occurred) - because in such situation the aerial object
owner will not be able to avoid responsibility and will lose a large amount of money in
fines and lawsuits. On the other hand, in military applications, since the enemy is most
interested in pretending that its aerial objects are belong to us (and therefore – avoiding
facing anti-aircraft defense system) so it is ready to spend almost unlimited time and
resources for this, and the result of a false-positive result of the aerial object recognition
can be airstrikes and human casualties, then protection against such situations should be
the highest priority of the state recognition system.
3. Protection against response imitation of a legitimate aerial object. Since the entire
exchange of information in the state identification system "friend-or-foe" is carried out
through the radio air, it is quite possible that all the data circulating between the legitimate
aerial object that gives a correct answer to the “friend-or-foe” request, and the recognition
center on the ground, can be intercepted by the enemy. After that, the enemy can try to
simply repeat the same responses to requests from the recognition center or try to change
them in a way that simulates a legitimate aerial object’s response. That is why the state
identification system must be reliably protected against this type of attacks.
4. Support for a large number of objects. As already mentioned above, “friend-or-foe”
military identification system must support the simultaneous recognition of multiple aerial
objects of different types to determine the identity of aircraft, helicopters, UAVs and their
swarms and cruise missiles in a timely manner.
5. Protection against cases of a legitimate air object loss. This requirement should be
considered if a legitimate air object was shot down over the enemy territory or fell into the
enemy’s hands by some other way. If there is no such protection, then the situation
described above will lead to the entire “friend-or-foe” identification system compromise
and the state will have to replace it on all legitimate aerial objects and recognition centers.
Such situation has already happened in the past, for example, in the Soviet Union [
The only overlap in the requirements for the recognition systems of civilian and military objects are the support of a large number of objects and special signals for emergency situations.
“friend-or-foe” object recognition system
To date, the "Parol-M" hardware/software complex, which is a modification of the Soviet Union system, developed in the 1980s and was itself developed as a replacement for the long-outdated “Kremniy-2” (2M) complex, which supported only 10 requesters and 10 respondents at the same time.
The technical capabilities of the " Parol-M" complex provide for the simultaneous recognition of up to 110 requesters and 110 responders [4]. At the same time, a similar system in the NATO countries– MarkXII – performs 400 polls per second in the nominal mode [5]. 4.1. Advantages of the “friend-or-foe” identification system currently used in Ukraine 1. Presence of an anti-imitation recognition mode. 2. Availability of guaranteed recognition mode. 3. The ability to perform the recognition procedure even in conditions of high-intensity interference application by EW means. 4. Availability of individual codes for recognition based on the principle "Who are you?" 5. Protection against receiving responses on the side lobes of the directional diagram. 6. Application of a high frequency range.
7. Variation in the frequencies of requests and responses [
Disadvantages of the
currently used in Ukraine
“friend-or-foe” identification
system
1. Support of an insufficient number of recognition objects.
2. Insufficient radio-electronic protection of the recognition process.
3. Insufficient imitation resistance – the probability of imitation of a correct response by the
enemy is as much as 0.5% [
Federation) in almost all details. 4.3.
Modern approaches to the “friend-or-foe” identification system development
In NATO countries, a large amount of works [6]-[8] is devoted to the issue of object recognition on the battlefield. Among the areas of development of the so-called Battlefield Combat Identification System (BCIS), the following should be highlighted: 1. Identification based on means of automatic radio data transmission about one's troops (Radio Based Combat Identification – RBCI). 2. Identification using radio tags (Radio Frequency Identification tags – RF tags). 3. Targets recognition on the battlefield (application of Battlefield Target Identification
Devices – BTID). 4.3.1. RBCI 4.3.2. RF tags
RBCI, also called Battlefield Force Tracking System (BFTS) or Blu-Force Tracking (BFT) System, is built on network-centric principles. Each legitimate aerial object equipped with the system transmits data about its location every 5 minutes by means of satellite communication or in the VHF communication network. In active mode, the requester sends a general request with coordinates – and the responder compares the received coordinates with their own, and if they match – sends a response. All data in wireless communication channels is encrypted.
The advantage of this approach is the ability to recognize objects outside the direct line of sight. Disadvantages are the need to use a complex system of repeaters on the battlefield, rapid aging of data for fast-moving objects, high impact of EW means and high cost of the system.
Recognition with the help of radio tags (RF tags) is also based on the "request-response" principle, as for civilian tags, used, for example, at a warehouse, the response is formed by modulating the incoming request. Active (similar to BTID), semi-active (have their own power source) and passive (powered by the energy of requests from the requester) tags may be used. The detection range of an active or semi-active tag can reach 40 km [6]. In fact, radio tags are currently the only potentially applicable identification method for determining the affiliation of individual enlistee or small units of them on the battlefield. And due to their small size and power requirements, they are potentially applicable for UAVs as well.
BTID systems are designed to recognize aerial objects in the "friend-unknown" format. Its essence does not differ from the general recognition of the “friend-or-foe” identification at Mk XII system. The term "unknown-friend" was introduced into military practice in view of the fact that an object of recognition that does not respond to a request is not necessarily an enemy object [6]. BTID systems also work on the "request-response" principle, the signals are encrypted and, to reduce the probability of interception, broadband.
The state “friend-or-foe” recognition algorithm must be built based on the national state standards. In fact, it should not be a single algorithm, but a family of algorithms – since the same algorithm cannot be used, for example, for the "land-aircraft" recognition line, as for the "aircraft-tank" line. At the same time, during the algorithms and corresponding software development the specified requirements, presented in the previous subsections, must be considered.
The algorithm and procedure for generating random keys should be highlighted separately. It should also be based on the national state standards. This procedure would be applied on an ongoing basis due to the requirement for constant key rotation. Generation of a pseudorandom sequences could be used for this goal. Another option is to use a physical generator to produce random sequences (for example, by saving the parameters of parasitic transistors capacitances etc.) [9]. But as the requirements to the software performance of these algorithms aren’t that severe (generation may take hours to complete and may be done with the parallel algorithms on the cluster, or just powerful hardware), so they wouldn’t be analyzed in more detail.
The recognition system includes (Figure 1): 1. The main recognition center. 2. Recognition centers (usually installed at radar complexes). 3. Centers for launching aerial objects (airports and military units, armed with UAVs). 4. Aerial objects (manned and unmanned). 5. Ground objects*.
6. Naval and underwater objects*.
* This work deals mainly with aerial objects. In case of expanding the work of the friend-or-foe” recognition system to land and naval objects, the recognition system should include headquarters (for distributing keys from the main recognition center to land objects) and ports (for water/underwater objects objects). At the same time, since ships and submarines can perform tasks autonomously for more than one day, this specificity should also be considered separately when implementing "aircraftship", "ship-aircraft" recognition lines etc.
The general scheme of the “friend-or-foe” recognition system (Figure 2): 1. The main recognition center generates random keys (general and individual, for recognition based on the "Who are you?" principle). New keys are generated every day. In this case, even if the current keys are compromised, the system will be protected again the very next day. 2. At the end of each day, the keys generated by the main recognition center are sent to all recognition centers and aerial objects launch centers (if necessary, to headquarters, etc.). 3. Next day, before each flight of manned and unmanned aerial vehicles, the keys are stored in their memory. 4. During the execution of the flight task, if necessary, recognition takes place along the required recognition line.
Due to the existence of two conflicting requirements for the “friend-or-foe” recognition system (recognition must occur as soon as possible to prevent friendly fire – but it must be reliable so that the enemy cannot deceive and bypass the air defense system), identification is proposed to be carried out in two stages: 1. Stage 1. Maximum speed, but reliable recognition is not fully guaranteed. Does not answer the questions "Who are you?", "Where are you?". 2. Stage 2. A slower, but better protected stage, which verifies the correctness of Stage 1 object recognition. Can answer the questions "Who are you?", "Where are you?".
These stages determine the statuses of objects in the “friend-or-foe” recognition system (Figure 3): 1. "Unknown". Before the first stage of recognition, the object to which the recognition procedure is applied is considered "unknown". 2. "Possibly friend". This status is assigned to an object that has successfully passed the first stage of recognition. 3. "Friend". This status is assigned to an object that has successfully passed the second stage of recognition. 4. "Foe". This status is assigned to an object that has failed the first or second stage of recognition and given a wrong answer.
From the software implementation point of view – as it was mentioned above, most of the UAVs have very limited power and space resources for installing “friend-or-foe” recognition system. So, it should be done as small and low power-consuming as possible. In the result mostly single-plate computers or even integrated circuits are used. Their performance in most cases is poor. Therefore, the first stage algorithm should be able to be implemented programmatically on such hardware and work fast on it. So, it must be as simple as possible. And as the most cryptographic algorithms (even symmetric ones) are pretty resource-consuming, it should be considered as well.
Now let’s describe the developed fast, but not fully guaranteed reliable algorithm of Stage 1 and aspects of its software implementation, improved corresponding to the currently existing algorithm: 1.
As described above in the general scheme of the recognition system, the main recognition center generates random keys (common to all objects of the recognition system) valid for Herewith:
Р1 ∩ Р2 = Р1 ∩ Р3 =…= РM-1 ∩ РM = Ø That means that no subset has elements in common with any other subset.
3. Before letting the recognition object (aircraft, UAV etc.) taking off on a task, the launch center transmits KI1 and KI2 keys, as well as a disjoint subset Рjk (k=1...S, where S is the expected, planned volume of requests from the object of recognition) of the subset Pj IZi. 4. To carry out Stage 1 of the recognition procedure, the requester (for any of the lines of recognition, the algorithm is unchanged) forms a request: with the operation of exclusive disjunction (which also has the names exclusive OR, or XOR) encrypts the previously unused IZi request identifier with KI1 key, at the output we receive a 128-bit encrypted recognition request Ra, which is sent to the respondent:
Ra = КІ1 IZi
XOR was chosen for speeding up software work – as it is very fast operation to perform on any hardware.
5. The respondent (for example, the UAV) receives the Ra request. 6. The respondent uses the key KI1 and decrypts the received Ra request using the exclusive disjunction operation (due to the double XOR operation using the key KI1, it deletes itself):
Rb = КІ1 Ra = КІ1 КІ1 IZi = IZi 7. The respondent performs an exclusive disjunction operation on the first and second halves of the deciphered IZi identifier – IZi1 and IZi2:
Rb = IZi1 IZi2
Rс = Rb КІ2 In this way, the respondent receives a block Rb with a length of 64 bits.
8. The respondent encrypts the received block Rb by an exclusive disjunction operation with the KI2 key: 9. The respondent sends the requester an answer – Rc. 10. The requester decrypts the received answer Rc with the KI2 key:
Rd = Rc КІ2 = Rb КІ2 КІ2 = Rb 11. Immediately after step 4, the requester performs the same action as the responder at the step 7 – an exclusive disjunction operation on the first and second halves of the used identifier IZi to obtain a control value:
Rk = IZi1 IZi2 12. After receiving and decoding the answer from the respondent, the requester compares the decoded answer Rd with the control value Rk. If they match, the respondent has successfully passed the check and receives the status "Possibly friend".
It is also possible to further simplify and speed up Stage 1 – to abandon the use of the KI2 key, and discard steps 8 and 10, then at the cost of a slight decrease in system reliability, you can get rid of two encryption/decryption operations.
Let's consider what data an attacker can get from a request and response exchange intercepted on the radio during the Stage 1.
An attacker can intercept two messages transmitted over the radio air – the request Ra = КІ1 IZi at step 4 and the response Rс = Rb КІ2 at step 9.
If the public key KI1 and the identifier IZi are presented in the form of two parts of 64 bits each (respectively KI11 and KI12, as well as IZi1 and IZi2), then an attacker in the request can intercept at the step 4 next data: КІ11 IZi1 DATAPACK1=
КІ12 IZi2 In response, he can intercept:
DATAPACK2= IZ1i IZ2i КІ2
Since all the initial data that should remain hidden from the adversary (KI11, KI12 and КІ2, as well as IZi1 and IZi2) according to statistical characteristics should not differ from random ones, the means of statistical analysis will not be able to give the adversary any clues about these initial data according to intercepted datapacks.
If he tries to use the intercepted data to find out more by performing XOR operation, he can get the following combinations: КІ11 IZ2i КІ2 КІ12 IZ21 КІ2 КІ11 IZ2i КІ12 IZ21
КІ11 КІ12 КІ2
If an attacker sends (for the purpose of determining secret keys) a false identification request according to Stage 1, for example (for ease of understanding), it will consist of all zeros – then the respondent will perform a decryption operation on this request, receiving a sequence of two halves of the key KI1 – KI11 and KI12, then perform XOR operation on these keys – KI11 and KI12 – and then perform a XOR operation on the received result with the key КІ2 – and sending the received result to the attacker.
That is, the attacker will again receive the sequence КІ11 КІ12 КІ2 (or, in the case of using the simplified algorithm of Stage 1 – КІ11 КІ12).
Since the statistical characteristics of the keys and identifiers will not differ from random ones, the obtained data will not give the attacker any useful information about the values of KI11, KI12, and КІ2 – there is a very large (limited only by the size of the keys) number of variants of KI11, KI12, and КІ2 sets (or in the simplified case – KI11 and KI12), which will correspond to the data known to the attacker. And he will not have the opportunity to determine which option from this set is correct, without going through them sequentially, using it as an answer to a request to the object of recognition, facing all the consequences of the fact that after an incorrect answer, this the object will be marked by the recognition system with the status "foe".
Now, after these crypto approaches analysis, let’s describe the advantages of the Stage 1: 1. Stage 1 is incredibly fast and does not require many hardware resources. To prepare a request, the requester should perform only one operation of exclusive disjunction (XOR) of two data blocks of 128 bits each. The length of the request is also only 128 bits. The respondent must perform three exclusive disjunction operations, the first for data blocks of 128 bits each, the second and third for blocks of 64 bits each. The response to the request is only 64 bits long. In the memory of the respondent (if it is not planned that he will perform the identification procedure as a requester) there should be space for only two keys with a total volume of 192 bits and the same amount of free memory should be available for performing decryption/encryption operations. So, from the software development point of view this algorithm is very fast. 2. Stage 1 recognition is sufficiently reliable. No data is transmitted with it in open form. The operation of exclusive disjunction is used for encryption, which is vulnerable to an attack on known plaintext. But in the case when each request identifier is used only once, and all request identifiers and encryption keys are generated in a truly random manner, then XOR provides sufficient robustness because: • Text and password have the same length. • No data in the message (request identifiers) is used more than once (reverse requirement of “no password is used more than once”). • Both the password and the message are random and cannot be guessed either by a dictionary or by other methods (an enhanced requirement compared to the classic "password is random"). • If we consider request identifiers as one-time keys, then in the absence of any statistical regularities during their generation, the algorithm is close to 100% security and is a form of the only known unbreakable cryptographic system, the one-time pad.
3. Due to the physical features and limitations of the recognition procedure (blocking repeated requests for some time to prevent the processing of reflected signals behind the side lobes of the multi-channel receivers’ radar etc.), it is protected from bruteforce attacks. In addition, you can set timeouts and block responses to requests from the same requester after a certain number of attempts. And within the military use any unsuccessful attempt to pick up correct response will cause severe consequences.
Disadvantages of Stage 1: 1. Does not answer the question "Who are you?" and "Where are you?". 2. Anyone can impersonate a legitimate requester. When performing Stage 1 recognition, the responder cannot verify the legitimacy of the request – and therefore will respond to any request of this format, including from an attacker sending a random 128-bit request without knowing any of the keys and identifiers. 3. Does not provide protection against “man-in-the-middle” attacks. If an attacker receives a Stage 1 request and sends it to a legitimate respondent, and then forwards its response to the requester, it can impersonate the legitimate respondent. The probability of this can be reduced by limiting the maximum response waiting time, but the theoretical possibility of such an attack remains.
4. Requires the daily generation of a large number of recognition identifiers.
To compensate for these shortcomings, the Stage 2 algorithm should be applied.
The essence of the Stage 2 algorithms work is the processing of flying object’s requests and responses, which are encrypted with a symmetric cryptographic algorithm. This approach is chosen because, as at Stage 1, the maximum performance of such a system is required, and the exchange of public keys in an asymmetric system may not work in the presence of natural noise or intentional interference created by the enemy's EW complexes. Another approach that can be used in step 2 is to use an asymmetric cryptographic algorithm only to encrypt the symmetric algorithm key for sending during request/response signaling. In this case, the ability to decrypt and use the symmetric algorithm’s key automatically means the presence of the asymmetric algorithm’s key.
As mentioned above in the requirements for military identification systems, in response, the flying can provide not only its identifier (if the request "Who are you?") was sent, but also data on coordinates, type of aircraft, etc., which can be additionally used to prevent replacing the response signal and checking the authenticity of the received code. Flying object, included in the subsystems of the Unified Automated Control System (EASU) of the Armed Forces, can also use information from civilian systems to verify data received through special networks that use symmetric and asymmetric cryptographic algorithms to protect information to ensure multi-level identification.
As an algorithm of Stage 2, it is possible to use the algorithm, based on Ukrainian symmetric cryptography standard [10] proposed and detailly described in [11-12]. In case the direct object-object recognition (“plane-plane”, “plane-tank” etc.) should be provided without any of the objects being the recognition center, analogue of Kerberos protocol [13] may be used.
In this work the main requirements for recognition systems of aerial objects of civil application were determined, which includes maximum compatibility; support for a large number of objects; outdated recognition complexes support; support for alternative ways of recognition; support for alternative data entry methods; determining the coordinates of aerial objects in an emergency situation.
“Friend-or-foe” recognition systems for military applications are also considered. In contrast to civilian systems, the following basic requirements have been identified for them: maximum speed of the recognition process; protection against false positive; protection against legitimate aerial object imitation; support for a large number of objects; protection against cases of loss of a legitimate aerial object; rotation of the secret part; protection against false-negative results to prevent friendly fire; protection against man-in-the-middle attacks; flexible integration with the NATO block recognition system; availability of opportunities for purely domestic production and support of the object recognition system; protection against electronic warfare means; support for several recognition modes; automatic blocking of the launch of ground-to-air and air-to-air weapons against objects that confirm their legitimacy by a correct response to a request; determining the coordinates of aerial objects in an emergency.
Based on the formulated requirements, a new friend-or-foe algorithm for the state identification system for military use is proposed, built based on the state standards, and considering the features of its software implementation in order to increase speed. Its implementation will ensure sufficient scalability, stability, reliability, and multi-level recognition.
The references should be formatted according to the following gudelines: A paginated journal
article [
Series, 1992. doi:10.4271/920983. [4] Zakrevskyi O., Friend-or-foe, 2014. URL: https://dou.ua/forums/topic/10097/. [5] W.-H. Kim, S.-Y. Jung, Y.-S. Lee, and S.-M. Chang, Mark XIIA(Mode 5) IFF system integration and certification test for surface to air missile system, Journal of the Korea Institute of Military Science and Technology, vol. 25, no. 2 (2022) 160–168. doi:10.9766/kimst.2022.25.2.160. [6] G.G. Kamaltinov, Recognition of objects on the battlefield. Analysis of world experience, Armament and military equipment. 4, 2016, pp. 22–26. doi: 10.34169/2414-0651.2016.4(12).2226. (in Ukrainian). [7] R. Putatunda, A. Gangopadhyay, R.F. Erbacher, C. Busart, Camouflaged Object Detection
System at the Edge, Automatic Target Recognition XXXII (2022). doi: 10.1117/12.2618869. [8] N. Pearce, S. Hamilton, IFF using Beamforming in Telemetry Beacons, 2021 IEEE Western New York Image and Signal Processing Workshop (WNYISPW) 2021 Oct 22, IEEE, 2021, pp. 1-5. doi: 10.1109/wnyispw53194.2021.9661287. [9] V.Y. Zhuykov et al, Adjustable power supply filters to protect information in microcontrollers,
Monograph, Kyiv, 2016, 184 p. [10] R. Oliynykov, I. Gorbenko, O. Kazymyrov, V. Ruzhentsev, O. Kuznetsov, Y. Gorbenko, D.
Kaidalov, A new encryption standard of Ukraine: The Kalyna block cipher, Cryptology ePrint Archive, 2015. [11] M.I. Ogurtsov, V.Yu. Korolyov, O.M Khodzinskyi, To the Problems of the National State Recognition System Improving, Cybernetics and Computer Technologies, 2, 2022, pp. 74–82. (in Ukrainian) doi: 10.34229/2707-451X.22.2.8. [12] V.Yu. Korolyov, M.I. Ogurtsov, A.I. Kochubinsky, Identification of technical objects in the special networks according to the principle of “Friend or Foe”, Control Systems and Computers, 4, 2021, pp. 3-12. (in Ukrainian) doi:10.15407/csc.2021.04.003. [13] M. Kuperberg, R. Klemens, Integration of Self-Sovereign Identity into Conventional Software using Established IAM Protocols: A Survey, Open Identity Summit 2022, 2022.